Public Duties for the New City

The first job of a government is to protect its people, and, in the United States, the government ostensibly performs this job through the police. But policing in America is deeply dysfunctional, as the police not only provide inadequate protection from violent crime, but simultaneously engage in outright acts of brutality against the citizenry. As awareness of these practices has swept across the nation, legal scholars and policymakers have offered numerous reforms and remedies to help solve policing’s problems. The responses have tended to focus on the top of the legal pyramid, using the big hammers of the federal government, the Constitution of the United States, the federal remedies of Section 1983, and the qualified immunity doctrine of federal courts as the requisite tools for reform. More recently, as these efforts have faltered, scholars and policymakers have begun to explore the possibilities for change at the state and local level.

This Article, too, begins at the bottom. While the proposed fixes to the federal framework are indeed important, this Article argues that changes at the lower, foundational level of cities, local governments, and common law duties of care are equally so. Policing is, after all, a fundamentally local matter, with thousands of municipal and county governments responsible for its administration. And duties of care are the most basic articulation of the norms and obligations flowing between members of our society, shaping not just private relations, but the government-constituent relationship as well. This Article argues that attending to these roots offers an opportunity to reorient the police-citizen relationship and recast the relational norms between local government actors and their constituents more generally. In particular, this Article argues that the “public duty doctrine”—a no-duty rule that immunizes municipalities from civil liability arising from police violence and failures to protect—has contributed to a profoundly unbalanced and perverse local-constituent relationship. To reestablish just relations, localities should bear, and indeed embrace, a legally enforceable duty of care to protect their constituents.

Such a duty would not open the liability flood gates, nor impose catastrophic expenses on cities, nor expand the already oversized footprint of policing. Such a duty would, however, achieve the usual tort goals of compensation and deterrence, significantly reduce the harms that police and other governmental actors visit on city constituents through both their action and inaction, align with corrective justice principles, enhance democratic accountability, advance the constitutional principle of equal protection, and accord with the thick conception of the city-constituent relationship that cities themselves put forward in the affirmative litigation context.

Further, implementing this duty on the ground would not be difficult. Neither courts nor legislatures need do anything at all; many cities could simply choose to not avail themselves of the public duty defense and instead accept an owed duty. Doing so would not only reorient the city-constituent relationship in a profoundly more positive way; adopting this duty would also serve cities’ broader self-interest. As cities increasingly vie for political recognition and acknowledgement as independently legitimate polities on both the domestic and international stage, this Article draws on the burgeoning sovereignty-as responsibility literature to argue that by embracing a duty to protect, cities can advance their own status as credible, politically important actors in the wider American democratic project.


The first job of government is to protect its people.1Ben A. McJunkin, Ensuring Dignity as Public Safety, 59 Am. Crim. L. Rev. 1643, 1647 (2022). To that end, as the Supreme Court has acknowledged, the people of the United States enacted their constitutions and created their governments specifically “to protect their persons and property from violence.”2Helen Gugel, Remaking the Mold: Pursuing Failure-to-Protect Claims Under State Constitutions via Analogous Bivens Actions, 100 Colum. L. Rev. 1294, 1295 (2010) (quoting Calder v. Bull, 3 U.S. (3 Dall.) 386, 388 (1798)). Without such protection, there would be little point in having a government at all: “If an individual could not appeal to the government for protection against imminent threats of private violence, there would be little advantage to forsaking the state of nature” in the first place.3Id. at 1295–96 (“According to Locke, when the government no longer fulfills its protective functions to ‘preserve the members of the commonwealth in peace from injury and violence,’ its raison d’etre is negated.” (footnote omitted)). Citizens agree to be subject to the laws of government because the government in turn agrees to provide the requisite protection to them; protection offered by government and citizen subjection to government are inherently reciprocal notions.4 Randy E. Barnett & Evan D. Bernick, The Original Meaning of the Fourteenth Amendment 94 (2021).

The main governmental agency tasked with providing this protection in the United States is the police.5The first modern police force was established in 1829 in London, England. American cities then soon adopted the model. Steven J. Heyman, The First Duty of Government: Protection, Liberty and the Fourteenth Amendment, 41 Duke L.J. 507, 544 (1991). Unfortunately, “[s]omething is seriously wrong with policing in the United States.”6Anthony O’Rourke, Rick Su & Guyora Binder, Disbanding Police Agencies, 121 Colum. L. Rev. 1327, 1337 (2021). In the contemporary United States, policing violates the fundamental tenet of protection in two main ways. First, police literally fail to protect. In moments when they are most needed, like during the recent mass school shooting in Uvalde, Texas that killed nineteen schoolchildren and two teachers,7See, e.g., Jake Bleiberg & Paul J. Weber, Uvalde Report: 376 Officers but ‘Egregiously Poor’ Decisions, AP News (July 17, 2022), []. or the mass shooting at a gay nightclub in Orlando, Florida that killed forty-nine people,8See, e.g., Bill Hutchinson & Justin Doom, 31 Orlando Police Officers Sued over Their Response to Pulse Nightclub Massacre that Left 49 Dead, ABC News (June 7, 2018, 2:38 PM), []. officers quite routinely fail to intervene and protect.9Alexandra Natapoff, Underenforcement, 75 Fordham L. Rev. 1715, 1717 (2006). Even when present at the scene, police have failed to intervene to save those in mortal danger from drownings,10See Emily Shapiro, Man Drowns as Officers Look on Without Coming to His Aid, Family Wants Them Fired, ABC News (June 10, 2020, 1:01 PM), []; Ross v. United States, 910 F.2d 1422, 1424–25 (7th Cir. 1990) (describing a county sheriff who not only neglected to rescue a drowning twelve-year-old boy himself, but refused to allow civilian divers to rescue him either; after twenty minutes during which multiple police officers simply stood there, the boy died). physical attacks from others,11See, e.g., Janelle Griffith, Actions of Officers Who Didn’t Stop Derek Chauvin Spur Duty-to-Intervene Laws, NBC News (Jan. 19, 2022, 4:32 AM), []; Ramenda Cyrus, Police Have No Duty to Protect the Public, Am. Prospect (April 18, 2022), [] (describing how a police officer failed to intervene while a man was attacked on the subway, because the officer feared the assailant had a firearm). and lethal exposure to the elements.12See Munger v. City of Glasgow Police Dep’t, 227 F.3d 1082 (9th Cir. 2000) (citing other relevant cases). At other times, the police simply do not show up at all: police bungle or do not respond to 911 calls for assistance,13See, e.g., Hartzler v. City of San Jose, 120 Cal. Rptr. 5 (Ct. App. 1975) (where the public duty doctrine was applied to shield police from liability when they failed to respond to a plea for assistance forty-five minutes before a homicide); Steve Papenfuhs & Eric P. Daigle, Addressing Cops’ Confusion over ‘the Public Duty Doctrine, Police1 (Jan. 5, 2012), []. frequently decline to even minimally investigate certain categories of crime such as crimes of a sexual nature14Natapoff, supra note 9, at 1721, 1731. or homicides in particular neighborhoods,15Id. at 1723. Even when police do investigate, they are not particularly effective. In 2020, the national clearance rate for homicides in the United States (meaning that an arrest was made in the case) was a dispiriting 54%. As a comparator, Nordic countries like Sweden and Norway regularly achieve clearance rates of approximately 98%. Ryan Cooper, Blame Police When They Fail Horribly, Am. Prospect (April 14, 2022), []. fail to respond to intimate partner violence,16Robert Lewis & David DeBolt, Police Officers Who Commit Domestic Violence Often Get to Keep Their Guns, Voice of San Diego (Nov. 10, 2019), []. and create entire “underenforcement zones” in communities of color while they prioritize resources in favor of protecting more affluent, white neighborhoods.17Natapoff, supra note 9, at 1721, 1723.

This underpolicing is ironically paired with overpolicing.18Id. at 1719. At the same time as police fail to protect citizens from the violence inflicted by others, police also fail to protect citizens when they engage in their own acts of violence against them.19Dallas Police Chief Eddie Garcia acknowledged this connection in the wake of the video showing police beating Tyre Nichols to death, stating, “As law enforcement we take an oath to protect and serve. The actions that day broke, violated and tarnished that oath.” Alex Sundby, Cara Tabachnick & Elias Lopez, Bodycam Footage of Tyre Nichols’ Violent Arrest Released by Memphis Police, CBS News (Jan. 28, 2023, 12:37 AM), []. Police kill an average of three people a day in America20Sam Levin, ‘No Progress’ Since George Floyd: US Police Killing Three People a Day, Guardian (March 30, 2022, 6:00 AM), [].—far more than in any other Western nation21Rob Picheta & Henrik Pettersson, American Police Shoot, Kill and Imprison More People Than Other Developed Countries. Here’s the Data, CNN (June 8, 2020, 7:13 AM), [] (noting a statistic showing that, in 2018, police in the United States fatally shot thirty-one per every ten million people. The equivalent rates in the United Kingdom, German, and Australia were less than one, one, and three per every ten million people, respectively).—and practices of brutality, excessive force, and racial profiling are routine and embedded in the American policing system.22See, e.g., Hannah Sarisohn, A Tennessee Police Officer Is on Administrative Leave as the Violent Arrest of a Man Is Investigated, CNN (July 23, 2022, 12:09 PM), []; N’dea Yancey-Bragg, Mississippi Police Chief Fired After Leaked Audio Captured Racist Rant, Him Bragging About Killing 13 People, USA Today (July 22, 2022, 6:50 PM), []. The nine-and-a-half-minute video of Derek Chauvin pushing his knee into George Floyd’s neck until he died,23How George Floyd Died, and What Happened Next, N.Y. Times (July 19, 2022), []. the barrage of bullets fired indiscriminately into the apartment of Breonna Taylor,24Josiah Bates, Breonna Taylor’s Killing Sparked Restrictions on No-Knock Warrants. But Experts Say Those Rules Don’t Actually Change Much, Time (March 11, 2022, 1:25 PM), []. and the millions of intrusive and unwarranted stop-and-frisk detentions that disrupted the lives of thousands of Black men and boys in New York City are all instances of police actively harming the public—particularly people of color.25See Brentin Mock, How Police Are Using Stop-and-Frisk Four Years After a Seminal Court Ruling, Bloomberg (Aug. 18, 2017, 7:30 AM), [] (“Between January 2004 and June 2012, the NYPD conducted over 4.4 million stops, over half (52 percent) of which were for African Americans compared to 10 percent for whites.”). A New York court declared stop-and-frisk unconstitutional in 2013. Floyd v. City of New York, 959 F. Supp. 2d 540 (S.D.N.Y. 2013). The police murder of George Floyd in May 2020 prompted what has been called the largest social movement in United States history,26Larry Buchanan, Quoctrung Bui & Jugal K. Patel, Black Lives Matter May Be the Largest Movement in U.S. History, N.Y. Times (July 3, 2020), []. yet virtually “no progress” in actually reducing the number of incidents of police violence has occurred.27Levin, supra note 20. Police killings continue unabated: to take one random sample in time, in early January 2023, Los Angeles police killed school teacher Keenan Anderson as he begged them to help him,28Shirin Ali, “We’re in the Process of Mass Forgetting, Slate (Jan. 21, 2023, 11:45 AM) [] (quoting Dr. Phillip Goff saying that Keenan Anderson “literally said, ‘please help me’ as the first words that are caught on camera. The result on the other end of it is that they tased him literally to death . . . .”); Emily Olson, A Million Claim Is Filed Against LA Over the Death of a Man Who Was Tased By Police, NPR (Jan. 21, 2023, 11:01 AM), []. and mere weeks later five Memphis police officers viciously beat twenty-nine-year-old Tyre Nichols to death.29Jonathan Franklin & Emma Bowman, What We Know About the Killing of Tyre Nichols, NPR (January 28, 2023, 4:50 PM), [].

The magnitude and complexity of the policing problem has naturally prompted legal scholars to offer equally large responses.30See, e.g., Alexander Reinert, Joanna C. Schwartz & James E. Pfander, New Federalism and Civil Rights Enforcement, 116 Nw. U. L. Rev. 737, 745 (2021). Often concentrating on issues of overpolicing (in part because overpolicing is more salient and has emerged as “one of the central lenses” for understanding problematic police practices),31Natapoff, supra note 9, at 1716. these efforts initially focused on the top of the legal pyramid and the big hammers of the federal government,32Reinert et al., supra note 30, at 739. the Constitution of the United States,33Id. the federal remedies of Section 1983,34E.g., Nancy Leong, Municipal Failures, 108 Cornell L. Rev. 345 (2023) (arguing that the failure-to-supervise path in the case law interpreting Section 1983 is underutilized). and the qualified immunity doctrine of federal courts35E.g., Joanna C. Schwartz, The Case Against Qualified Immunity, 93 Notre Dame L. Rev. 1797 (2018); William Baude, Is Qualified Immunity Unlawful?, 106 Calif. L. Rev. 45 (2018); see also Joanna C. Schwartz, Municipal Immunity, 109 Va. L. Rev. 1181, 1183 (2023) (“Qualified immunity has taken center stage in recent debates about police misconduct and paths to reform.”). as the tools for potential remedy.36Mandatory insurance for police officers has also been proposed. In Minneapolis, an advocacy group tried to advance a ballot measure requiring mandatory professional liability insurance for police officers, but that measure was determined to be in conflict with Minnesota’s state indemnification statute. Reinert et al., supra note 30, at 777. Scholarly works suggesting that insurance can be beneficial in this context include Deborah Ramirez, Marcus Wraight, Lauren Kilmister & Carly Perkins, Policing the Police: Could Mandatory Professional Liability Insurance for Officers Provide a New Accountability Model?, 45 Am. J. Crim. L. 407 (2019), and John Rappaport, How Private Insurers Regulate Public Police, 130 Harv. L. Rev. 1539 (2017), for the suggestion that insurance markets can help improve police practices. However, a recent article argues that proposals targeting insurance as a path of reform are “over-optimistic.” Kenneth S. Abraham & Daniel Schwarcz, The Limits of Regulation by Insurance, 98 Ind. L.J. 215 (2022). More recently, as these efforts have faltered, scholars and policymakers have begun to explore the possibilities for change at the state and local level.37Charlotte Plantive, Efforts Since George Floyd Killing Fail to Stem US Police Violence, Barron’s (Jan. 28, 2023) [] (noting that in 2020, considerable reform efforts attempted to target qualified immunity—a form of immunity officers receive in suits brought under section 1983 of the Civil Rights Act. Federal legislation in that regard was proposed, but stalled. Reform efforts then began to focus more seriously on the local level, but “modestly and unevenly, producing a patchwork of different approaches.”).

This Article, too, begins at the bottom. While the proposed fixes to the federal framework are important, this Article argues that changes at the lower, foundational level of cities, local governments, and common law duties of care are equally so.38Other scholars that have actively voiced encouragement for increased attention to the subfederal levels include Reinert et al., supra note 30; Leong, supra note 34; and Joanna C. Schwartz, Backdoor Municipal Liability, 132 Yale L.J.F. 136 (2022). Policing is, after all, a fundamentally local matter, with thousands of municipal and county governments responsible for its administration.39For a thorough examination of how localities administer police departments, see O’Rourke et al., supra note 6. Indeed, “[e]ven a cursory observation of local governments confirms” their significant role in carrying out police powers and activities.40Fred Smith, Local Sovereign Immunity, 116 Colum. L. Rev. 409, 456 (2016). It is the nation’s municipal police officers who respond when someone calls 911, who are dispatched to disturbances, who patrol city streets and apartment buildings, who stop individuals on the street who look “suspicious,” and who enforce all manner of state and local laws.41See id. at 456–57. By the numbers, of the approximately 850,000 total law enforcement officers in the United States, an estimated 650,000 are local government police officers.42 Rachel Harmon, The Law of the Police 6 (2021).

And although tort law is often conflated with purely private law and its role in the government-citizen relationship has been largely ignored,43Peter Cane, Tort Law and Public Functions, in Philosophical Foundations of the Law of Torts 148, 148–49 (John Oberdiek ed., 2014). tort has a first-order, foundational role in shaping the relationship between government and citizen.44Id. at 150–51. Duties of care are, at their core, our most basic articulation of the relational obligations that flow between members of our society, including public institutions and the public they are meant to serve.45Id. at 149. Tort law is both social and deeply political: it helps to “construct community” and “operates as a vehicle through which communities perpetually reexamine and communicate their values,” including the values that inform the government-citizen relationship.46Cristina Carmody Tilley, Tort Law Inside Out, 126 Yale L.J. 1320, 1324 (2017).

This Article argues that attending to these roots offers an opportunity to reorient the city-constituent relationship from the ground up and generate better behavior from all government actors, including police. Specifically, this Article argues that the “public duty doctrine”—a “no-duty” rule that immunizes municipalities from tort liability arising from police violence and failures to protect—has contributed to a profoundly unbalanced and perverse local-constituent relationship. To reestablish just relations, localities should bear, and indeed embrace, a legally enforceable duty of care to protect their constituents.

Such a duty would not open the liability floodgates, nor impose catastrophic expenses on cities,47For example, at least a dozen states have either abolished or significantly cabined the doctrine in other contexts outside of police, and none have experienced catastrophic liability as a result. See infra Part II. nor expand the already oversized footprint of policing.48Remedying underpolicing does not require “more police,” but it does require a different kind of policing that is democratically accountable and in service to all members of the local public, a vision wholly compatible with a smaller overall police footprint and with calls for abolition. See Natapoff, supra note 9, at 1321 for a discussion of this issue. Numerous additional mechanisms already operate to keep litigation volume reasonable, insurance and damage caps readily manage damage awards, and the problem of underpolicing is only rarely related to a lack of resources (policing often receives around 30% of total municipal budgets).49. For example, “65 out of the nation’s 300 largest cities spend 40 percent or more of their general budgets on policing.” Katharina Buchholz, How Much Do U.S. Cities Spend on Policing?, Statista (Feb. 6, 2023), []. In fact, a duty of care in this context may result in a smaller overall footprint for policing in general, as an expansive understanding of the duty of care flowing from localities to constituents would push in favor of creating “well-funded communities that are able to eliminate the criminogenic conditions (for example, poverty, poor schools, and lack of affordable housing) that lead to crime.”50India Thusi, Policing Is Not a Good, 110 Geo. L.J. Online 226, 232–33 (2022).

Further, such a duty would also achieve the usual tort goals of compensation and deterrence, significantly reduce the harms that police and other governmental actors visit on city constituents through both their action and inaction, align with corrective justice principles, enhance democratic accountability, advance the constitutional principle of equal protection, and accord with the thick conception of the city-constituent relationship that cities themselves put forward in the affirmative litigation context.51See infra Section II.C.4.

Implementing this duty on the ground would not be difficult. Neither courts nor legislatures need do anything at all; many cities could simply choose to not avail themselves of the public duty defense and instead acknowledge an owed duty.52See infra Part III. Doing so would not only reorient the city-constituent relationship in a profoundly more positive way; adopting this duty would also serve cities’ broader self-interest. As cities increasingly vie for political recognition and acknowledgement as independently legitimate polities, this Article draws on the burgeoning sovereignty-as-responsibility literature in international law to argue that by accepting a duty to protect, cities can advance their own status as credible, politically important actors.53Tove Dannestam, Rethinking Local Politics: Towards a Cultural Political Economy of Entrepreneurial Cities, 12 Space & Polity 353, 353–55 (2008). While sovereignty and recognition as an independent, legitimate polity is often thought of in terms of what rights are associated with this status, this literature reveals that responsibility offers another viable path to such recognition. Through rejecting the public duty doctrine, cities can advance their larger, ongoing project of creating themselves as a relevant and increasingly important political category.54See id. Embracing an enforceable duty can be “part of a general movement of municipal empowerment and maturation” occurring across the nation, enhancing both the theoretical and practical meaning of a city.55Sarah L. Swan, Plaintiff Cities, 71 Vand. L. Rev. 1227, 1286 (2018).

To this end, this Article proceeds as follows. Part I describes the local nature of policing and the specific democratic and relational harms of under- and overpolicing. Part II analyzes the public duty doctrine and weighs its various rationales against the benefits of its elimination. Part III focuses on the two largest benefits of eliminating the public duty doctrine: first, that relinquishing this no-duty rule would fundamentally reorient the city-constituent relationship, leading to better behavior from all local government actors, including police; and second, that through embracing a duty to protect, cities can advance their own status as credible, politically important actors in the American democratic project.

I. Local Under- and Overpolicing

This Part describes the local nature of policing, the paradoxical problem of under- and overpolicing, and the infirmities of the current legal responses used to address police accountability. Neither federal constitutional law nor state tort law currently offer reasonable avenues of remedy or relief for police misconduct.

A. Policing is Local

Although policing’s problems occur across the country with little regard to geography, policing in the United States is fundamentally a local endeavor.56 Harmon, supra note 42, at 4. Indeed, the first modern police force in the Western world was a local one: the Metropolitan Police of London established in 1829.57Heyman, supra note 5, at 544. By the mid-1800s, Philadelphia, Boston, and New York City had also created municipal police forces, seeding the field for the current locally administrated system of policing now spread across the United States.58Id.

Municipalities and counties are authorized to create and run local police departments through state law, and police provision throughout the nation occurs primarily through this means.59As Harmon describes:

State law authorizes the creation of municipal, and sometimes county, departments, which are run by politically appointed chiefs. That means city (or county) officials hire and fire those chiefs, and local (or county) governments set police department budgets, allowing them to require the agencies to adhere to specific goals and means set by the public officials. Those officials in turn are elected or appointed by those who are elected.

Harmon, supra note 42, at 9.
The number of local law enforcement officers dwarfs that of all other levels of government combined, with nearly 650,000 local law enforcement officers compared to approximately 130,000 federal and 60,000 state ones.60Id. at 6. New York City by itself employs an estimated 36,000 local police officers, which is two-and-half-times more officers than are employed by the entire Federal Bureau of Investigation.61Id. at 10. Finances provide another point of scale: approximately $90 billion is spent on local policing, as opposed to $28 billion spent on federal law enforcement and $16 billion on state law enforcement.62Id. at 5. Local law enforcement officers also have the broadest mandates: “[N]early all municipal and county police agencies enforce criminal laws, maintain order, and provide miscellaneous services to the public on a day-to-day basis,” whereas state and federal agencies typically have more specific or cabined responsibilities.63Id. at 4. At the state and county levels, police are primarily responsible for traffic enforcement on highways and specialized enforcement (e.g., fish and game police, harbor police, and officers assigned to protect state buildings). Police at these levels also cover jurisdictions that may not have their own police forces, and some state law enforcement agencies operate under general policing mandates. For a more detailed explanation of the roles of state and county police, see id. at 4–11.

At the federal level, “responsibilities of federal agencies are generally very specific and defined by federal law—for example, the Customs Bureau enforces import and export laws.” Id. at 4–5 (quoting Nat’l Rsch. Council of the Nat’l Acads., Fairness and Effectiveness in Policing (Kathleen Frydl & Wesley Skogan eds., 2004)).

In addition to municipal police departments with general mandates, special districts such as neighborhood business districts sometimes have their own law enforcement agencies as well. These police agencies can be found at sites like public schools, transportation systems, and college campuses, for example, and they operate under more specialized mandates. Id.
In fact, the leading casebook on the law of the police explicitly focuses “almost exclusively on local policing” because that is what policing in America is.64Id. at 3. There are approximately 12,500 local police departments across America, and only eighty-three federal law enforcement agencies.

In practical terms then, for better or for worse, most interactions between the public and the police are really between individuals and local police officers. In their daily lives, Americans are much more likely to confront “local municipal policing” than any other kind of law enforcement agency.65Id. at 4. Because municipal police “engage in routine patrol throughout the communities they serve and respond to requests for service, they are the agencies most visible to the public and also have the most direct contact with them.”66 Nat’l Rsch. Council of the Nat’l Acads., supra note 63, at 49. Note, though, that not every city has a municipal police force. Fiscally strapped smaller cities often outsource law enforcement tasks to the county sheriff. Additionally, counties may have law enforcement jurisdiction over some smaller townships and unincorporated communities which lack the authority to create their own municipal departments. O’Rourke et al., supra note 6, at 1382. Perhaps surprisingly, although these “unincorporated communities tend to be rural and sparsely populated, that category also includes some densely populated urban areas like East Los Angeles (with more than 100,000 residents) and Paradise, Nevada, where the Las Vegas strip is located.” Id.

These public-police interactions are paradigmatic encounters between individuals and the state.67Amy E. Lerman & Vesla M. Weaver, Staying Out of Sight? Concentrated Policing and Local Political Action, 651 Annals Am. Acad. Pol. & Soc. Sci. 202 (2014). Police are representatives of the state in the broad, amorphous sense of the government generally, but they are more specifically the perceived agents of a particular government: the city. Because an “officer in the line of duty acts as an agent of the government,” and indeed a human agent is the only way a government can act,68Gregory Sisk, Recovering the Tort Remedy for Federal Official Wrongdoing, 96 Notre Dame L. Rev. 1789, 1810 (2021). police officers are essentially the embodiment of the city. Through contact with officers, “[c]itizens learn about how government works and whether government officials value them.”69Lerman & Weaver, supra note 67, at 206. These interactions are thus deeply political—a moment when the citizen and the city come face to face.

B. The Under- and Overpolicing Paradox

The political significance of these encounters heightens the harms that result when they go awry. Interactions between local law enforcement and the public tend to go wrong in one of two ways. First, the problem might be that the encounter doesn’t happen at all: calls for assistance may go unanswered,70Natapoff, supra note 9, at 1723. 911 emergencies calls may be ignored for inordinate periods of time,71Family Sues over Boy’s Neglected 911 Calls, ABC News (June 5, 2019, 11:46 AM), [] (describing failures to respond to 911 calls in a number of American cities). complaints may be inappropriately shelved,72Ashley Southall & Troy Closson, N.Y.P.D.’s Handling of Sex Crimes to Be Investigated by Justice Dept., N.Y. Times (June 30, 2022), [] (describing the NYPD Special Victims Division’s systemic failures to investigate sexual crimes). and certain homicides may never actually be investigated.73See German Lopez, There’s a Nearly 40 Percent Chance You’ll Get Away with Murder in America, Vox (Sept. 24, 2018, 1:20 PM), []. Indeed, the police failure to protect or respond to crime has become one of the central features of many high-crime urban neighborhoods.74Natapoff, supra note 9, at 1722. In these neighborhoods, “[p]olice openly let certain sorts of criminality slide, and residents know that their complaints may be responded to weakly or not at all.”75Id. For instance, in one predominantly Black neighborhood in Los Angeles, each square mile contains forty-one unsolved homicides.76Jill Leovy, The Underpolicing of Black America, Wall St. J. (Jan. 23, 2015, 4:38 PM), []. These distributive impacts connect to an overall low crime-solving rate on the part of police. One recent study estimates that “97% of burglars, 88% of rapists, and over 50% of murderers get away with their crimes.” Shima Baradaran Baughman, How Effective Are Police? The Problem of Clearance Rates and Criminal Accountability, 72 Ala. L. Rev. 47, 55 (2020). Similarly, the high-poverty neighborhood of southern Dallas, Texas experiences a homicide and assault rate twice that of the rest of the city, yet that neighborhood receives the slowest police response times when compared to other Dallas neighborhoods.77Natapoff, supra note 9, at 1726 (“And the officers patrolling those neighborhoods are the department’s least experienced . . . . ‘We’ve abandoned the people and the neighborhoods,’ [says a police chief.]” (first alteration in original)).

Second, the police-citizen encounter itself might occur in a violent, derogatory, disrespectful, or discriminatory way.78 Alex S. Vitale, The End of Policing 2 (2017). The interaction might be the result of racial profiling.79Id. The police may use demeaning words, violence, or unwarranted lethal force.80Id. They might harass citizens in a manner meant to convey that some constituents do not enjoy the benefits of full citizenship in the polis.81See Emily Badger, The Lasting Effects of Stop-and-Frisk in Bloomberg’s New York, N.Y. Times (March 2, 2020), []. They might plant evidence,82See, e.g., Vitale, supra note 78, at 1. unjustifiably engage in arrests on trumped-up charges,83Sam Levin, Beverly Hills: 99% of People Arrested by ‘Safe Streets’ Unit Were Black, Suit Says, Guardian (Sep. 1, 2021, 8:51 PM), []. or execute warrants in reckless ways that endanger both the targeted person and all within the vicinity.84 Vitale, supra note 78, at 25–26; see, e.g., C.J. Ciaramella, Illinois Town Will Pay Million to Family After SWAT Officer Shot 12-Year-Old in Kneecap, Reason (Feb. 1, 2023, 1:39 PM), [].

Paradoxically, then, for communities of color in particular, modern policing manages to create both “the convictions of innocents, over-representation in prison, and deaths of unarmed youths, while also leaving a void when protection is needed.”85Zanita E. Fenton, Disarming State Action; Discharging State Responsibility, 52 Harv. C.R.-C.L. L. Rev. 47, 66 (2017) (citations omitted). A 2009 Department of Justice investigation into policing practices in Maricopa County, Arizona, illustrates this paradox.86Deborah Tuerkheimer, Underenforcement as Unequal Protection, 57 B.C. L. Rev. 1287, 1313–14 (2016). The three-year-long investigation revealed that the Maricopa County Sheriff’s Office had unlawfully “engaged in a ‘pattern or practice of unconstitutional policing’ of Latinos” that included both overpolicing and underpolicing.87Id. The overpolicing practices consisted of racial profiling (“[p]olice were between four to nine times more likely to stop Latino drivers than similarly situated non-Latino drivers”), excessive unlawful detentions, and excessive arrests.88Id. But these practices were simultaneously paired with underpolicing: the same investigation also revealed a general reduction of policing services to the Latino community as well as a gender and racial bias that manifested in an acute failure to investigate sex crimes.89Id. For example, the agency neglected to investigate “more than four hundred cases of sexual assault and child molestation” thought to involve Latino victims.90Id.

Both overpolicing and underpolicing express the same kind of disregard for the wellbeing of others. For example, instances of police violence and excessive force are often followed by a corresponding failure to render aid or medical assistance.91See, e.g., Rebecca Butcher, Breaking Down How the Memphis Police Officers Involved in Tyre Nichols’ Death Broke Department Policy, ABC24 (Jan. 24, 2023, 10:56 PM), []; Becky Sullivan, Three Ex-Minneapolis Police Officers Guilty of Violating George Floyd’s Civil Rights, NPR (Feb. 24, 2022, 10:07 PM), []; Rosh Lowe, New Video Shows Hollywood Police Dragging Man into Elevator After Shooting Him, Local10 (Feb. 3, 2023, 6:02 PM), [] (detailing how thirty minutes elapsed between police officers shooting an innocent person and police officers obtaining medical aid for him).Although they sound like opposites, under- and overpolicing are in fact “corollaries.”92Tuerkheimer, supra note 86, at 1319. In addition to both occurring when police engage in acts of brutality, under- and overpolicing are, at a more macro level, connected parts of “a cycle [or] continuum of targeted police underservice and blindness to crime that eventually serves as its own rationale for overpolicing and pursuing more aggressive tactics.”93Vann R. Newkirk II, What We Are Getting Wrong About Police Reform, Gawker (Nov. 9, 2015, 12:50 PM), []. Beneath this cycle is the “idea that [some groups] have fewer civil rights, that the task of the police is to keep them under control, and that the police have little responsibility for protecting them from crime within their communities.”94Aya Gruber, Policing and “Bluelining, 58 Hous. L. Rev. 867, 880 (2021). This attitude is also evident in Coffee County District Attorney Craig Northcott’s expressed refusal to bring domestic violence charges when that violence occurred within a same-sex relationship. Relying on his “prosecutorial discretion,” he stated that “the social engineers on the Supreme Court now decided we have homosexual marriage. I disagree with them.” He later added, “There’s no marriage to protect with homosexual relationships, so I don’t prosecute them as domestic.” Deanna Paul, This ‘Good Christian’ Prosecutor Is Overlooking Domestic Violence Charges for Same-Sex Couples, Wash. Post (June 5, 2019, 3:35 PM), [].

C. Legal Responses to Under- and Overpolicing

For individuals and communities harmed by these over- and underpolicing practices, the current legal remedies are demonstrably inadequate. Prospective plaintiffs face a doctrinal “obstacle course” teeming with legal barriers that they “must climb over and crawl under” in order to have a viable claim against the police.95 Harmon, supra note 42, at 612. These hurdles were purportedly meant to “weed out insubstantial cases, save officials time and money, and avoid inappropriate interference in local government,” but in practice they have been so onerous that even meritorious suits rarely achieve success in the courtroom.96Id. Such suits do, however, sometimes result in settlements.

Theoretically, Section 1983 of the federal Civil Rights Act offers victims of unconstitutional excessive force or unlawful search and seizures a potential path of recourse against both individual police officers and the cities that employ them.97In Monell v. Department of Social Services of New York, 436 U.S. 658 (1978), the Supreme Court clarified that although states are not “persons” for the purposes of Section 1983, police departments, towns, cities, counties, and other local agencies or governments can be held liable as “persons” under Section 1983. However, the Court also made clear that municipalities are liable only for their own actions and are not vicariously liable for the conduct of individual officers. Id. at 692.

Municipal liability may be based on “(1) a formally promulgated policy; (2) a well-settled custom or practice; (3) a final decision by a municipal policymaker; or (4) deliberately indifferent training, supervision, or screening of employees,” though it is in general quite challenging to bring a successful claim on any of these bases. Harmon, supra note 42, at 633 (citing Martin A. Schwartz, Section 1983 Litigation § 7.06 (3d ed. 2014)).
First enacted in 1871, Section 1983 was actually a response to the problem of underpolicing.98Olivia B. Waxman, The Tyre Nichols Case Shows How Police Have Become Increasingly Unaccountable, Historian Says, Time (Feb. 3, 2023, 8:34 AM), []. As Professor Joanna Schwartz stated: Section 1983 “was a product of the Reconstruction after the Civil War, when Black Americans were being assaulted and killed by white supremacist groups, and law enforcement was either participating or standing idly by.” Id. The statute housing Section 1983 was “[e]nacted in 1871 and referred to as the Ku Klux Klan Act,” and “it gave people the power to bring claims in federal court, instead of relying on state court where Black people weren’t even allowed to testify.” Id. It was designed to address the historic refusal of police in many states “to enforce the law to protect African-Americans.”99Julie Goldscheid, Rethinking Civil Rights and Gender Violence, 14 Geo. J. Gender & L. 43, 64 (2013). Now, however, Section 1983 is almost exclusively used in suits over police behavior as a remedy for the problems associated with overpolicing.100See infra notes 118–134 and accompanying text.

Unfortunately, Section 1983 is not a meaningful remedy, in large part because of the doctrine of qualified immunity.101But see Joanna C. Schwartz, After Qualified Immunity, 120 Colum. L. Rev. 309 (2020) (arguing that eliminating qualified immunity would not affect the vast majority of outcomes in Section 1983 litigation). Qualified immunity operates to shield “government officials performing discretionary functions” from liability unless their conduct violates “clearly established statutory or constitutional rights of which [a] reasonable person would have known.”102Harlow v. Fitzgerald, 457 U.S. 800, 818–19 (1982). Under this standard, “only those officers who are ‘plainly incompetent’ or who ‘knowingly violate the law’ may be held liable,”103 Harmon, supra note 42, at 621 (quoting Kisela v. Hughes, 138 S. Ct. 1148, 1152 (2018)). and “the contours of the right must be sufficiently clear that a reasonable official would understand that what he is doing violates that right.”104Anderson v. Creighton, 483 U.S. 635, 640 (1987). In practice, qualified immunity has shielded officers who have engaged in facially brutal activities, like body-slamming an alleged shoplifter so hard that he “suffered a broken collarbone, a skull fracture that required 52 stitches, and a traumatic brain injury that caused permanent mental impairment”;105McDaniel v. Neal, 44 F.4th 1085, 1089 (8th Cir. 2022). shooting a ten-year-old child when trying to shoot the family dog;106Corbitt v. Vickers, 929 F.3d 1304 (11th Cir. 2019). killing a distressed man with mental illness outside his hospital room;107See Aaron L. Nielson & Christopher J. Walker, Qualified Immunity and Federalism, 109 Geo. L.J. 229 (2020). and shooting a man who was sleeping in his car after he woke up and attempted to slowly drive away.108Stewart v. City of Euclid, 970 F.3d 667 (6th Cir. 2020). Notably, although qualified immunity applied to the Section 1983 claims, the claims based on Ohio state law were remanded to the district court to determine whether they could survive summary judgment. Outcomes like this have prompted numerous scholars and policymakers to critique qualified immunity and suggest its elimination.109For example, the “George Floyd Justice in Policing Act,” the “Ending Qualified Immunity Act,” and the “Reforming Qualified Immunity Act” have been introduced in Congress in recent years. Although none of these bills were passed, qualified immunity reform remains a point of discussion. Jay Schweikert, Qualified Immunity, Am. Bar Assoc. (Dec. 17, 2020), []. These proposals are gaining traction and support,110Id. but as of the time of this writing, qualified immunity continues to stand as a formidable impediment to accountability for police wrongdoing.111Even if qualified immunity were eliminated, an enforceable duty of care flowing from cities to constituents would still be valuable. See infra Part III.

In addition to this impediment, the usual rules of respondeat superior for employers and employees do not apply in the Section 1983 local government context. Instead, in order to make the city itself liable, a plaintiff must overcome the Monell test and show that “the local government had an unlawful policy or custom that caused their employee to violate the Constitution.”112Schwartz, Municipal Immunity, supra note 35, at 1185. This hurdle is so onerous that one recent study of the federal court docket found that “very few” of these claims “made it to trial; even fewer succeeded.”113Id. “The statute should also impose vicarious liability on local governments for wrongs committed by their officers, instead of requiring plaintiffs to meet the challenging Monell standard.” Reinert et al., supra note 30, at 769.

Fortunately, Section 1983 is not the only potential pathway to police liability: state tort law provides an alternative avenue of relief.114Reinert et al., supra note 30, at 759 (“States typically allow victims to pursue state law tort claims, for such common law torts as assault and battery, but often impose limits on and immunities from such liability to protect official defendants.”). Although states cannot reform qualified immunity at the federal level, they can still pass state laws analogous to Section 1983 and expressly preclude a qualified immunity defense to those causes of action. One such example of a state civil rights law is Colorado’s “Law Enforcement Integrity and Accountability Act,” which passed in 2020 with bipartisan support. Schweikert, supra note 109. But, like Section 1983, tort suits for police misconduct are difficult to win.115See Reinert et al., supra note 30, at 755–56. Additional governmental and sovereign immunities, special procedural rules, and damage caps all impede the possibility of successful tort suits.116 Harmon, supra note 42, at 630. Even though municipalities frequently enter into settlements for high-profile police brutality claims, if the municipality is not so inclined, plaintiffs face significant difficulty when suing the police for affirmative wrongdoing.117Id.

When the basis for a suit is not affirmative conduct but instead an officer’s failure to act, things get even more difficult. Overpolicing is a difficult problem to legally remedy; underpolicing is even more so. Indeed, Section 1983 does not even apply to underpolicing: the Supreme Court held in two seminal cases, DeShaney v. Winnebago County Department of Social Services118DeShaney v. Winnebago Cnty. Dep’t of Soc. Servs., 489 U.S. 189 (1989). and Town of Castle Rock v. Gonzales,119Town of Castle Rock v. Gonzales, 545 U.S. 748 (2005). that state failures to protect do not trigger constitutional protection under Section 1983.120Natapoff, supra note 9.

In DeShaney v. Winnebago, Joshua DeShaney sued the Winnebago County Department of Social Services for failing to protect him from his father’s extreme beatings that left him partially paralyzed, severely brain-damaged, and in need of constant care for the rest of his shortened life.121DeShaney, 489 U.S. 189. The Supreme Court term in which DeShaney was heard was “the first in over forty years in which conservative justices held a majority on the Court.” Lynda G. Dodd, DeShaney v. Winnebago County: Government Neglect and the “Blessings of Liberty,” in Civil Rights Stories 189, 200 (Myriam E. Gilles & Risa L. Goluboff eds., 2008). Joshua DeShaney died on November 9, 2015, when he was thirty-six years old. Crocker Stephenson, Boy at Center of Famous ‘Poor Joshua!’ Supreme Court Dissent Dies, Milwaukee J. Sentinel (Nov. 11, 2015), []. Despite the fact that Joshua was officially “under the care and supervision of the Winnebago County Department of Social Services” at the time of the final beating that resulted in these catastrophic injuries (as he had been for fourteen months),122Dodd, supra note 121, at 186. the Court rejected the argument that the Department’s practice of returning Joshua to his father’s custody without doing more than “dutifully record[ing]” the abusive incidents in their case files was a constitutional violation.123DeShaney, 489 U.S. at 193. According to the majority, “nothing in the language of the Due Process Clause itself requires the State to protect the life, liberty, and property of its citizens against invasion by private actors.”124Id. at 195. Nevertheless, the Court acknowledged that exceptions existed, including if a plaintiff could show that the state actor established a “special relationship” with them or if the state actor made the plaintiff’s situation worse than it would otherwise have been. Goldscheid, supra note 99, at 67–68. In other words, Joshua’s claim was unsuccessful because it failed to satisfy the “state action requirement” of the Fourteenth Amendment, under which there can only be a substantive due process violation if a state actor is the direct cause of harm.125See Natapoff, supra note 9, at 1758.

Town of Castle Rock v. Gonzales also involved disturbing failure on the part of governmental actors.126Town of Castle Rock v. Gonzales, 545 U.S. 748 (2005). There, the plaintiff mother had a restraining order in place against her abusive estranged husband, requiring him to remain away from her and her children except during specified visiting times.127Id. at 751–53. The restraining order explicitly required police to make reasonable attempts to arrest if the order were violated.128Id. at 751. One evening, the estranged husband kidnapped the three daughters from the plaintiff’s front yard.129See id. at 753. The plaintiff, profoundly concerned for the children’s safety, telephoned the police multiple times, went to the station, and pled with officers to enforce the order and attempt an arrest of her ex-husband. The police refused, at one point calling her requests “a little ridiculous.”130Jessica Gonzales’ Statement Before the IACHR, ACLU (Oct. 27, 2014), []. At 3:20 a.m. that morning, her husband drove to the police station, parked his truck in front of it, and began shooting at the officers inside.131Gonzales, 545 U.S. at 754. He was killed, and the bodies of the plaintiff’s three daughters were discovered in the cab of the truck.132Id.

As in DeShaney, the Court held that no constitutional relief was available. The Court held that “even where a court orders police to perform a protective function—in that case by issuing a domestic violence protective order—and even where state law mandates police protection, this does not give rise to a protected property interest as a matter of procedural due process.”133Natapoff, supra note 9, at 1758. The Court also noted that “the framers of the Fourteenth Amendment and the Civil Rights Act of 1871 did not create a system by which police departments are generally held financially accountable for crimes that better policing might have prevented.”134Gonzales, 545 U.S. at 768–69 (citation omitted). According to the holdings of Deshaney and Gonzales, the Constitution does not require a duty to protect, even when legislative or judicial mandates seem to impose such a duty.135Natapoff, supra note 9, at 1758.

Responding to the uproar that followed the release of each of these decisions, some scholars highlighted the federalism concerns running through these cases.136Gugel, supra note 2, at 1308. On this view, a dominating concern in the decisions was that if the Court were to read the Fourteenth Amendment as imposing a positive obligation on states to protect, the judiciary would undermine state autonomy.137Id. These scholars argued that a federal constitutional duty to protect would thus end up operating like “a decree that state governments employ their budgets for specific social services,” and consequently state officials would no longer have discretion to act in accordance with the particular political vision, manner of resource allocation, and identity that the state had set.138Id. at 1309.

Commentators adopting this read therefore argued that the Court’s reluctance to provide relief in federal court was actually premised on a belief that state law is the appropriate place for such plaintiffs to seek relief.139Id. at 1310. Indeed, in both the DeShaney and the Gonzales opinions, the majority explicitly suggested that state tort law claims were a more appropriate avenue. In DeShaney, the majority stated that while there was no constitutional violation, basic common law duties in tort may have been implicated.140Deshaney v. Winnebago Cnty. Dep’t of Soc. Servs., 489 U.S. 189, 201–02 (1989). Citing the Restatement (Second) of Torts, the Court noted that “by voluntarily undertaking to protect Joshua against a danger it concededly played no part in creating,” the State may have “acquired a duty under state tort law to provide him with adequate protection against that danger.”141Id. The Court cited the restatement for the principle that “one who undertakes to render services to another may in some circumstances be held liable for doing so in a negligent fashion.” Id. at 202 (citing Restatement (Second) of Torts § 323 (Am. L. Inst. 1965)). The Court also cited a leading treatise “discussing ‘special relationships’ which may give rise to affirmative duties to act under the common law of tort.” Id. (citing W. Page Keeton, Dan B. Dobbs, Robert E. Keeton & David G. Owen, Prosser and Keeton on the Law of Torts § 56 (5th ed. 1984)).

In DeShaney, the Court noted in dicta that even though “the Fourteenth Amendment . . . does not transform every tort committed by a state actor into a constitutional violation” and not “all common-law duties owed by government actors were . . . constitutionalized by the Fourteenth Amendment,” those duties may nevertheless persist and be cognizable under state tort law.142Id. at 202 (quoting Daniels v. Williams, 474 U.S. 327, 335 (1986)). The Court opined that Wisconsinites were free to design their state tort law in any way they liked and that they should be able to do so without federal overreach: “The people of Wisconsin may well prefer a system of liability” under which the State would be responsible for such failures to protect, “[b]ut they should not have it thrust upon them by this Court’s expansion of the Due Process Clause of the Fourteenth Amendment.”143Id. at 203. The majority in Gonzales echoed this sentiment, relying on DeShaney to acknowledge the Court’s “continuing reluctance to treat the Fourteenth Amendment as ‘a font of tort law,’ ” but that “the people of Colorado are free to craft such a system [of liability] under state law.”144Town of Castle Rock v. Gonzales, 545 U.S. 748, 768–69 (2005) (quoting Paul v. Davis, 424 U.S. 693, 701 (1976)).

To a large extent, though, states have not crafted such systems of liability.145Nonetheless, at least four states (Colorado, New Mexico, Connecticut, and Massachusetts) have passed reforms both allowing plaintiffs to bring suits alleging law enforcement officers violated certain state constitutional provisions and disallowing qualified immunity as a defense to those. See Reinert et al., supra note 30, at 739–40. Police misconduct like the excessive use of force or inappropriate arrest can be actionable in state tort law under causes of action like battery, assault, and false imprisonment—and failures to protect and affirmative wrongful conduct can prompt state tort suits in negligence—but most state tort law systems deploy a complicated matrix of common law and statutory tools like immunity rules and damage caps to limit governmental liability.146Cassandra R. Cole & Chad Marzen, A Review of State Sovereign Immunity Statutes and the Management of Liability Risks by States, 32 J. Ins. Regul. 45, 49 (2013). For example, state tort law was not a viable option in the DeShaney case because of a $50,000 state-legislated damages cap that applied anytime Wisconsin was a defendant.147 Wis. Stat. § 893.80(3) (2012). This amount would not cover even one year of Joshua’s annual medical expenses.148Dodd, supra note 121, at 196. The federal lawsuit sought “ million in compensatory damages and million in punitive damages.” Id. at 196 n.63. Damage caps may also have dissuaded the plaintiff in Gonzales, as the damage cap in Colorado in 2013 was 0,000 for an individual claim (an increase from the previous amount of 0,000). Changes to Government Immunity Caps in Colorado, Levine Law (Nov. 8, 2013), []. Damage caps have raised their own constitutional issues. “For example, a court in Missouri struck down the state’s non-economic damages cap of 0,000, stating that it violated a plaintiff’s constitutional right to have a jury determine the amount of damages that are to be awarded.” Id. The general position of state and local governments is usually to disavow and limit financial obligations to those injured by either governmental actions or failures to act.149Cole & Marzen, supra note 146, at 48.

II. The Public Duty Doctrine

The public duty doctrine assists in this disavowal. An important part of the civil rights ecosystem,150See Joanna C. Schwartz, Civil Rights Ecosystems, 118 Mich. L. Rev. 1539 (2020). the public duty doctrine holds that although governmental entities may owe duties to the public, individuals cannot legally enforce them.151See infra Section II.A. Functionally, this means that when plaintiffs harmed by police inaction—or, sometimes, action—assert state tort law claims of negligence against municipal governments, cities often defend on the basis that under the public duty doctrine, they owe no legally enforceable duty of care to their constituents.152Id.

This Part describes the public duty doctrine and situates it within the broader context of municipal immunity. This Part then surveys the arguments both in favor of and against the retention of this no-duty rule, and concludes that eliminating the doctrine would increase accountability, further democratic goals, and create consistency with the thick conception of the city-constituent relationship which municipalities advance in the affirmative litigation context.153See infra Section II.C.4.

A. Public Duty Doctrine and the Landscape of Municipal Immunity

Tort law falls generally under the domain of the state, and states have multiple doctrines and legislative shields that protect them and local governments from civil liability. As a starting position, states historically benefited from sovereign immunity,154The doctrine of sovereign immunity was believed to have inured itself in the states prior to the creation of the federal Constitution. Once the federal Constitution was enacted, however, it was unclear what Article III—which provides that “[t]he judicial Power shall extend . . . to Controversies . . . between a State and Citizens of another State . . . and between a State . . . and foreign States, Citizens or Subjects”—meant for state sovereign immunity. In Chisolm v. Georgia, 2 U.S. 419 (1793), the Supreme Court helped resolve the issue by “abolish[ing] the doctrine of sovereign immunity” in these regards. State Sovereign Immunity and Tort Liability in All 50 States, Matthiesen, Wickert & Lehrer, S.C. (Jan. 13, 2022), []. However, the fallout from that decision led to the passage of the Eleventh Amendment, which “reinstated states’ sovereign immunity, at least to the extent that Article III encroached upon it,” and by the beginning of the nineteenth century, “this sovereign immunity was adopted by nearly every state.” Id. At the turn of the next century, however, the Supreme Court ruled in Ex parte Young, 209 U.S. 123 (1908), that an unconstitutional action by the state could defeat sovereign immunity claims. Then, following the federal government’s waiver of sovereign immunity for torts with the passage of the Federal Torts Claims Act, state legislatures followed suit and started passing their own state statutes waiving immunity. Id. a holdover from the English principle that “the King can do no wrong.”155Erwin Chemerinsky, Against Sovereign Immunity, 53 Stan. L. Rev. 1201, 1202 (2001). Over time, states waived this blanket immunity insofar as it applied to certain lawsuits from their constituents, opening themselves up to some forms of liability.156For example, in 1929, New York State waived its sovereign immunity in certain cases. In 1946, Congress passed the Federal Tort Claims Act waiving some of its sovereign immunity, and by the 1970s, many states had followed suit. See Dan B. Dobbs, Paul T. Hayden, & Ellen M. Bublick, Torts and Compensation 476 (8th ed. 2005). However, states have not made themselves vulnerable to every kind of liability. As a leading casebook puts it:

[T]hese waivers of immunity have been partial rather than total, and usually set up special procedural rules designed to provide greater protection from liability for governmental entities than are enjoyed by private actors. As a result, the present contours of government liability at the federal, state, and local levels are quite complex.

John C.P. Goldberg, Leslie Kendrick, Anthony J. Sebok & Benjamin C. Zipursky, Tort Law: Responsibilities and Redress 543 (5th ed. 2021).
However, states have also implemented complicated matrices of fine doctrinal distinctions, damage caps, and statutory forms of immunity to keep state tort liability to a minimum.157 Goldberg et al., supra note 156, at 543.

In contrast to the original widespread acceptance of state sovereign immunity,158See supra note 154. municipal immunity has always been a more malleable concept.159 Briffault et al., Cases and Materials on State and Local Government Law 621 (9th ed. 2022). The public duty doctrine thus occurs within a broader municipal immunity landscape that is both complex and continually contested.160Smith, supra note 40, at 477. As “municipal corporations,” cities have a unique hybrid legal status, rendering their ability to partake in state sovereign immunity questionable.161Historically, counties in some states shared in state sovereign immunity. The court in Madden v. Lancaster, 65 F. 188, 191 (8th Cir. 1894), reasoned that immunity was appropriate because “[a]ll the powers” entrusted to counties “are the powers of the state, and the duties imposed upon them are the duties of the state.” See Smith, supra note 40, at 426–27. For example, as a historical matter, some courts were comfortable ascribing “quasi-sovereign status” to counties—a local government unit thought of as the state’s administrative arm—but those same courts nevertheless held that the ambiguous municipal corporate status of cities took them out of this ambit.162Smith, supra note 40, at 427. On the other hand, some courts took the opposite approach and did rely on notions of state sovereign immunity to shield municipalities from liability.163Id. at 428.

Other courts held that, leaving aside the issue of sovereign immunity, municipalities should receive immunity for policy reasons.164Id. at 430–40. A pivotal moment in municipal immunity took place in 1788 when, in the case of Russell v. Men of Devon, the Court of the King’s Bench in England gave immunity to the defendant town of Devon. 165Russell v. Men of Devon (1788) 100 Eng. Rep. 359, 362 (KB). In Russell, the plaintiffs had suffered injuries after a bridge failure and brought suit against the town. The court granted the town immunity, finding that “the novelty” of the plaintiff’s claim, “an infinity of actions” that would result, and the “importance of unimpeded public functions” all weighed in the town’s favor.166Smith, supra note 40, at 425; Russell, 100 Eng. Rep. at 361–62. One of the deciding judges specifically stated that if municipal liability were to be imposed, the public would be harmed, and “it is better that an individual should sustain an injury than that the public should suffer an inconvenience.”167Russell, 100 Eng. Rep. at 362. In the nineteenth century, some state courts, including those in Massachusetts, New York, South Carolina, New Jersey, and Mississippi, adopted Russell in their respective jurisdictions to grant municipal immunity against common law tort liability.168Smith, supra note 40, at 425–26.

Even states that generally granted immunity to municipalities, however, sometimes had special statutes that opened municipalities up to liability in certain circumstances.169Heyman, supra note 5, at 541. For example, one historical source of liability was found in “statutes of incorporation.”170Id. at 539. These statutes provided a path for individuals to bring suit against local governments when state sovereign immunity prevented an individual from obtaining redress for an officer’s failure to protect.171Where these statutes were not in place, courts generally held that municipalities should receive immunity because they were performing functions which the state delegated to them. Id. These statutes hearkened back to the English tradition of “communal liability,” where a local community (known as “the hundred”) had a duty to pursue felons and to compensate victims in instances where it failed to fulfill that duty.172Id. at 542. The Riot Act of 1714, for instance, imposed liability on the hundred when riots resulted in property damage.173Id. Eventually, in England, these statutes delegated to municipal governments—rather than the hundred—responsibility for protecting their constituents and property.174Id. at 534–44.

In the mid-eighteenth century, many states adopted statutes similar to these English acts.175Id. at 542. These new state laws essentially “reaffirmed the community’s duty of protection and made it the basis of a legal action” free from an immunity rule.176Id. By the time the Fourteenth Amendment was ratified in 1868, “at least nine states” had similar riot laws, and in the mid-nineteenth century, plaintiffs used these statutes to sue local governments for failures to protect them from violence.177Id.

In the century following the ratification of the Fourteenth Amendment, the landscape of municipal immunity was muddled and had few observable trends. By the mid-twentieth century, however, there was a clear trend toward abolishing municipal immunity and allowing at least some forms of liability.178. Smith, supra note 40, at 424 n.64. Some courts became increasingly opposed to ideas of sovereign immunity generally, finding it anti-American and noting that “the Revolutionary War was fought to abolish that ‘divine right of kings’ on which the theory is based.”179E.g., Molitor v. Kaneland Cmty. Unit Dist., 163 N.E.2d 89, 94 (Ill. 1959). These courts found the policy choice to absolve the government of liability while requiring an injured individual to solely bear the consequences of that harm grotesque,180Id. at 93–94. with one court declaring it “incredible” that an injured individual should shoulder the burden of damage caused by the government instead of more “justly” spreading the loss across the entire community.181Id. at 94 (quoting Barker v. City of Santa Fe, 136 P.2d 480, 482 (N.M. 1943)).

Nevertheless, in the 1980s, courts began to favor municipal immunity once again, largely because a national insurance crisis meant liability insurance rates skyrocketed at the same time as municipalities were faced with very large damages awards.182 Briffault et al., supra note 159, at 822. Today, the contemporary municipal immunity landscape is “a patchwork of state constitutional provisions and state tort law and immunity statutes.”183Id at 1010. Sometimes municipalities are included within state tort claims statutes such that “municipal immunity parallels state government immunity”; other times states and municipalities have different immunity rules.184See also Dan B. Dobbs, The Law of Torts § 149, at 334 (2000). Overall, the landscape is such that “[f]lexible, juridically created doctrines, along with frequent statutory modification, expand and contract the limits of governmental immunity as public sensibilities change.”185. Briffault et al., supra note 159, at 1010.

Within this landscape, one common approach is to make municipal immunity turn on which side of a binary, categorical test the impugned action falls on. Often framed in terms of a governmental/proprietary or discretionary/ministerial test, these tests are, in practice, notoriously indeterminate and exceedingly difficult to apply.186Id. at 1009–35. The governmental/proprietary test is meant to correspond with the dual nature of local governments as both a public, governmental entity and a private, corporate one.187Smith, supra note 40, at 426. Under this approach, local governments cannot be sued in their governmental capacity, but may be liable if acting in their corporate capacity.188Id. Other courts rely on a discretionary/ministerial distinction, under which an act requiring the exercise of judgment is immune, while one which is routine or prescribed may create liability.189 Eugene McQuillin, The Law of Municipal Corporations § 53:15 (3d ed. 2005), Westlaw (database updated June 2023). For example, a police failure to follow specific protocol could be a ministerial act, and thus subject to potential liability.190See, e.g., Robert Weisberg, Preventing Crime: Private Duties, Public Immunity, 2 J.L. Econ. & Pol’y 365, 379 n.66 (2006) (“[P]olice can be held liable when they failed to comply with [their] own protocol for responding to emergency calls.”) (discussing Austin v. City of Scottsdale, 684 P.2d 151 (Ariz. 1984)); see also Ferreira v. City of Binghamton, 975 F.3d 255, 272 (2d Cir. 2020) (where the failure to follow “established police procedures and acceptable police practice” rendered the governmental immunity defense inapplicable).

In addition to specific immunity rules like these, courts have also created liability “exemptions” that, despite being formally different, functionally grant municipalities further immunity.191. Goldberg et al., supra note 156, at 562. The most significant of these is the public duty doctrine.192Id. The public duty doctrine is recognized in most jurisdictions193See, e.g., Cope v. Utah Valley State Coll., 290 P.3d 314 (Utah Ct. App. 2014); Dobbs et al., supra note 156, at 501. That said, there is wide variation in how jurisdictions understand the applicability of the doctrine. Some courts hold that it only applies to statutory duties. See, e.g., Dobbs, supra note 184, § 345, at 375 n.4 (“[P]ublic duty rule relieves public entity of a duty of care ‘only when an action is founded upon a statutory duty; when duty is based on common law, then its existence is analyzed as it would be with a private defendant which is not a government entity;’ immunity is a separate issue.” (citing Madison ex rel. Bryant v. Babock Ctr., Inc., 638 S.E.2d 650, 660 (S.C. 2006))). Other courts find that the doctrine applies to both statutory and common law duties. Id. at 375. Some courts hold that the doctrine only applies to nonfeasance; other courts hold that it applies to misfeasance as well. Id. at 375 n.6 (“[A]pparently barring relief for negligent police affirmative acts of police in investigating murder, which led to second murder . . . .” (citing Varner v. District of Columbia, 891 A.2d 260 (D.C. 2006))). Some courts also hold that the presence of liability insurance voids the doctrine. Id. (“[W]here by law the entity’s purchase of insurance waives immunity, insurance coverage waives public duty doctrine . . . .” (citing Kunzie v. City of Olivette, 184 S.W.3d 570 (Mo. 2006))). Additionally, “[s]ome courts have said that the public duty doctrine will not apply to protect the entity when it is guilty of egregious misconduct, intentional wrongdoing, malice, or recklessness.” Id. § 346, at 378 (footnote omitted). and has been described as “perhaps the most daunting of the remaining hoops that plaintiffs in many jurisdictions still have to jump through to sue a municipality.”194 McQuillin, supra note 189, § 53:21 (footnote omitted) (noting that “[t]he public duty doctrine generally applies to governmental functions, but not to proprietary ones”). Often described by the phrase “a duty to all is a duty to none,” the public duty doctrine in broad strokes provides that any duty a government owes to the general public “does not impose a specific duty of due care on the government with respect to individuals who may be harmed by governmental action or inaction, unless there is some specific connection between the government agency and the individuals that makes it reasonable to impose a duty.”195Eklund v. Trost, 151 P.3d 870, 878 n.2 (Mont. 2006). In other words, the public duty doctrine holds that, as a threshold matter, the duties a municipality owes to the general public are moral or political, but barring some special circumstance, “no individual has ‘standing’ ” to legally enforce them in a tort suit.196. Goldberg et al., supra note 156, at 562. Cities typically turn to this no-duty rule when plaintiffs allege that a local government has carelessly failed to protect them from third-party harm, and some assert this defense when a police officer is the direct source of the harm as well.197Id. at 562–64.

The public duty doctrine is a judicially created rule, crafted by the Supreme Court in South v. Maryland in 1855.198South v. Maryland, 59 U.S. 396 (1855); see Briffault et al., supra note 159, at 621. In that case, the plaintiff sued a sheriff for failing to protect him from a kidnapping, but the Court held that the police were “not liable for injuries to an individual member of the community, even where the police unreasonably neglected to assert their authority.”199G. Kristian Miccio, If Not Now, When? Individual and Collective Responsibility for Male Intimate Violence, 15 Wash. & Lee J. C.R. & Soc. Just. 405, 429 (2009).

Since South, many state courts have adopted the public duty doctrine.200Id. It has achieved wide—but not universal—acceptance and is frequently used to functionally immunize both states and their local governments from civil liability.201Id. In particular, it has gained significant traction shielding police from liability.202Id. at 429–30. Importantly, when applied in the police context, the doctrine characterizes police—enormously important representatives of government—as owing no duty distinct from that of a private citizen.203Id. at 429. Just as the average individual person owes no duty to rescue others, protect others from third-party criminality, or prevent crime from happening, so too is there no legally enforceable “duty to rescue, protect or prevent crime on the part of the police.”204Id. Some courts explicitly argue that parity between private and public liability is a good reason to have the public duty doctrine. For example, in City of Rome v. Jordan, the Georgia Supreme Court “stressed parity between governmental and private tortfeasors” as the “primary ground for embracing the public duty doctrine.” R. Perry Sentell, Jr., Georgia’s Public Duty Doctrine: The Supreme Court Held Hostage, 51 Mercer L. Rev. 73, 77 (1999). Yet, given the difference between cities and private individuals in terms of resources, power, and structural positioning, it is not at all clear why parity in this context would be appropriate. As iterated earlier, the government is tasked with its “first job” of protecting its people; individuals are entirely differently situated. Moreover, the wisdom and validity of the no-duty-to-rescue doctrine as applied to individuals is itself the subject of widespread and scathing critique. See, e.g., Sarah L. Swan, Bystander Interventions, 2015 Wis. L. Rev. 975 (2015).

The most well-known case illustrating the public duty doctrine in action is likely Riss v. City of New York.205Riss v. City of New York, 240 N.E.2d 860 (N.Y. 1968). In that case, the plaintiff, Linda Riss, was dating an attorney, Burton Pugach. After discovering he was already married, she tried to terminate the relationship, and he repeatedly threatened her with death or maiming if she did so.206Id. at 862 (Keating, J., dissenting). Frightened, Riss repeatedly begged the police for protection, but to no avail. Several months later, on the occasion of a party celebrating the plaintiff’s engagement to a new suitor, Riss received a telephone call warning her that this was her “last chance.”207Id. Again, she asked the police for protection, and again they declined. The following morning, Pugach arranged for men to attack Riss and throw lye in her face, which left her partially blind and permanently scarred.208Id. She sued New York City for negligence, but the trial court gave a directed verdict to the city and the New York Court of Appeals affirmed.209Id. at 861.

Since Riss, municipalities often deploy the public duty doctrine to shield themselves from liability for even egregious actions and failures to protect.210Another example of an egregious public duty case is Gonzales v. City of Bozeman, 217 P.3d 487 (Mont. 2009). Although the public duty doctrine began in the context of nonfeasance or a failure to act, the line between misfeasance and nonfeasance is notoriously blurry, and courts have applied and continue to apply the doctrine to shield municipalities from liability for overt acts. Dobbs et al., supra note 156, § 346, at 381–82. Some courts, though, explicitly state that the public duty doctrine should not apply to affirmative acts of harm. See, e.g., Scott v. City of Charlotte, 691 S.E.2d 747, 752 (N.C. App. Ct. 2010). For example, in Ferreira v. City of Binghamton, the plaintiff was shot in the stomach during a botched execution of a no-knock warrant and the city invoked the public duty doctrine to avoid liability.211Ferreira v. City of Binghamton, 194 N.E.3d 239, 244 (N.Y. 2022). In Ferreira, the New York State Court of Appeals considered the following certified question:

Does the “special duty” requirement—that, to sustain liability in negligence against a municipality, the plaintiff must show that the duty breached is greater than that owed to the public generally—apply to claims of injury inflicted through municipal negligence, or does it apply only when the municipality’s negligence lies in its failure to protect the plaintiff from an injury inflicted other than by a municipal employee?

Id. at 245 (quoting Ferreira v. City of Binghampton, 975 F.3d 255, 291 (2020)). In answering that the New York version of the public duty doctrine applies to affirmative acts of municipal negligence, not just to municipal failures to protect individuals from private third-party violence, the court stated,

We have never limited the special duty rule to instances in which it is alleged that a non-governmental third party directly inflicted the harm. Instead, we have applied the special duty rule where a party asserts a negligence claim against a municipality acting in its governmental capacity, regardless of who most immediately inflicted the injury.

Id. at 249. A strongly worded dissent noted that in fact many prior decisions were at least ambiguous on this point, and would have held that that it was not necessary for a plaintiff to establish a special duty in circumstances where a municipal employee like a police officer has engaged in an affirmative act causing harm. Id. at 254 (Wilson, J., dissenting). For a discussion of this case, see John C.P. Goldberg and Benjamin C. Zipursky, Getting the Law Right: An Essay in Honor of Aaron Twerski, 18 Brook. J. Corp. Fin. & Com. L. (forthcoming 2024).
And in Varner v. District of Columbia, a deaf college student was murdered after a grossly negligent police investigation misidentified the killer of another student, leaving the actual murderer free to kill again.212Varner v. District of Columbia, 891 A.2d 260, 264 (D.C. 2006). The court held that even though the evidence suggested that the police both negligently investigated one homicide and negligently failed to prevent another, the public duty doctrine nonetheless barred the claim.213Id. at 276; see also Carl Rizzi, A Duty to Protect: Why Gun-Free Zones Create a Special Relationship Between the Government and Victims of School Shootings, 25 Cornell J.L. & Pub. Pol’y 499 (2015).

Despite such holdings, the public duty doctrine is not absolute, and exceptions often soften the application of the rule.214 Goldberg et al., supra note 156, at 562. These exceptions mostly map onto those governing the common law duty to rescue—namely, that the public duty doctrine will not apply if the government created the risk of harm, assumed a duty through an undertaking, or is in a special relationship with the plaintiff, nor will it apply if legislative intent weighs in favor of liability.215Id.

These exceptions can mitigate the severity of the public duty doctrine. For example, in Beltran-Serrano v. City of Tacoma, the Supreme Court of Washington declined to apply the public duty doctrine where a police officer shot (multiple times) a non-English-speaking homeless man with mental illness who “became scared and started to run away” when the officer requested his identification.216Beltran-Serrano v. City of Tacoma, 442 P.3d 608, 610 (Wash. 2019). The trial court had granted the city’s motion to dismiss on the basis of the public duty doctrine, but the Washington Supreme Court held that there was potential liability in this case. First, the court noted that a negligence claim could coexist even in the presence of an intentional tort claim, finding that “ordinary negligence principles apply in situations that involve both a claim of battery or unprivileged use of force and the duty to act reasonably in carrying out law enforcement functions.”217Id. at 612. Over two dissents, the court held that even though the incident culminated in an intentional shooting, the negligence claim that the officer “unreasonably failed to follow police practices calculated to avoid the use of deadly force” was still viable.218Id. at 611. Further, the court found that the “affirmative interaction” between the officer and the plaintiff rendered the public duty doctrine inapplicable to the negligence claim.219Id. at 615.

As the ruling in Beltran-Serrano suggests, many courts recognize an “affirmative interaction” exception and hold that where a police officer directly causes an injury to the plaintiff, the public duty doctrine does not apply.220In Jones v. State, the court held that the public duty doctrine’s application is limited to where “the breach of the alleged duty has involved the governmental entity’s negligent control of an external injurious force or of the effects of such a force.” Jones v. State, 38 A.3d 333, 346 (Md. 2012). Citing Liser v. Smith, 254 F. Supp. 2d 89, 102 (D.D.C. 2003), the court also noted that the public duty doctrine

is wholly inapposite in a case such as this, where the alleged harm was brought about directly by the officers themselves . . . . The claim that the government has no general duty to protect particular citizens from injury is simply a non-sequitur where the government itself is solely responsible for that injury, which it has caused by the allegedly negligent use of its own police powers.

Jones, 38 A.3d at 346–47; see also District of Columbia v. Evans, 644 A.2d 1008, 1017 n.8 (D.C. 1994) (noting that “the harm . . . was caused directly by the officers at the scene. There is no allegation of failure to protect. The public duty doctrine, therefore, has no relevance to this case”); Lewis v. City of St. Petersburg, 260 F.3d 1260, 1263 (11th Cir. 2001) (holding that under Florida law, a police officer whose conduct creates a “foreseeable zone of risk” has a duty to act with reasonable care to all individuals within that zone).
On this view, the public duty doctrine applies only when officers fail to protect a plaintiff from harm caused by a third person, but has no application when the officers themselves are the source of the harm. Other courts, however, continue to apply the public duty doctrine even in the face of affirmative police misconduct.221In Brothers v. Monaco, 363 F. Supp. 3d 1138 (D. Mont. 2019), for example, the court held that a negligent investigation, which involved affirmative misleading acts rather than a mere failure to investigate, was covered by the public duty doctrine. The court held that “because investigation is a core law enforcement function, a duty owed to an individual under investigation is apparently subsumed under the broader duty to the general public.” Id. at 1153. See also Peschel v. City of Missoula, 664 F. Supp. 2d 1149 (D. Mont. 2009), where the court held that the public duty doctrine did not shield an officer for using excessive force, but only because the exception that the plaintiff was in custody functioned to defeat the doctrine. For example, in Blair v. District of Columbia, the trial court held that the public duty doctrine applied to shield an off-duty police officer who was convicted of criminal assault for beating a plaintiff and leaving him blinded in one eye.222Blair v. District of Columbia, 190 A.3d 212 (D.C. 2018). Over the plaintiff’s objections that the public duty doctrine only applied when third-party harm was at issue, the trial court held that the public duty doctrine insulated the officer in these circumstances.223Id. at 218–19. Because of the applicability of a statute of limitations issue, the appellate court did not reach a ruling on this issue. Id. at 221–24. Similarly, in Todd v. Baker, officers who tased and injured a young man were also protected by the public duty doctrine.224Todd v. Baker, No. CV 10-127-M-DWM, 2012 WL 1999529, at *12 (D. Mont. June 4, 2012).

In addition to the “affirmative interaction” exception, the special relationship exemption also sometimes alleviates the harshness of the public duty doctrine.225In some jurisdictions, what qualifies as a “special relationship” has been interpreted so broadly that the exception essentially swallows the rule. Dobbs et al., supra note 156, at 724–26. Under this exception, if the plaintiff can show that the duty owed to them was particular and distinct from that owed to the general public, the public duty doctrine will not apply. However, there is little consensus among the states as to what exactly a plaintiff must show to bring themselves within this exemption.226. White v. Beasley, 552 N.W.2d 1, 4 (Mich. 1996) (“A survey of special-relationship tests reveals deep disagreement among the states regarding the exact nature of the ‘special-relationship’ the plaintiff must prove.”). Perhaps the most generally accepted test is that created by the New York Court of Appeals in Cuffy v. City of New York, which has been adopted by multiple states, including Ohio and Georgia.227Id. at 5. Georgia’s version of the Cuffy test is slightly modified. See City of Rome v. Jordan, 426 S.E.2d 861, 863 (Ga. 1993). The Cuffy test requires that the municipality assumed, through its actions or promises, an affirmative duty to act on behalf of the plaintiff; knew inaction could result in harm; made direct contact with the plaintiff; and that the plaintiff justifiably relied on the municipality’s undertaking.228. White, 552 N.W.2d at 5.

In the context of policing, at least one court has noted that the special relationship exception presents a high burden, shielding police officers from liability in many instances.229Id. However, that same court believed the test would offer legal relief “in the particularly egregious case of an officer promising police protection, but negligently carrying out that promise.”230Id. In the court’s view, a high standard was appropriate because “police officers are employed to work in a ‘milieu of criminal activity where every decision is fraught with uncertainty,’ ” and “the unusual and extraordinary nature of police work” made it “unfair to allow ‘a jury of laymen with the benefit of 20/20 hindsight to second-guess the exercise of a policeman’s discretionary professional duty.’ 231Id. (citing Ezell v. Cockrell, 902 S.W.2d 394, 398 (Tenn. 1995); Shore v. Stonington, 444 A.2d 1379 (Conn. 1982)). In sum, the court noted that “[p]olice officers must work in unusual circumstances. They deserve unusual protection.”232White, 552 N.W.2d at 5.

When successfully applied, the affirmative interaction and special relationship exceptions can soften the otherwise harsh application of the public duty doctrine.233Id. But the continuing presence of a default no-duty rule has radiating pernicious effects, even when these exceptions mitigate its application. First, setting a baseline of no legally enforceable duty to protect simply makes it less likely that officers will protect. It creates a culture of impunity where due care is not seen as necessary or required. Second, this atmosphere then impacts all police-citizen encounters—not just failures to protect from third-party harms, but failures to take due care for the wellbeing of citizens in direct interactions as well. The public duty doctrine creates an uphill battle for plaintiffs, sends a troubling expressive message, and sets the stage for an unbalanced, politically unaccountable city-constituent relationship. In other words, even when an exception does apply such that there is a successful outcome for a particular plaintiff, the lingering default no-duty rule continues to perpetuate noxious effects.

B. Justifications and Rationales

Courts typically rely on five main rationales to justify the continued application of the public duty doctrine.234Some cases cite additional reasons. For example, in Tipton v. Town of Tabor, 567 N.W.2d 351, 358 (S.D. 1997), the court noted that it was upholding the public duty doctrine defense in part to “promote[] accountability for offenders, rather than police who through mistake fail to thwart offenses.” To hold otherwise, according to the court, would mean that “lawbreaker culpability becomes increasingly irrelevant with liability focused not on the true malefactors, but on local governments.” Id. First, courts point to a floodgates argument, citing the concern of vastly increased litigation if a duty were imposed. Second, courts argue that such liability could devastate municipal finances. Third, courts argue that imposing a duty could lead cities to reduce the services they offer. Fourth, courts argue that the public duty doctrine offers the correct vision of deference from the judicial branch to the legislative. Finally, courts argue that this form of immunity is necessary for policing to function.

1. Floodgates

Public duty doctrine defenders often allege that removing the no-duty default would open the floodgates of litigation.235E.g., Gleason v. Peters, 568 N.W.2d. 482, 484 (S.D. 1997) (where the court “decline[d] the opportunity to open the floodgates of litigation”). On this view, without the public duty doctrine, cities would be overwhelmed by a deluge of claims that would distract them from their ability to administer municipal activities.236Harlow v Fitzgerald, 457 U.S. 800, 806 (1982). However, local governments have been able to manage similar expansions of liability in the past. For example, when the Supreme Court held in 1978 that local governments were subject to Section 1983 liability in some circumstances, there was no evidence this expansion “unduly disrupted the actions of government.”237Chemerinsky, supra note 155, at 1218.

Two additional counterpoints, one philosophical and one practical, further defang the floodgates argument. First, it is a famed adage of American jurisprudence that “[t]he very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury.”238Marin K. Levy, Judging the Flood of Litigation, 80 U. Chi. L. Rev. 1007, 1010 (2013) (alteration in original) (citing Marbury v. Madison, 5 U.S. (1 Cranch) 137, 163 (1803)). As Marin Levy explains, “[t]his phrase originated with Chief Justice Marshall.” Id. at 1010 n.10. But the floodgates argument “implies that one’s ‘protection of the laws’ depends, at least in part, on a speculation about how many others intend to claim that same protection.”239Id. Rendering such protection unavailable on the basis that multiple parties may seek it is arguably morally repugnant and requires significant justification, yet none has been offered.240Id.

Further, the floodgates argument ignores the multiple additional barriers to bringing forward municipal litigation. Municipal litigation comes with a host of procedural and substantive barriers (including the immunity discussed previously), and many of these obstacles will continue to suppress the quantity of municipal litigation regardless of changes to the public duty doctrine. Moreover, for cases that do not involve police, some jurisdictions have already relaxed the public duty doctrine or drastically expanded their use of exceptions,241See infra Section II.B.5. and there is no evidence that a discernible tsunami of litigation has resulted.

2. The Municipal Fisc

Another argument advanced in favor of retaining the public duty doctrine is the protection of the municipal fisc. Particularly in regard to police liability, “[j]udicial fear of the financial impact of expanded police liability” motivates much of its application of the public duty doctrine.242Note, Police Liability for Negligent Failure to Prevent Crime, 94 Harv. L. Rev. 821, 833 (1981).

Recognizing an enforceable duty of care and rejecting the public duty doctrine defense may initially result in some large damage awards.243Id. However, a number of factors make these damages manageable. First, municipal insurance is often both available and financially feasible for local governments.244Id. This would allow municipalities to avoid the fiscal distress that can follow large exogenous shocks like significant damage awards.245Indeed, some courts also hold that the presence of liability insurance voids the doctrine. See supra note 193.

Second, if tort law is functioning properly, those initial awards will give rise to operational and policy changes that will reduce the possibility that similar wrongful acts will take place in the future. In other words, if these suits have the deterrent effect associated with tortious liability such that police departments are more efficient, these awards could ultimately result in retaining, rather than draining, the municipal fisc.246Note, supra note 242, at 833. Ideally, “[a] government suffering large damage judgments should modify institutional practices to reduce the incidence of actionable police negligence,”247Id. and there is evidence that such deterrence effects bear out in practice.248See Schwartz, The Case Against Qualified Immunity, supra note 35, at 1811–12. Some departments evidently do employ better practices than others. For example, studying the practices of the Kansas City Police Department revealed that officers spent “approximately twenty-five percent of their preventive patrol time running personal errands and engaging in other unofficial activities,” while others spent far less. Note, supra note 242, at 833. For example, the Department of Justice has required police departments suspected of civil rights violations to make structural reforms that have included changes in training practices, processing of citizen complaints, and department oversight.249 Reinert et al., supra note 30, at 797–98. The local governments that implemented such reforms subsequently faced fewer liability payments.250 Id. After the Los Angeles Police Department implemented the changes required by the Department of Justice, for example, payouts dropped from about $17 million in 2001 to only about $627,000 in 2009.251Id. The deterrence effect of opening municipalities up to liability can therefore, in the long run, save, rather than increase, municipal expenses.

Moreover, retaining the public duty doctrine is not costless. Even with it intact, municipalities pay out significant amounts for police misconduct through settlements. The reality is that many cities make massive settlement payments, often repeatedly, as a result of their misconduct.252Nick Wing, We Pay a Shocking Amount for Police Misconduct, and Cops Want Us Just to Accept It. We Shouldn’t., HuffPost (May 29, 2015, 7:39 AM), []. For instance, looking at some of the largest metropolises in the United States, Chicago paid $521 million for police misconduct settlements between 2004 and 2014,253Andy Shaw, Key to Cutting Costly Brutality Beefs? Change Chicago Police Culture, Better Gov’t Ass’n Pol’y (Apr. 14, 2014), []. Los Angeles paid more than $100 million in a similar time frame, and New York City paid $348 million between 2006 and 2011.254Wing, supra note 252; Joanna C. Schwartz, Police Indemnification, 89 N.Y.U. L. Rev. 885, 962 (2014). For mid-sized cities, Philadelphia paid $40 million between 2010 and 2014, and Boston paid $36 million over a ten-year period.255Wing, supra note 252. In 2021, Minneapolis paid $27 million for the murder of George Floyd alone.256Nicholas Bogel-Burroughs & John Eligon, George Floyd’s Family Settles Suit Against Minneapolis for Million, N.Y. Times (Mar. 12, 2021), []. These ad hoc settlements sometimes include agreements to change particular policies or trainings, but they are unlikely to be as effective as jury findings in having a more systemic impact on departmental practices.257See Wing, supra note 252.

Second, in both monetary and human terms, “brokenness carries a higher cost than repair.”258A.R. Moxon, Reframe (July 26, 2022) (on file with author). Continuing to apply the public duty doctrine continues the status quo and allows the pace of injury to persist unabated. Discarding the public duty doctrine and imposing a duty on local government actors to protect their constituents could ultimately result in less liability and, more importantly, less injury to constituents. It would do so by incentivizing different police practices that are crafted in light of this duty and by changing the default position to one in which police actors are more likely to adhere to a duty of care in all circumstances.259See Chemerinsky, supra note 155, at 1214 (“Damages are often essential to ensuring accountability. The prospect and actuality of damages can be crucial in creating the incentive for the government to comply with the law.”).

Third, removing a no-duty rule like the public duty doctrine is not equivalent to unfettered liability. As one scholar noted in the context of police liability, it is important to consider the “self-limiting nature of a negligence regime” given that plaintiffs will still have to establish elements such as breach of duty and causation.260Note, supra note 242, at 834. Eliminating the public duty doctrine does not impose strict liability for all injuries related to municipal or police activity or nonactivity.261Estate of Graves v. Circleville, 922 N.E.2d 201, 207–08 (Ohio 2010) (“The absence of the public-duty rule will not automatically result in the creation of new duties and new causes of action . . . . Claimants who seek recovery . . . must still establish [] the [required tortious elements].” (citation omitted)). Rather, it simply places the analysis of whether there should be liability for such activities back in the hands of the jury, the decisionmaker frequently tasked with determining whether liability in any given circumstance is warranted.262Patrick J. Kelley & Laurel A. Wendt, What Judges Tell Juries About Negligence: A Review of Pattern Jury Instructions, 77 Chi.-Kent L. Rev. 587, 587–88 (2002).

This point is particularly important when considering the consequences of removing the no-duty rule on communities, both big and small. Most municipalities in the United States are small, and “[a]bout half of all local police departments employ fewer than ten sworn officers” (though “agencies with more than 100 officers employ nearly two-thirds of all local officers”).263 Harmon, supra note 42, at 9. Smaller communities usually have differing capacities than large ones—though capacities can differ for all sorts of reasons. For example, a large police department might “police a struggling Rust Belt city with an eroding industrial base, entrenched poverty, budgetary cutbacks, and drug and violent crime,” and the resources of that department might differ significantly from a department that polices “a booming suburb with a large and progressive professional class, good schools, and medical care.”264Id. at 11. Indeed, “[c]ommunities vary by population density; housing stock; wealth; education levels; age distributions; racial and ethnic diversity; risk of terrorism and other critical incidents; and strength of social, educational, and medical services,” all of which affect policing practices.265Id. at 11–12. Juries will no doubt consider such factors when deciding whether the policing offered met the standard of care under the circumstances.266See Kelley & Wendt, supra note 262, at 595.

In other words, in rejecting the public duty doctrine, the city would be held to a standard of care requiring it to “act reasonably to promote the security of individuals,” including those affirmatively seeking police aid and those avoiding such interaction. 267David L. Attanasio, The State Obligation to Protect 93 (2015) (Ph.D. dissertation, University of California, Los Angeles) (eScholarship). But “what is reasonable depends on what a normal [municipality] would do to promote security.”268Id. There may be instances where a city is “completely overwhelmed by security problems such that it is simply unable to establish a decent level of security or . . . available resources are insufficient to implement all of the systematic and operational measures of protection that individuals would normally want.”269Id. Because the standard is not perfection but rather what is reasonable under the circumstances, cities need not fear excessive liability in these instances.270See id. It would be reasonable, for example, “for the [city] to do what is possible in the near term to ensure individual security and to attempt to change the circumstances so that [it] might better protect in the future.”271Id.

Finally, if juries are finding liability to the extent that it is becoming detrimental to city functioning, damage caps can keep awards within a payable range.272Note, supra note 242, at 836. While many of these are currently set at absurdly low amounts, it is possible to imagine the existence of reasonable damage caps that appropriately balance the need to compensate individuals harmed by government misconduct and the need to provide services to the community more broadly.

3. Reduced Services

Another proffered rationale for upholding the public duty doctrine is that if cities are not insulated from liability in this way, they may respond by offering reduced police services.273In Maldovan v. County of Erie, 205 N.E.3d 393 (N.Y. 2022), a twenty-three-year-old woman with an intellectual age of eight was beaten and tortured to death by her mother and brother. Neighbors and concerned associates had made numerous calls to child protective and adult protective services, which were negligently investigated. Maldovan, 205 N.E.3d at 395. The court held that the New York version of the public duty doctrine applied to bar the plaintiff’s claim, noting that “ ‘[t]he rationale for this rule is that the cost to municipalities of allowing recovery would be excessive [and] the threat of liability might deter or paralyze useful activity,’ endangering the ability of government agencies to provide crucial services to the public.” Id. (quoting Laratro v. City of New York, 861 N.E.2d 95, 96 (N.Y. 2006)). For example, the Illinois Supreme Court suggested this when it stated that if municipalities could be held liable for failure to enforce ordinances, “the tremendous exposure to liability would certainly dissuade the city from enacting ordinances designed for the protection and welfare of the general public, and thereby the general public would lose the benefit of salutary legislative enactments.”274White v. Beasley, 552 N.W.2d 1, 4 (Mich. 1996) (citing Stigler v. Chicago, 268 N.E.2d 26, 29 (Ill. 1971)).

Statements like this ignore the realities of the city-constituent relationship. A city that chose to make no legal protections for its citizens would likely soon find itself either empty of residents or under new leadership.275See Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J. Pol. Econ. 416, 418 (1956). Moreover, although the precise line has not been determined, there is likely some level of police services that municipalities are legally required to meet.276Michelle Wilde Anderson, The New Minimal Cities, 123 Yale L.J. 1118, 1195 (2014). As Michelle Wilde Anderson describes, in the context of cities that are in extreme fiscal distress, bankruptcy courts often proceed on the basis that there is some minimal level of services that cities are required to provide.277Id. When situations have arisen where local governments have declared that they no longer have the resources to offer protection, as when one local leader advised residents to “lock your doors and load your guns” because financial realities had rendered their police department unable to offer adequate protective services, other local governments like counties have been required to step in and fill that gap.278Id. at 1120. As Anderson writes, given the lack of affirmative duties in the context of tort and constitutional law, it is surprising that in the context of bankruptcy law, courts “assume that city residents are entitled to some degree of basic services provided by the government.”279Id. at 1195. Cases have been adjudicated on the basis that residents as a class “are entitled to have a 911 emergency system that dispatches police officers and firefighters, along with solid waste pick up, wastewater treatment, and other basics.”280Id. In these cases, “[w]ithout resting on either extra-legal natural rights or affirmative, positive rights articulated in constitutional text, Americans do seem to have legally defensible, affirmative rights to basic local services,” at least when a city enters insolvency, and perhaps more broadly.281Id.

4. Judicial Deference

In addition to arguments that floodgates, municipal treasury problems, and reduced services may result from erasing the public duty doctrine, courts also point to principles of judicial deference. For example, in White v. Beasley, the Supreme Court of Michigan upheld the public duty doctrine on the grounds that it “serves a useful purpose by protecting governments from unreasonable interference with policy decisions.”282White v. Beasley, 552 N.W.2d 1, 3 (Mich. 1996). Courts frequently state that the question of whether municipalities should receive immunity is a state policy decision with which the judiciary should not interfere. But “judicial deference to the legislature” is not neutrality. It is, instead, “ultimately a tacit endorsement of the soundness of its existing policies and practices: ‘[T]he decision not to consider whether a duty has been breached is a decision to defer to, and ratify, the political choices government makes.’ 283Gugel, supra note 2, at 1336 (quoting Susan Bandes, The Negative Constitution: A Critique, 88 Mich. L. Rev. 2271, 2329 (1990)). When courts “[p]reclud[e] a . . . failure-to-protect claim on the grounds that it poses no cause of action, or find[] no relief because of the blurry line between action and inaction,” they are functionally legitimating those government actions.284Id.

Moreover, relying on the public duty doctrine as a form of deference to state legislatures is deeply ironic, given that the public duty doctrine is itself a judicial creation, and an equally logical position on this reasoning would actually be to eliminate the doctrine.285Smith, supra note 40, at 424 n.64. In fact, some courts have refused to apply the public duty doctrine on the basis that this judicially created doctrine conflicts with the state’s statutory immunity legislation.286See, e.g., Coleman v. E. Joliet Fire Prot. Dist., 46 N.E.3d 741, 757 (Ill. 2016) (noting that “the legislature’s enactment of statutory immunities has rendered the public duty rule obsolete”).

5. Immunity Is Necessary for Policing to Function

A final reason courts offer for upholding the public duty doctrine is that it allows police officers to make the quick, difficult decisions that the job demands.287See supra notes 231–232. This same argument is also used to support qualified immunity in the Section 1983 context.288See, e.g., Ashcroft v. al-Kidd, 563 U.S. 731, 743 (2011); see also supra Part I. But, just as the Constitution already “gives officers the protection they need to make reasonable mistakes on the job” when it bans only “unreasonable search and seizures,” tort law, too, uses reasonableness as the governing standard for liability in negligence.289See Madeleine Carlisle, The Debate Over Qualified Immunity Is at the Heart of Police Reform. Here’s What to Know, Time (June 3, 2021, 6:35 PM), [] (emphasis added) (referencing interview with Professor Joanna Schwartz). The rate at which police are failing to protect and actively harming constituents suggests that the current liability framework is calibrated too heavily toward impunity.290See Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Allocation of Constitutional Costs, 67 U. Chi. L. Rev. (2000). Reorienting it towards a standard of reasonableness may move officers away from the current regime under which, in the memorable words of Justice Sotomayor, they “can shoot first and think later.”291Kisela v. Hughes, 138 S. Ct. 1148, 1162 (2018) (Sotomayor, J., dissenting).

* * *

When interrogated, none of these proffered rationales, individually or combined, provide a persuasive basis for retaining the public duty doctrine. Indeed, a handful of states have never been persuaded by these rationales and have eschewed the doctrine; currently, at least a dozen state supreme courts either do not apply the doctrine or have significantly cabined its application.292See Cope v. Utah Valley State Coll., 342 P.3d 243, 249 (Utah 2014) (noting that as of 2014, “twelve state supreme courts have explicitly rejected or abandoned [the public duty doctrine]”). Other decisions rejecting or abandoning the doctrine include Fick v. Morken, 685 N.W.2d 98 (N.D. 2004); Natrona Cnty. v. Blake, 81 P.3d 948 (Wyo. 2003); Jean W. v. Commonwealth, 610 N.E.2d 305, 308 (Mass. 1993); Schear v. Bd. of Cnty. Comm’rs, 687 P.2d 728, 730–31 (N.M. 1984); Wilson v. Nepstad, 282 N.W.2d 664, 671 (Iowa 1979); Brennen v. City of Eugene, 591 P.2d 719, 724–25 (Or. 1979); and Adams v. State, 555 P.2d 235, 241–42 (Alaska 1976). See Briffault et al., supra note 159, at 1046. State courts that have criticized the doctrine often do so on the basis that it conflicts with the state’s “abrogation of absolute sovereign immunity,” because the public duty doctrine is really sovereign immunity by another name.293Rollins v. Petersen, 813 P.2d 1156, 1162 n.3 (Utah 1991); see also Beaudrie v. Henderson, 631 N.W.2d 308, 317 (Mich. 2001). Other courts have expressed concern that the doctrine “creates confusion in the law and produces uneven and inequitable results in practice.”294 McQuillin, supra note 189, § 53:21. Some courts have opted to formally retain the public duty rule but soften it through exceptions like not applying the doctrine if the actions were “particularly ‘egregious.’ 295Id. Notably, though, even those jurisdictions that “have drastically restricted the public duty defense” under one of these means tend to “leave it intact in the case of a claim that the police generally failed to protect citizens from foreseeable crime.”296Weisberg, supra note 190, at 379 (citing Hamilton v. Cannon, 482 S.E.2d 370 (Ga. 1997) (“[I]mmunity removed for such public functions as operating swimming pools, but immunity remains for police.”)); see also Thompson v. Waters, 326 S.E.2d 650, 651 (N.C. 2000) (holding that the public duty doctrine protects law enforcement agencies but does not extend beyond that).

C. Towards a Duty of Care

A major problem with the public duty doctrine is that it creates a faulty foundation for a thriving, balanced city-constituent relationship. Tort law possesses an “inherent political power” for structuring the terms of our relationship to each other and to the state, and the presence or absence of a duty of care shapes our understanding of what obligations flow from these relationships.297Sarah L. Swan, Tort Law and Feminism, in The Oxford Handbook of Feminism and Law in the United States (Deborah Brake, Martha Chamallas & Verna Williams eds., 2021). Eliminating the public duty doctrine would reorient the city-constituent relationship and further the goals of compensation and deterrence; align with equal protection principles as expressed at the state, federal, and international levels; enhance democratic accountability in the city; and accord with the thick city-constituent relationship municipalities themselves articulate in the affirmative litigation context.298See infra Section II.C.4.

1. Compensation and Deterrence

The public duty doctrine requires harmed individuals to “bear the entire loss” resulting from the government’s wrongful acts or failures to act, rather than spreading the cost of that injury over the collectivity of the municipality.299Chemerinsky, supra note 155, at 1217. Implicitly underlying the doctrine is the idea that safeguarding the government treasury is more important than compensating harmed individuals for their injuries. The Supreme Court, though, has never explicitly rationalized or “justified this value choice.”300Id. In essence, the public duty doctrine “requires victims to subsidize local governments and the public by bearing the financial losses inflicted by police-involved violence” and other governmental wrongs. However, a compelling stated reason for why individuals should bear such a loss and subsidize the community in this way remains elusive.301Reinert et al., supra note 30, at 784.

Eliminating the public duty doctrine would allow victims to potentially receive compensation for the injuries they suffer at the hands of a city’s wrongful actions and failures to act. Given that many police misconduct harms fall on marginalized communities, cities that claim the mantle of social justice seekers “should readily acknowledge the unfairness of requiring vulnerable communities to bear these losses” and acknowledge “the wisdom of assigning the losses to the local government agencies responsible.”302Id. Corrective justice principles dictate that wrongdoers should be accountable for the losses they inflict, and it is a core purpose of tort law to provide compensation to those who are wrongfully injured.303Id. An injustice exists when a community “allow[s] victims to go uncompensated while culpable police defendants are not held accountable for their actions. As between the two parties, the one who has acted unreasonably ought to pay.”304Note, supra note 242, at 834 (footnote omitted). As the Supreme Court once explained when considering potential municipal liability under Section 1983, “the public should pay the costs” of government misconduct because it receives the benefits that flow from having that government in place; leaving a harmed individual uncompensated and required to solely bear the costs of governmental misconduct, when others share equally in the benefits of government, seems fundamentally unjust.305Id. (citing Owen v. City of Indep., 445 U.S. 622, 655 (1980)).

This compensation function connects to tort’s deterrence function. Cities, and the police in particular, are in a better position to bear the costs of governmental wrongdoing, not just because cities “generally possess superior loss-bearing capacity,” but also because they are able to alter their behavior to create fewer losses and harms.306Id. As noted above, a core premise of tort law is that liability can create deterrence and make it less likely for actors to engage in similar behavior.307Id. Removing the public duty doctrine would not only provide a path to compensation for those already harmed, but would also prevent future harms from occurring in the first place, thereby resulting in less harm overall.

2. Equal Protection Principles

In addition to fulfilling this core compensation and deterrence purpose of tort law, eliminating the public duty doctrine and imposing a duty of care would advance the equal protection values expressed in state, federal, and international instruments. At different historical moments, state, federal, and international governing bodies have all embraced a robust duty to protect their citizens.

i. State Equal Protection Principles

At the state level, there was a historical era when at least some states believed that equal protection principles required them to affirmatively protect their citizens. In addition to the communal liability discussed earlier,308See supra Section II.A. states like Massachusetts, at the time when the states stood on the brink of the Civil War, took their duty to protect seriously and understood the Citizenship Clauses in their state constitutions to require an affirmative duty of protection.309Maeve Glass, Citizens of the State, 85 U. Chi. L. Rev. 865, 893 (2018). At that time, “state citizenship . . . had come to represent a status that carried with it a duty of protection on the part of the state—a duty said to define the very essence of what it meant to be a government of the people.”310Id. at 918–19. State sovereignty thus was understood to be “measured not simply by the degree of autonomy from the national government, but by the duties owed to its peoples.”311Id. at 929.

The idea that state sovereignty was intimately connected to the duties owed was heavily influenced by the political idea of the social contract. Indeed, state constitutions often clearly invoke the Lockean notion that a social contract formed the basis of government.312Gugel, supra note 2, at 1342. In accordance with this pact, “[i]n exchange for consent to be governed,” the state assumed obligations to “provid[e] civil liberties both in positive and negative terms . . . states accepted responsibility for third-party harms and the need to affirmatively act to ensure each citizen’s welfare.”313Id. at 1342–43. In other words, the sovereign-state relationship was understood as defined by “mutual bond and obligation,” including “the sovereign’s obligation to protect against private [and arguably public] violence.”314Id. This obligation was enforceable in courts through another state constitutional provision: “the right of court access for redress against injury.”315Id.

Indeed, through their constitutions, some states expressly acknowledged that protection was “the right of every individual,” not an amorphous duty owed to the general public.316Heyman, supra note 5, at 523. The Massachusetts Constitution declared that “[e]very individual of the society has a right to be protected by it in the enjoyment of his life, liberty, and property, according to standing laws.”317 Mass. Const., art. X, pt. I. The Massachusetts Constitution went on to explain that in return, the citizen was

obliged, consequently, to contribute his share to expense of this protection; to give his personal service, or an equivalent, when necessary; but no part of the property of any individual can, with justice, be taken from him, or applied to public uses, without his own consent, or that of the representative body of the people.”

The Pennsylvania Constitution of 1776 asserted that “every member of society . . . hath a right to be protected in the enjoyment of life, liberty and property,” and this was echoed in the Delaware, Massachusetts, and New Hampshire state constitutions as well.318Heyman, supra note 5, at 512. The content of this right of protection was three-pronged.319Id. at 530. First, the protection “related to the status of the individual: To be under the protection of the law meant to have the status of a freeman and a citizen.”320Id. at 530–31. Second, the protection had a substantive component: “[p]rotection meant that the law recognized and secured an individual’s rights to life, liberty and property.”321Id. at 531. The third prong was remedial and “referred to the enforcement of rights: the specific ways in which government prevented violations of substantive rights, or redressed and punished such violations.”322Id. These three prongs are implicated to an even greater extent in situations where the government itself is the violator of those substantive rights: the offense to the “protection of laws” is even greater. Where there is no redress for that harm, as the public duty doctrine ensures, insult is added to injury, and the idea of equal protection is further thwarted. If the state can harm or fail to protect its citizens with legal impunity, one’s status as “a person entitled to protection as a member of the political community” is continually in jeopardy.323Id.

ii. Federal Equal Protection Values

At the federal level, a growing number of scholars argue that, despite the rulings in Deshaney and Gonzales, the original meaning of the Fourteenth Amendment did include an affirmative state obligation to protect.324 Barnett & Bernick, supra note 4, at 94. Parsing that history, one scholar notes specifically that “[t]he 1866 debates make clear that the Framers understood the Civil Rights Act and the Fourteenth Amendment to incorporate a fundamental right to protection by the government, with a corresponding obligation on the states to afford such protection.”325Heyman, supra note 5, at 554. Over time, this right to protection was obscured as the clause came to be interpreted more through a discrimination frame than one involving affirmative protection.326Christopher R. Green, The Original Sense of the (Equal) Protection Clause: Subsequent Interpretation and Application, 19 Geo. Mason U. C.R.L.J. 219, 222 (2009). But reviving this lost history and reincorporating a duty-to-protect reading of the Equal Protection Clause would impose on governments “a positive duty to act properly, not merely a negative duty to refrain from misbehavior.”327Id. Such a “duty-to-protect reading of the Equal Protection Clause would have great significance today,” and serve as “both an equality provision and a fundamental-right provision.”328Id. Serving “[a]s an equality provision, it would require that the state perform its enforcement and remedial functions for the equal benefit of all persons within the jurisdiction.”329Id. And “[a]s a fundamental-right provision, it would require that the state protect all individuals from violence in a minimally adequate way.”330Id. Returning to this original meaning would thus have radical implications for governmental affirmative duties in general.331For example, it would legitimate federal legislation offering federal court remedies to those who experience a failure to protect. In United States v. Morrison, 529 U.S. 598, 619 (2000), the Supreme Court held that Congress may not use its Section 5 power under the Fourteenth Amendment to establish a federal cause of action under the Violence Against Women Act for failures to protect. Barnett & Bernick, supra note 4, at 36–37.

Eliminating the public duty doctrine could have a similar effect. Imposing a basic, enforceable duty of care on municipal police can similarly “require that the state protect all individuals from violence in a minimally adequate way” and “require that the state perform its enforcement and remedial functions for the equal benefit of all persons within the jurisdiction.”332Green, supra note 326, at 222. Each police-public interaction would be imbued with this duty, and the duty would prohibit practices “that unreasonably leave the personal security of some people more uncertain than that of others.”333 Barnett & Bernick, supra note 4, at 35. Importantly, this duty would impact both underpolicing and overpolicing: not only persons seeking police aid and calling 911, but also people about whom 911 is called, would receive the benefit of that duty of care.334See, e.g., Haven Orecchio-Egresitz, A Police Officer Said ‘I Don’t Have Time for This’ Before Shooting an 18-Year-Old with a Mental Illness Dead in His Home, His Mother Says, Bus. Insider (Jan. 27, 2021, 12:06 PM), [].

iii. Evolving International Norms

At the international level, human rights principles also recognize a governmental duty to protect. Under international human rights principles, states have “an affirmative obligation to protect individuals from private acts of violence, to investigate alleged violations and publicly report the results, and to provide an adequate and effective remedy when these duties are breached.”335Gugel, supra note 2, at 1344 (quoting Caroline Bettinger-Lopez, Human Rights at Home: Domestic Violence as a Human Rights Violation, 40 Colum. Hum. Rts. L. Rev. 19, 21 (2008)). In other words, the protection from private violence is understood as a basic human right, which the state is obliged to protect and secure.336See Dimitris Xenos, The Protection Against Crime as a Human Right: Positive Obligations of the Police, in The Police and International Human Rights Law 181, 190 (Ralf Alleweldt & Guido Fickenscher eds., 2018).

Protection from private violence gained the status of a human right following World War II.337Jacob Weinrib, Sovereignty as a Right and as a Duty: Kant’s Theory of the State, in Sovereignty and the New Executive Authority 21, 21 (Claire Finkelstein & Michael Skerker eds., 2019). Prior to that era, most public law at both national and international levels tended to view sovereignty in terms of which unassailable rights accompanied that status.338Id. at 21–22. In the aftermath of the war, however, a view of sovereignty rooted in the idea that the underlying basis of the “right to rule” was the “fulfillment of the human rights of the ruled” became ascendant.339Id. at 21. Both the Universal Declaration of Human Rights and many postwar constitutions came to reflect this conception,340Id. and it is “[n]ow entrenched as a ‘fundamental precept of international human rights law’ ” that states have an obligation to protect their citizens from private as well as public violence.341Gugel, supra note 2, at 1344 (quoting Caroline Bettinger-Lopez, Human Rights at Home: Domestic Violence as a Human Rights Violation, 40 Colum. Hum. Rts. L. Rev. 19, 46 (2008)). Accordingly, the prevailing international view holds states accountable for private individual acts when they fail to exercise due diligence in preventing or responding to the violation. Under this standard, the state can be responsible for a human rights violation if it “knew or ought to have known of a situation presenting a real and immediate risk to the safety of an identified individual from the criminal acts of a third party,” yet “failed to take reasonable steps within the scope of its powers, which might have had a reasonable possibility of preventing or avoiding that risk.”342Id. at 1345. States have been held liable under this standard by domestic and transnational courts for failures to protect constituents from third-party harms.

Indeed, in 2011, the Inter-American Commission on Human Rights found that the United States fell below this standard in the circumstances of the Gonzales case. The Commission found that the United States’ failure to protect the plaintiff in the first instance, and its failure to provide a remedy when that initial duty was breached in the second, “violated the ‘obligation not to discriminate and to provide for equal protection before the law under Article II of the American Declaration.’ 343Sarah Rogerson, Domesticating Due Diligence: Municipal Tort Litigation’s Potential to Address Failed Enforcement of Orders of Protection, 21 Am. U. J. Gender, Soc. Pol’y & L. 289, 292 (2012).

Further, after investigating complaints of police brutality in the United States, that same Commission concluded that “the United States has systematically failed to adopt preventive measures and to train its police forces to perform their duties in an appropriate fashion. This has led to the frequent use of force based on racial bias and prejudice and tends to result in unjustified killings of African Americans.”344 Inter-Am. Comm’n on Hum. Rts., Police Violence Against Afro-Descendants in the United States 2 (2018), []. The Commission also emphatically denounced the institutional problems with policing in the United States following the murder of George Floyd and police misconduct during the subsequent Black Lives Matter protests, and again urged that appropriate forms of legal redress were crucial to remedying the discriminatory application of force.345See Press Release, Inter-Am. Comm’n on Hum. Rts: The IACHR Expresses Strong Condemnation for George Floyd’s Murder, Repudiates Structural Racism, Systemic Violence Against Afro-Americans, Impunity and the Disproportionate Use of Police Force, and Urges Measures to Guarantee Equality and Non-Discrimination in the United States (June 8, 2020), []. The press release notes that the IACHR “expresses its strongest condemnation for the cruel murder of George Floyd at the hands of the police, the systemic violence against Afro-Americans” and “urges the State to . . . adopt urgent measures to guarantee equality; as well as to put an end to police violence and impunity as a symptomatic element of the persistence of this discrimination based on ethnic-racial origin.” Id.

As local governments have been trying to increase their political and legal status, they have often turned to the international stage as a forum for accomplishing this. For example, local governments have frequently been banding together in networks now referred to as the “glocal” in order to enhance their political power and further goals like human rights and environmental protections.346See, e.g., Michael Riegner, International Institutions and the City: Towards a Comparative Law of Glocal Governance, in The Globalisation of Urban Governance (Helmut Philipp Aust & Anél du Plessis eds., 2018). In many instances, local governments have voluntarily adopted and held themselves to the standards set in various human rights declarations and rulings, even in circumstances where those same instruments have not been ratified at the federal level.347See Rogerson, supra note 343, at 303–04. Embracing a duty of care would be consistent with this approach, and would incorporate the international norm of a duty to protect and align localities with the recommendations of the Commission.

3. Democratic Accountability

Embracing an owed duty to protect would not only compensate harmed individuals and align with general equal protection principles; it would also help to ensure the democratic accountability necessary for a healthy community.348See generally John C.P. Goldberg, The Constitutional Status of Tort Law: Due Process and the Right to a Law for the Redress of Wrongs, 115 Yale L.J. 524 (2005). Awarding damages is not only about compensation to the particular person harmed; “[f]rom a corrective justice perspective, the award of damages may also restore community faith in the system of law, providing redress and righting a wrong.”349Reinert et al., supra note 30, at 796–97. When a wrongful injury occurs, “tort liability helps mend the community by allowing the individual victim of a wrong to pursue redress.”350Id. at 797. To promote this community faith and “persuade people of the system’s legitimacy,” “governments owe a basic duty to provide victims with access to tribunals in which they can pursue vindication for the wrongs they have suffered” and in which they can be assured of “[f]air process and a right to be heard.”351Id.

When the government is the tortfeasor, this need is even stronger. A functioning American government “is based on the fundamental recognition that the government and government officials can do wrong and must be held accountable.”352Chemerinsky, supra note 155, at 1202. Indeed, as Abraham Lincoln once noted, “[i]t is as much the duty of Government to render prompt justice against itself in favor of citizens as it is to administer the same between private individuals.”353Id. at 1224 (emphasis added) (quoting Abraham Lincoln). The public duty doctrine, much like sovereign immunity, “undermines that basic notion.”354Id. at 1202.

i. The Right to Petition

Deeply tied to the notion of democratic accountability is the constitutional right to petition. Under the Petition Clause of the First Amendment, the people have a right “to petition the Government for a redress of grievances.”355 U.S. Const. amend. I. This right is part of the core political rights protected by the federal Constitution,356Many state constitutions also protect this fundamental right. Steven G. Calabresi, Sarah E. Agudo & Katherine L. Dore, The U.S. and the State Constitutions: An Unnoticed Dialogue, 9 N.Y.U. J.L. & Liberty 685, 711 (2015). and although a narrow, conventional view of the Petition Clause emphasizes its relation to petitioning the legislative branch of government, powerful textual and historical arguments suggest that the Petition Clause equally encompasses a “guaranteed right to pursue judicial remedies for unlawful government conduct.”357James E. Pfander, Sovereign Immunity and the Right to Petition: Toward a First Amendment Right to Pursue Judicial Claims Against the Government, 91 Nw. U. L. Rev. 899, 899 (1997) (emphasis added). Indeed, the drafting of the constitutional Petition Clause itself shows that it initially referred to petitions to the Legislature, but a subsequent draft changed that language to petitions to the “Government,” encompassing all three of its branches. This is in keeping with the British right to petition, which served as the template for the American version and was rooted in the notion that British subjects had a legal right to turn to the judiciary to adjudicate certain claims against the King for governmental misconduct. Such “petitioning” provided a legal path for the Court of Chancery to hear these claims for redress. Id. at 959 n.219.

Understood as guaranteeing that the people have a right to apply to the judiciary for redress for governmental misconduct, the Petition Clause provides constitutional heft to the political right of citizens to hold the government responsible for misconduct in the judicial forum.358Id. The public duty doctrine, along with other iterations of sovereign immunity, conflicts with this right.359See id. at 980. Because it is a no-duty rule, the public duty doctrine bars judicial redress, often on the grounds that such redress is a political, not judicial, matter, despite the constitutionalization of the idea that such claims are in fact required to be heard in court.

ii. The Importance of the Jury

Like the ability to seek redress of grievances generally, an important part of the local democratic project is the power to seek dispute resolution from the community in the form of the jury.360Kelley & Wendt, supra note 262, at 591. Many state constitutions speak to the importance of the civil jury and community standards for adjudicating claims of negligence.361Id. at 587–88. When a negligence claim goes to the jury, the jury “make[s] a qualitative judgment about that action” that applies community standards, and this process is constitutionalized because of its centrality to fair process and community cohesion.362. Id. at 590. A no-duty rule like the public duty doctrine, on the other hand, blocks a claim before it can be considered by a jury and prevents a community from ever deciding the difficult question of potential liability for police failures to protect or affirmative misconduct.363Id. Thus, even though “negligence is considered a fact to be determined by the jury under state constitutional provisions,” the public duty doctrine subverts this constitutional underpinning as well.364Kelley & Wendt, supra note 262, at 587. “Central to liberal democracy is the idea that the exercise of public power should, to a significant and meaningful extent, be under the control of the people—individual citizens, groups of citizens and the citizen body as a whole,” and submitting the potential liability of police misconduct to jury members fits squarely within this purview.365Cane, supra note 43, at 164. Further, because police agencies are agents of the people and their compensation comes out of the public fisc, the people have a heightened claim to assist in the adjudication of these matters through the presence of a jury.

4. The City-Constituent Special Relationship

Many cities have cast themselves as deeply concerned with protecting their citizens and enhancing their civil rights.366Hannah J. Wiseman, Rethinking Municipal Corporate Rights, 61 B.C. L. Rev. 591, 595–96 (2020). The city of Toledo, for instance, recently bought the medical debt owed by its residents, spending .6 million to eliminate approximately 0 million in debt owed by roughly 25,000 residents. The Toledo City Council was inspired in part by a similar program in Chicago. See Steffi Cao, ‘We Have a Broken Healthcare System’: An Ohio County Is Trying to Erase up to 0 Million in Medical Debt, Buzzfeed News (Nov. 12, 2022, 5:07 PM), []. For example, cities have offered their constituents enhanced voting opportunities,367Wiseman, supra note 366, at 634. tried to pass regulations curtailing environmental harms and raising the minimum wage,368Id. at 596, 648. signed international agreements to uphold human rights,369Id. at 638. and incorporated international human rights into local ordinances.370For instance, Cincinnati, Baltimore, and Miami enacted local ordinances modeled on human rights law “affirming that freedom from domestic violence is a basic human right.” Goldscheid, supra note 99, at 78. In 2021, New York City even created a cause of action against the police “for excessive force and unlawful searches and seizures.”371Reinert et al., supra note 30, at 742.

When cities engage in these endeavors, they often do so on the basis that the important relationship between them and their constituents requires them to further these interests.372See Swan, supra note 55, at 1255. This conception of a thick city-constituent relationship is particularly prominent in the context of affirmative litigation. Affirmative litigation refers to a trend, growing in the last few decades, where cities have increasingly been bringing forward tort claims against third parties for harms caused to them and their constituents.373See id. at 1230 n.11. Many major cities, including “San Francisco, Los Angeles, and New York now have specialized units within their law departments devoted to bringing these claims and developing their role as plaintiff cities,” and “[o]ther less well-resourced cities pursue such claims as well, though often on a more ad hoc basis or through partnering with private law firms.”374Id. at 1229–30. Cities have brought forward hundreds of claims, including against “gun manufacturers for the scourge of gun violence,” “against banks for the consequences of the subprime mortgage crisis,” and “against pharmaceutical companies for opioid-related deaths and injuries,” among others.375Id. at 1227.

When they bring these claims on behalf of their constituents and communities, cities assert that the city-resident relationship is thick and capacious enough to support such actions.376Id. at 1251. Cities rebuff criticisms that they lack standing or are inappropriate entities to bring such suits with arguments that they are increasingly important political bodies that are deeply bound to their residents and their wellbeing.377Id. They frame themselves as protecting their constituents’ interests, and in particular as protecting the interests of the city’s most vulnerable groups (the groups most impacted by the harms alleged in affirmative litigation).378Id. at 1246. Cities use these plaintiff city claims to create the city as “a polis, a political collectivity, a place where public interest is defined and realized,” and define that political collectivity to decisively include often marginalized groups.379Mustafa Dikeç, Justice and the Spatial Imagination, in Searching for the Just City 72, 83 (Peter Marcuse et al. eds., 2009) (quoting Étienne Balibar, La Crainte des Masses 66 (1997)).

This “signaling function and community-building impact” of affirmative litigation matters for the public duty context.380Swan, supra note 55, at 1288. In the affirmative litigation context, cities assert that they are in such a thick relationship with constituents that it in many ways comes to resemble a special relationship. In fact, the city-constituent relationship maps specifically onto the “special relationship” exception to the public duty doctrine on multiple fronts. For example, “[t]he factors considered by many courts in deciding whether to impose a duty based on the special relationship rubric” include “whether ‘the plaintiff is . . . in some respect particularly vulnerable and dependent upon the defendant who, correspondingly, holds considerable power over the plaintiff’s welfare.’ 381W. Jonathan Cardi & Michael D. Green, Duty Wars, 81 S. Cal. L. Rev. 671, 713 (2008) (quoting Harper v. Herman, 499 N.W.2d 472, 474 n.2 (Minn. 1993)). In many instances, the would-be plaintiffs asserting either inaction or wrongful action on the part of the city are particularly vulnerable and dependent on the city, and the city “holds considerable power over the plaintiff’s welfare” by virtue of the structural relationship between them.382Id. Cities and residents are in a relationship of what one scholar calls “substantive relationality,” with the kind of “ex ante” relationship between them that “informs virtually every court’s reasoning about certain affirmative duties.”383Id. at 715. This substantive relationality should equally inform how courts understand the relationship between cities and constituents in this context.

Some scholars have gone so far as to argue that the relationship between cities and constituents is so thick that cities stand in a fiduciary relationship to their constituents.384Max Schanzenbach & Nadav Shoked, Reclaiming Fiduciary Law for the City, 70 Stan. L. Rev. 565 (2018). On this view, a city has “a special moral obligation to protect” because it is “the fiduciary of those individuals in its legal territory” and thus owes “a fiduciary obligation to advance their [constituents’ interests] reasonably. Since obtaining security from violence is a normal and fundamental purpose of individuals, the state has a fiduciary obligation to protect them from private violence.”385Attanasio, supra note 267, at ii. Returning to political theory, “Locke claimed that government holds a fiduciary power and so must act in accordance with the trust placed in it, principally to ensure the preservation of society and all of its members.”386Id. at 205. On this view, “[t]he state has a special obligation to protect . . . because it assumes political power over certain persons, meaning that it treats them as subject to its formulation, application, and enforcement of law and rights.”387Id. at 207.

Courts have stated that the public duty doctrine does not impose a specific duty of care to individuals harmed by governmental action or inaction “unless there is some specific connection between the government agency and the individuals that makes it reasonable to impose a duty.”388Cope v. Utah Valley State Coll., 290 P.3d 314, 318 (Utah Ct. App., 2012). But in the modern city, the city constantly alleges that specific connection in other contexts and encourages its constituents to believe in this relationship.

In this way, the city-constituent relationship comes close to the “well-established rule of tort law that ‘one who assumes to act, even though gratuitously, may thereby become subject to the duty of acting carefully.’ 389Note, supra note 242, at 824 (quoting Schuster v. City of New York, 154 N.E.2d 534, 541 (N.Y. 1958) (McNally, J., concurring) (quoting Glanzer v. Shepard, 135 N.E. 275, 276 (N.Y. 1922))). By skirting so close to alleging a special relationship in the affirmative litigation and other contexts, cities may give their constituents the impression that the city is operating under an assumed duty to protect. This creates an “illusion of protection,” and when this illusion bumps up against the reality of no duty to protect, the corresponding clash and alienation may be “worse than no protection at all.”390Miccio, supra note 199, at 408 (emphasis omitted).

III. Public Duties and the New City

Relying on the public duty doctrine to avoid liability for harm caused to constituents by police action and inaction exists in tension with cities’ otherwise often socially progressive character and works against their own claims to political legitimacy as independently valuable polities. To be consistent seekers of social justice and advance their own political standing, cities should reject the public duty doctrine and instead embrace a duty to protect.

A. Embracing a Duty to Protect

The city’s opposing faces depending on which side of the litigation aisle it occupies present a fundamental conundrum.391See Alexander A. Reinert, The Influence of Government Defenders on Affirmative Civil Rights Enforcement, 86 Fordham L. Rev. 2181, 2182 (2018) (identifying this tension in the federal Department of Justice, and noting it likely exists at the state level as well). When bringing affirmative litigation, the city “can be one of the most efficient and powerful vindicators of civil rights.”392Id. at 2182. Yet when operating from a defensive crouch, the city can simultaneously be “one of the most effective advocates for imposing barriers to affirmative civil rights enforcement.”393Id. City attorneys can be in one courtroom “vindicat[ing] rights and remedy[ing] inequities,” but in another vigorously trying “to prevent the same.”394Id. (quoting Vanita Gupta, Head of the C.R. Div., Dep’t of Just., Remarks at the National Legal Aid & Defender Association Annual Conference (Nov. 10, 2016), []).

In numerous contexts, however, cities suggest that they are committed to the aspirational goal of being a “just city.”395See Susan S. Fainstein, The Just City 3 (2010). A just city recognizes some of the social ills that are associated with urbanization—including exploitation, insecurity, and economic inequality that run along lines of race, class, and gender396James Connolly & Justin Steil, Introduction: Finding Justice in the City, in Searching for the Just City 1, 8 (Peter Marcuse et al. eds., 2009).—and turns to three guiding principles—equity, diversity, and democracy—to help cities overcome these social problems397 Fainstein, supra note 395, at 5. and offer a “ ‘richer account of democratic life’ for their citizens.”398Swan, supra note 55, at 1289 (quoting Richard C. Schragger, The Political Economy of City Power, 44 Fordham Urb. L.J. 91, 132 (2017)). The first guiding value, equity, focuses on orienting municipal policy decisions towards achieving distributive justice.399Id. The second, diversity, focuses on “building capacity for all community members.”400Id. And the third, democracy, is concerned with “meaningful public participation and reasoned public deliberation about the best measures to secure the general community interest.”401Id. (quoting David A. Dana, Public Interest and Private Lawyers: Toward a Normative Evaluation of Parens Patriae Litigation by Contingency Fee, 51 DePaul L. Rev. 315, 322 (2001)). Part of democracy involves “a right to inclusion and social responsibility—a right that suggests that one’s [injury] . . . is not solely one’s own concern,”402Wendy E. Parmet, Tobacco, HIV, and the Courtroom: The Role of Affirmative Litigation in the Formation of Public Health Policy, 36 Hous. L. Rev. 1663, 1711–12 (1999). but “the concern of a just polity as well.”403Swan, supra note 55, at 1290.

These three principles are themselves supported by an underlying or overarching “duty of justice.”404See Attanasio, supra note 267, at 45–86. The just city is one which acknowledges its “moral duty to contribute to the realization of justice, including to help ‘ensure that all persons’ have their rights respected and fulfilled.”405Id. at 49 (quoting Allen Buchanan, Political Legitimacy and Democracy, 112 Ethics 689, 703–04 (2002)). As municipal affirmative litigation efforts demonstrate, contemporary American cities are crafting a new political identity, one in which they are “establishing themselves as entities responsible for and capable of achieving justice for their populations.”406Swan, supra note 55, at 1285. Because cities insist that this socially equitable vision is their goal,407See e.g., Wiseman, supra note 366 and accompanying text. they should reconcile these two faces, aligning their defensive approaches to more accurately further the ends of social justice.408See Reinert, supra note 391, at 2196 (noting that this will “require[] a form of institutional design that permits the communication of priorities between affirmative and defensive bureaus, as well as a means to resolve conflicts when they arise”). The choices a city makes in relation to both reducing failures to protect and other harms caused by police, as well as their litigative choices when these harms do happen, matter a great deal for this self-construction of the city as a relevant and increasingly important political category.409See Dannestam, supra note 53, at 367. As cities forge their new identity, their defensive litigative choices must be reconciled with their affirmative ones if they are to create a coherent picture of an engaged polity intent on social justice. Right now, police across the nation “kill more than 1,000 people annually . . . [and] mistreat more than one million people each year . . . . [P]olice disproportionately use their powers against the most marginalized: people who are Black, Latino, and Indigenous; immigrants; protestors; homeless; mentally ill; and people who identify as LGBTQ+.”410Joanna C. Schwartz, An Even Better Way, Calif. L. Rev. (forthcoming) (manuscript at 3), []. Cities nevertheless routinely defend egregious police behavior, a portrait that stands in sharp contrast to the aspirational goal of a just city.

Indeed, the tension between the public duty doctrine and the vision of a just city is profound. In essence, the underenforcement often alleged in public duty doctrine cases is a reflection of city abdication: a municipal “failure to address social harm, to attend to the consequences of lawbreaking, and generally to care for its constituents.”411Natapoff, supra note 9, at 1774. Underenforcement is not neutral: it is a powerful signal concerning “many distributive and democratic issues.”412Id. It articulates “society’s background assumptions about who matters and who is entitled to public support.”413Id.

At its most basic level, the public duty doctrine essentially creates a norm of not taking reasonable care for city residents. A somewhat elusive yet deeply important concept, “[n]orms are the language a society speaks, the embodiment of its values and collective desires, the secure guide in the uncertain lands we all traverse, the common practices that hold human groups together.”414Rogerson, supra note 343, at 298 (quoting Cristina Bicchieri, The Grammar of Society, at ix (2006)). These governing yet implicit values “ ‘do not appear out of thin air; they are actively built by agents having strong notions about appropriate or desirable behavior in their community.’ They are the most powerful yet analytically elusive guideposts of human behavior.”415Id. (footnote omitted) (quoting Martha Finnemore & Kathryn Sikkink, International Norm Dynamics and Political Change, 52 Int’l Org. 887, 896 (1998)).

Norms are heavily influenced by law—laws that say there are no duties to others impact behavior. As one theorist noted, “[a] past wrong . . . , standing in your history without apology, atonement, retribution, punishment, restitution, condemnation, or anything else that might recognize it as a wrong, makes a claim”: that it is acceptable to treat you in that way.416Scott Hershovitz, Treating Wrongs as Wrongs: An Expressive Argument for Tort Law, J. Tort L., Mar.-Oct. 2017, at 1, 8 (emphasis omitted) (quoting Pamela Hieronymi, Articulating an Uncompromising Forgiveness, 62 Phil. & Phenomenological Rsch. 529, 546 (2001)). When norms suggest that particular groups need not be protected and the law reinforces that through the public duty doctrine, discriminatory effects result. And just as a no-duty rule in the form of the public duty doctrine both reflects and creates the norm that failing to protect certain groups is tolerable, it also helps to perpetuate the norm that excessive force in policing is acceptable. The public duty doctrine essentially holds that police need not take care for you. Without the grounding of an owed, enforceable duty of care, the default approach all too quickly slides into the opposite of care: violence and exploitation.417Jia Tolentino, Can Motherhood Be a Mode of Rebellion?, New Yorker (May 8, 2022), [].

Fortunately, “hold[ing] officials accountable can help prevent unjustified actions from becoming ongoing norms” and counteract them where they already exist.418Andrew Parker, Austin Sarat & Martha Merrill Umphrey, Introduction, in Subjects of Responsibility 1, 10 (Andrew Parker ed., 2011). Eliminating the public duty doctrine could create a new norm of care. It would switch the default position from “police need not take reasonable care to protect” to “police owe a basic duty of care to all constituents.” This change in ethos would impact both underpolicing and overpolicing, moving away from disregard and exploitation and towards a governing ethic of care.

Policing—and the norms it does or does not embody—is acutely important to the city-constituent relationship.419See O’Rourke et al., supra note 6. Notably, the Court in Deshaney and Gonzales appears to view policing like any other city service, akin to “trash collection, health care, or housing,” and thus “subject to the more general proposition that the state is not obligated to provide such services in the first place, and therefore the failure to provide is not a constitutional harm.”420Natapoff, supra note 9, at 1768 (footnote omitted). Inherent in this framing “is an implicit rejection of any special, democracy-maintaining quality of policing, or the idea that policing could be a unique state obligation to which citizens might have a right above and beyond their entitlement to conventional government benefits.”421Id. But policing is a distinct kind of government service, and “police duties of protection against crime usually relate to the most fundamental human rights and basic societal guarantees, such as security and public order.”422Xenos, supra note 336, at 4. So the inquiry is “not simply about the duties of the police, but also about the legal system of protection of human [and civil] rights.”423See id. Policing has significant “participatory and citizenship implications. Insofar as policing represents the state’s respectful and responsive engagement with its citizens, the state’s failure to engage through underenforcement”—and overenforcement—“is a . . . democratic harm.”424Natapoff, supra note 9, at 1770–71.

A significant body of research reveals the important connections between police activity and political engagement. Essentially, a dysfunctional police-constituent relationship excises citizens from the political community425 Amy E. Lerman & Vesla M. Weaver, Arresting Citizenship 111 (2014). and results in multiple forms of reduced engagement from that public with local government.426See Jonathan Ben-Menachem & Kevin T. Morris, Ticketing and Turnout: The Participatory Consequences of Low-Level Police Contact, 117 Am. Pol. Sci. Rev. 822 (2022). Additionally, police misconduct can lead citizens to protest, which is itself a form of political engagement, though one that is arguably more consistent with objecting to one’s status as outside the realm of protection offered by the polis. Id. Impacted citizens are less likely to vote,427See id. at 1. less likely to call the local government for assistance via mechanisms like New York City’s general 311 telephone line,428See Lerman & Weaver, supra note 67, at 210. and more likely to experience a sense of alienation and exclusion from the body politic.429Id. at 203. In terms of voting, studies demonstrate that the “most disruptive forms of criminal legal contact,” like arrest and incarceration, significantly “discourage [people from] voting.”430Ben-Menachem & Morris, supra note 426, at 831. Indeed, even low-level versions of this—such as a misdemeanor charge and a day or two in jail—can also markedly decrease the potential of voting in the future.431Ariel White, Misdemeanor Disenfranchisement? The Demobilizing Effects of Brief Jail Spells on Potential Voters, 113 Am. Pol. Sci. Rev. 311, 311, 322 (2019). One study noted that this decrease was especially pronounced for black arrestees: “[w]hite people sent to jail voted a bit less often, but black people turned dramatically away from voting,” with their participation falling by approximately half.432Ariel White, Even Very Short Jail Sentences Drive People Away from Voting, Wash. Post (Mar. 28, 2019, 11:45 AM), []. The decrease was from 26% to 13%. Even inappropriate stops that fell short of arrest had similar impacts on political participation: research has confirmed that “ ‘warrior style’ policing tactics that criminalize routine behaviors and result in arbitrary stops of residents as they move through public space” deter voting and stymie “citizen engagement . . . with the local government.”433Vesla Weaver, Gwen Prowse & Spencer Piston, Withdrawing and Drawing In: Political Discourse in Policed Communities, J. Race, Ethnicity, & Pol. 604, 606 (2020). The authors of one study concluded that people generalized their experiences with the police to the government more broadly. Positive experiences with the police encouraged civic engagement; negative experiences bred “political alienation, distrust, and withdrawal” from civic participation generally.434Lerman & Weaver, supra note 67, at 203.

In other words, police activity “shapes how communities interact with the state more broadly. Citizens learn about how government works and whether government officials value them (and people like them) through their contact with government agencies.”435Id. at 206. Whatever happens with the police interaction will be translated into the broader context: “[i]f citizens experience government as unresponsive, arbitrary, unfair, and authoritarian, as they often do in these encounters, these perceptions may come to inform broader perceptions of government and shape resulting behavior vis-à-vis the state writ large.”436Id. Because “the government officer in the line of duty acts as an agent of the government,” when victims experience harm “at the hands” of a local police officers, “the harm is likely to be experienced as a wrong by the responsible government itself.”437Gregory Sisk, Recovering the Tort Remedy for Federal Official Wrongdoing, 96 Notre Dame L. Rev. 1789, 1810 (2021).

When cities harm citizens through the police, and then go on to assert a public duty defense to claims of police misconduct, they compound the initial harm and inflict an institutional betrayal made even more painful because of the city’s increasingly important role in the average resident’s life. The city is increasingly the most salient form of government for individuals in the United States. This is in some sense a return, as the centrality of local government in hearts and mind has a long history: “[i]n its earliest English meaning, ‘citizenship’ was residence in a city.”438Nancy Fraser & Linda Gordon, Contract Versus Charity: Why Is There No Social Citizenship in the United States?, Socialist Rev., Jan. 1992, at 45, 51. Cities “are the new states: ‘[T]he states’ role as our most salient community has decreased,’ and states ‘no longer play the role in today’s polity that they did at the time of the Founders.’ 439Swan, supra note 55, at 1271 (quoting Kaitlyn Ainsworth Caruso, Associational Standing for Cities, 47 Conn. L. Rev. 59, 87 (2014); Mark C. Gordon, Differing Paradigms, Similar Flaws: Constructing a New Approach to Federalism in Congress and the Court, 14 Yale L. & Pol’y Rev. 187, 209–11 (2010)). Instead, that role now belongs to cities. A 2012 Pew Research Center poll found that “Americans trust local governments at a significantly higher level than both state and federal government,” and many people report feeling greater affinity to their locality than to their state.440Raymond H. Brescia & John Travis Marshall, Introduction, in How Cities Will Save the World: Urban Innovation in the Face of Population Flows, Climate Change, and Economic Inequality 1, 8 (Raymond H. Brescia & John Travis Marshall eds., 2018) (referencing Growing Gap in Favorable Views of Federal, State Governments, Pew Rsch. Ctr. (Apr. 26, 2012), []).

A consequence of these closely felt ties is that when the city betrays that affinity, as happens via police misconduct or failure to protect and the city’s corresponding response, the harm of institutional betrayal compounds the harm inflicted. Institutional betrayal refers to “wrongdoings perpetrated by an institution upon individuals dependent on that institution, including failure to prevent or respond supportively to wrongdoing by individuals . . . committed within the context of the institution.”441Jennifer J. Freyd, Institutional Betrayal and Institutional Courage, Univ. of Or., [] (last updated Feb. 7, 2023). In other words, the “failure of a system that is supposed to protect people” magnifies the injury.442Jennifer M. Gómez & Jennifer J. Freyd, Institutional Betrayal Makes Violence More Toxic, Register-Guard (Aug. 22, 2014, 12:01 AM), []. The injury itself is bad, and the institutional betrayal experienced when that injury goes unredressed and unacknowledged exacerbates that harm.

To fix this and demonstrate the kind of “institutional courage” that can serve as an antidote to this compounded injury,443Id. cities seeking social justice and democratic inclusion should distance themselves from this no-duty rule. There are multiple paths to doing so.444See Reinert et al., supra note 30, at 780. Most conveniently, cities can simply choose not to use it.445Additionally, cities could set up entire alternative forums for these types of claims, as “some would argue that monetary rights against public agents should primarily be a matter of bureaucratic accountability (via ombudsmen and administrative—statutory or non-statutory—compensation schemes) rather than legal accountability (to courts and other judicial bodies).” Cane, supra note 43, at 166–67. City attorney officers craft agendas and priorities for affirmative litigation and enforcement, and they could easily do so in the defensive realm as well.446Reinert et al., supra note 30, at 780. Local governments could simply consciously decide to not rely on the public duty doctrine at all or only in limited circumstances.447Id. at 781 (suggesting that furthering a “public-facing objective” could constitute a legitimate reason to not invoke qualified immunity). They could “instruct their attorneys to decline to insist” on the public duty doctrine’s application,448Id. and instead “determine, as a matter of policy, that they will accept” a duty to protect as a threshold matter, leaving it to the jury to decide whether the standard of care has been met in any given case.449Id.

There is precedent for this kind of action. For example, in Arkansas, Governor Asa Hutchinson directed the agencies of that state to refrain from using sovereign immunity arguments in their defenses because it conflicted with the policy goal of allowing citizens access to courts to petition for redress regarding governmental actions they found unjust.450Alan Greenblatt, A Whopper of a Court Case: Can Citizens Sue States?, Governing (Mar. 22, 2018), []. The governor now requires agencies seeking to assert sovereign immunity to first get approval from the governor’s office so that it can ensure that the use of immunity is appropriate and that “citizens have access to the courts whenever it’s appropriate.”451Id.

Indeed, as Professor Joanna Schwartz and others note, “the manner in which local governments choose to mount legal defenses to liability has a significant impact on the success of civil rights litigation.”452Reinert et al., supra note 30, at 766. The legal barriers that exist are in some sense waivable—“these barriers to effective remediation enter litigation only when the local government defendants present them to the courts for consideration.”453Id. at 761. And defendants need not do so. City attorneys, like government attorneys more generally, have both the authority and indeed the ethical responsibility “to determine what ‘seeking justice’ means.”454Id. at 790 (citing Bruce A. Green, Must Government Lawyers “Seek Justice” in Civil Litigation?, 9 Widener J. Pub. L. 235, 277–79 (2000)). Within that inquiry, “particularly when representing [public] entities or sovereigns,” public attorneys can consider “other relevant government objectives” beyond simply “winning in court,” like “ ‘whether actions are consistent with democratic norms, whether their outcomes enhance the capabilities of the comparatively disadvantaged, and whether they recognize and respect who benefits from a [particular] policy’ or course of action.”455Id. at 789–90; Swan, supra note 55, at 1285 (alteration in original) (quoting Justin Steil & James Connolly, The Just City, in The Wiley-Blackwell Encyclopedia of Urban and Regional Studies (Anthony M. Orum ed., 2018)). These queries would all support embracing an owed duty of care as a threshold matter.456This path may also avoid some of the difficulties that cities encounter when they do attempt police reforms by other means. See, e.g., Benjamin Levin, What’s Wrong with Police Unions?, 120 Colum. L. Rev. 1333 (2020).

B. Sovereignty-as-Responsibility

Even though cities play the most tangible governmental role in the life of the average American457Kaitlyn Ainsworth Caruso, Associational Standing for Cities, 47 Conn. L. Rev. 59, 61 (2014). and cities like New York City have an economy equivalent to that of many countries,458See Richard Florida, The Economic Power of Cities Compared to Nations, Bloomberg: CityLab (Mar. 16, 2017, 11:10 AM), [] (noting that “New York City’s .5 trillion GDP places it among the world’s twenty largest economies, just a tick under those of Spain and Canada,” and “Los Angeles’ 8 billion GDP is [a] bit smaller than Australia’s, with .1 trillion”). the formal legal status of cities in America does not reflect this reality. Perhaps surprisingly, “[t]he American city’s legal and political autonomy has long been precarious.”459Richard C. Schragger, The Attack on American Cities, 96 Tex. L. Rev. 1163, 1166 (2018). Of particular note, in 1907, the Supreme Court in Hunter v. City of Pittsburgh held that municipalities are mere “political subdivisions of the State” which the state may create or destroy at whim,460Hunter v. City of Pittsburgh, 207 U.S. 161, 161 (1907). a holding which continues to haunt cities and their claims to local power.461See Sarah L. Swan, Constitutional Off-Loading at the City Limits, 135 Harv. L. Rev. 831, 883 (2022). Relatedly, cities also have no constitutional status to speak of: localities are “all but invisible to the Constitution”462Richard C. Schragger, The Limits of Localism, 100 Mich. L. Rev. 371, 469 (2001). and there is “no constitutional principle of local autonomy.”463Richard T. Ford, Law’s Territory (A History of Jurisdiction), 97 Mich. L. Rev. 843, 849 (1999). Consistent with the holding in Hunter, “[f]or constitutional purposes, local governments are not solemn political associations but rather subdivisions of state government.”464Id.

This subordinate legal status hamstrings cities in multiple ways. The most obvious one occurs in the context of state preemption. While states have always had at least some preemptory power to override local laws, in the last few decades, states have punitively exercised their preemptive powers to block city action and strip cities of regulatory powers.465Sarah L. Swan, Preempting Plaintiff Cities, 45 Fordham Urb. L.J. 1241, 1243–44 (2018); see also Erin Adele Scharff, Hyper Preemption: A Re-Ordering of the State-Local Relationship?, 106 Geo. L.J. 1469 (2018). Often in response to municipal socially progressive regulation,466Scharff, supra note 465, at 1483. states have been turning to hyperpreemption: state laws which provide that individual local officials can be “sued, fined, or removed (or some combination of all three)” for attempting to enact laws in now-prohibited areas.467Swan, supra note 465, at 1244. Additionally, these hyperpreemption laws provide that local government entities themselves can similarly be financially punished, through either fines or state refusal to provide funds that once were available.468Id.

Cities are constantly trying to overcome such barriers to their ability to govern effectively. They do have some successes: some courts acknowledge that of course cities are more than mere creatures of the state and “not purely administrative components of state government.”469David J. Barron, Why (and When) Cities Have a Stake in Enforcing the Constitution, 115 Yale L.J. 2218, 2238 (2006). Despite the holding in Hunter, in other cases the Supreme Court at least implies—if not outright states—that cities are in fact independently important polities.470Id. And both state legislation empowering cities with home rule powers to broadly legislate on a multitude of issues—and state case law—sometimes reflect this view as well.471“Home rule” refers to a system of municipal empowerment whereby states give municipalities broad regulatory powers, as opposed to requiring them to seek state approval each time they want to legislate. See James D. Cole, Constitutional Home Rule in New York: ‘The Ghost of Home Rule, 59 St. John’s L. Rev. 713, 713 n.1 (1985). Along with these legislative and judicial victories, cities draw upon “state constitutional ‘home rule’ immunity, state constitutional constraints on unfairly targeting local governments, as well as federal equal protection, due process, and other constitutional arguments” to counteract these preemption incursions into their regulatory powers.472 Richard Briffault et al., Am. Const. Soc’y for L. & Pol’y, The Troubling Turn in State Preemption 1–2 (2017). But since cities start from a weaker structural position, their battle is uphill, and “given the contemporary landscape of increasing state hostility, there is an urgency to thinking creatively about new legal arguments, as well as building on instances of successful local advocacy, where they can be found” to bolster city empowerment.473. Id. at 2.

As cities think creatively about how to claim political legitimacy and enhance their legal status, embracing responsibility may provide a nonconventional yet viable path. Although the status of a particular political entity is often discussed in terms of what rights that entity is entitled to, a burgeoning literature in the field of international law suggests that a political entity’s responsibilities are actually key to its status.474See, e.g., Roberta Cohen & Francis M. Deng, Sovereignty as Responsibility: Building Block for R2P, in The Oxford Handbook of the Responsibility to Protect (Alex J. Bellamy & Tim Dunne eds., 2016); Sarah Teitt, Sovereignty as Responsibility, in The Globalization of International Society (Tim Dunne & Christian Reus-Smit eds., 2017). This literature focuses on sovereignty-as-responsibility, arguing that sovereignty, a concept delineating the highest possible status a political body can gain, is intimately connected to the responsibilities a political body owes to its people. In other words, sovereignty—meaning a government’s “ability to exercise supreme legal authority within its borders”475Philip Mayor, Note, Borderline Constitutionalism: Reconstructing and Deconstructing Judicial Justifications for Constitutional Distortion in the Border Region, 46 Harv. C.R.-C.L. L. Rev. 647, 647–48 (2011). and its “right to structure its own internal affairs”476Sovereignty also has an “external component,” focused on the relationship between sovereigns. Id. at 648.—hinges on responsibility.

As cities strive to gain more political and legal recognition, this linkage of sovereignty and responsibility highlights the deep connections between rights, responsibilities, and status. Although responsibility is often the most obscured of the three, the trio of responsibility, rights, and status have long been entwined concepts. For example, traditionally a badge of the status of sovereignty has been the immunity previously discussed: the inherent right to take no legal responsibility because “the king can do no wrong” and the status of the entity is higher than those subject to its laws.477John W. Miller, II, Comment, Sovereign Immunity: The King Can Do No Wrong, 8 Am. J. Trial Advoc. 471, 471 n.1 (1985). But the more modern position, reflected in the voluntary relinquishment of full blanket immunity on the parts of many states and the federal government, is to acknowledge at least a minimal responsibility to one’s citizens.478“Professor Akhil Amar has argued that this accountability is embodied in the first words of the Constitution, ‘We the People,’ a phrase which makes the people sovereign.” Chemerinsky, supra note 155, at 1214. There is a growing recognition that “[s]overeignty . . . does not have to mean that a state can behave in any way it wants toward its own citizens without consequence.”479Cohen & Deng, supra note 474, at 80. Instead, sovereignty, or enhanced political status, also “carries with it a responsibility on the part of governments to protect their citizens.”480Id.

Notably, the very status of legal personhood itself is defined by both rights and responsibilities. Entire conceptions of legal personhood are rooted in the idea of an entity that is both “right-holding” and “duty-bearing.”481 Visa A.J. Kurki, A Theory of Legal Personhood 15 (2019). Although battles for increased status are usually waged in the register of rights, “the ascription of rights has historically been connected with the imposition of societal obligations and duties”482People ex rel. Nonhuman Rts. Project, Inc. v. Lavery, 998 N.Y.S.2d 248, 250 (App. Div. 2014). and “imposing responsibility has been a means of recognizing a certain kind of status held by a particular person or entity.”483Swan, supra note 55, at 890 (emphasis added). For example, those who have not historically received the full benefits of legal personhood, including those subject to slavery and married women,484Fraser & Gordon, supra note 438, at 55 (“In one of the greatest ironies in the history of civil citizenship, the first U.S. Married Women’s Property Act, passed in Mississippi in 1839, was aimed at securing slaveholders’ wives[’] rights over slaves.”). were long denied rights, but less obviously they were also denied legal responsibility. Under coverture for example, husbands, not married women, were held legally responsible when wives committed petty crimes or torts.485Sarah L. Swan, Conjugal Liability, 64 UCLA L. Rev. 968, 978 (2017). Similarly, masters were held liable for the actions of their slaves; enslaved people themselves bore only limited forms of personal responsibility.486Thomas D. Morris, “As If the Injury Was Effected by the Natural Elements of Air, or Fire”: Slave Wrongs and the Liability of Masters, 16 Law & Soc’y Rev. 569 (1981).

Sovereignty-as-responsibility pairs responsibility with status in a similar way. A once myopic focus on the rights of sovereign political entities has, over time, grown to encompass concern about their responsibilities.487Cohen & Deng, supra note 474, at 75. On this view, sovereignty is reframed “in a positive way as a form of responsibility towards a state’s population”;488Id. it is “not merely the right to be undisturbed from without, but the responsibility to perform the tasks expected of an effective government” that is sovereignty’s key defining feature.489 Francis Deng et al., Sovereignty as Responsibility: Conflict Management in Africa, at xviii (1996). Sovereignty-as-responsibility suggests that governments are “first and foremost responsible to protect the human rights of their own citizens in accordance with national and international human rights standards,” and sovereign rights flow from fulfillment of this duty.490Cohen & Deng, supra note 474, at 76–77. So, “[i]f governments fail to meet their obligations, they risk undermining their legitimacy.’ 491Id. at 80.

Under a sovereignty-as-responsibility framework, “the best way to protect sovereignty is to discharge the responsibilities of sovereignty.”492Id. at 82. Cities can use this connection between sovereignty and responsibility to enhance their own status. By accepting responsibility for constituent wellbeing in the form of accepting a legally enforceable duty in the context of policing and beyond, cities can increase their own political legitimacy. Notions and “[i]deas of responsibility are instantiated in, and constituted by, the workings of social and political institutions; appeals to responsibility are used to define the parameters of inclusion in American culture, and those appeals become standards against which new social arrangements are judged.”493. Parker et al., supra note 418, at 1.

Indeed, taking responsibility for police activity specifically has an important history as a core concern in conflicts over sovereignty and democratic governance, and has been a lever for the exercise of enhanced localism in the past.494See O’Rourke et al., supra note 6, at 1366. Fights over the local nature of policing drove much of the initial movement for city autonomy and was part of the basis for the long fight for home rule or localism itself. In the mid-nineteenth century, in an effort to rein in cities and exert more control over them, states attempted to seize control over municipal police forces.495Id. at 1367–68. For example, in 1857, the New York state legislature removed the city’s control over the New York City police department, and gave it instead to a state agency overseen by state commissioners.496Briffault et al., supra note 159, at 361. City constituents protested, such that by 1870, this transfer had been reversed and municipal control reinstated.497See Rick Su, Anthony O’Rourke & Guyora Binder, Defunding Police Agencies, 71 Emory L.J. 1197, 1226 (2022). Indeed, frustration over this state power grab eventually led to heightened state constitutional protections for local home rule and increased local power, particularly over policing.498See O’Rourke et al., supra note 6, at 1367. By also accepting responsibility for the harms policing causes, cities can continue to both gain legitimacy as significant political entities and repair the democratic and practical harms policing wreaks in municipalities across America.


As they vie for increased political and legal recognition, cities are in a complicated “moment of emergent state building.”499Swan, supra note 55, at 1290. This moment is occurring at the same time as “citizens are increasingly expecting their local governments to further their opportunity for human flourishing.”500Id. Banishing the public duty doctrine from cities can further both goals. “[T]he city is the scale where questions of justice are felt concretely as part of everyday life,”501Connolly & Steil, supra note 396, at 1. yet state and local governments have currently “taken relatively few steps to foster an independent local culture of rights recognition and enforcement.”502Reinert et al., supra note 30, at 751. Removing the public duty doctrine and starting with the basic principle that the city and its actors, especially police officers, owe a legally enforceable duty to protect would provide the proper basis for a new, just relationship between cities and their constituents, enhancing both the city’s status in the American polity and the civil rights of all city residents.

* Professor of Law, Rutgers Law School (Newark). Many thanks to the following for their comments and conversations: Rachel Barkow, Anita Bernstein, Stephanos Bibas, Mike Cahill, Martha Chamallas, Jake Charles, Michelle Dempsey, Doug Husak, Brenner Fissell, Paul Heaton, Maggie Lemos, Ben Levinson, Wayne Logan, Sandy Mayson, Darrell Miller, Justin Murray, Anthony O’Rourke, Michael Pollack, Blaine Saito, Cathy Sharkey, the students in the 2022 Civil Rights Seminar at Duke Civil Rights Enforcement, and the participants of the Rutgers Junior Faculty Workshop, the Northeastern Faculty Workshop, and the January 2023 Markelloquium.