Municipal Liability – The Leading Indicator

Most legal outcomes are decided before a court examines the evidence, before a jury hears the facts, and before anyone argues what actually happened. The machinery that produces outcomes runs on doctrines too abstract for most people, administered by institutions with strong incentives to keep them obscure, whose product is a gap between the right the Constitution names and the remedy the legal system provides. ‘Before the Merits’ strips down the American legal machine to its component parts one case at a time, for general readers to discover how a precedent-based system is run and maintained.

Before the Merits—Monell v. Department of Social Services, 436 U.S. 658 (1978)

In 1871 Congress built a federal emergency room for injuries that local courts had demonstrated they would not treat. For most of a century that emergency room turned away the patients for whom it was built. Then, in 1961, the Supreme Court codified the exclusion by ruling that cities, the institutions most responsible for the injuries, could not be admitted to the facility at all. The case that changed this began with a pregnancy: a class of female city employees forced onto unpaid leave before their doctors required it sued the New York City Department of Social Services, and their lawsuit forced the Court to reopen a door it had sealed seventeen years earlier.

The ruling that followed opened institutional liability and simultaneously installed, inside the opening, a set of intake requirements so demanding that most patients who enter the emergency room never reach a treatment room. The calibration of those requirements was not a discovery from legislative history but a policy choice, and its consequences have fallen for nearly fifty years with systematic precision on the people the 1871 statute was written to protect. Read on to learn the exact intake requirement the plaintiff must satisfy that has nothing to do with whether the constitutional violation occurred.

Case Header:

Case Name: Monell v. Department of Social Services of the City of New York
Citation: 436 U.S. 658 (1978)
Court / Jurisdiction: United States Supreme Court; certiorari to the U.S. Court of Appeals for the Second Circuit
Date Decided: June 6, 1978
Docket No.: No. 75-1914

Legal Domain(s):

Primary Issue(s): Civil Rights / § 1983 Municipal Liability / Official Policy or Custom
Practice Area: Federal Civil Rights Litigation; Constitutional Law; Municipal/Government Liability
Procedural Posture: Class action filed in S.D.N.Y.; district court denied backpay on Monroe v. Pape immunity grounds; Second Circuit affirmed; Supreme Court granted certiorari to decide whether local government officials and school boards are “persons” within 42 U.S.C. § 1983 when equitable or damages relief is sought against them in their official capacities

Holding (One Sentence Rule):

Local governments are “persons” subject to suit under 42 U.S.C. § 1983 and may be held liable for constitutional deprivations caused by an official policy, ordinance, regulation, or governmental custom, but may not be held liable solely on a theory of respondeat superior.

Precedent

The federal emergency room that Congress built in 1871 was constructed for a specific population: the freed people and their allies whom Southern state governments were systematically terrorizing while state courts watched with indifference or complicity. The building was federal, the remedy was damages, and the statute’s language was unqualified. Any person who, acting under color of state law, deprived any citizen of constitutional rights shall be liable to the party injured. Congress knew what it was writing. The legislative debates documented the precise category of official conduct the statute was meant to reach, the sheriff who stood aside while the mob worked, the official who organized the violence, the government that maintained the infrastructure enabling it, and the language it produced was designed to make none of them unreachable.

The emergency room sat largely unused for most of a century, its doors open in principle and functionally closed in practice.

The Supreme Court revived the statute in 1961 in a case involving a Black Chicago man whose home had been subjected to an unconstitutional warrantless search by Chicago police officers. The ruling was significant in what it permitted: officers could now be sued under the federal statute. It was more significant in what it excluded. The Court held that municipalities were not persons within the meaning of the 1871 Act and therefore could not be sued at all.

The reasoning rested on a reading of the legislative history of the Sherman Amendment, a provision Congress had rejected during the Act’s passage, which the Court interpreted as evidence that Congress had not intended local governments to bear liability under the statute. The emergency room was now open to individual patients. The institution that had trained, equipped, directed, and indemnified the person who caused the injury could not be admitted.

The exclusion was the wrong reading of the wrong amendment for the wrong reason, as the Monell majority would later establish with considerable documentary precision.

Congress had rejected the Sherman Amendment because it created new affirmative duties on municipalities to prevent private lawlessness, which raised federalism concerns about imposing obligations local governments had never possessed. It had nothing to do with whether municipalities bore liability for their own existing constitutional violations, which was the question the 1961 ruling had answered by reading the wrong legislative record. The exclusion stood for seventeen years.

The women who forced the correction were not suing over violence. Jane Monell and a class of female employees of the New York City Department of Social Services were challenging a mandatory leave policy that required pregnant employees to take unpaid leave before any medical necessity arose. The policy was enforced as written departmental practice; no individual officer was acting aberrationally. The institution itself had adopted, maintained, and enforced the violation. Their case presented the institutional-liability question in its cleanest possible form, and the Court that received it was positioned, in 1978, to correct a historical misreading it had carried for nearly two decades.

What the correction would cost, and who would pay it, was the question the majority had not fully answered when it wrote the opinion.

Doctrine

Monell v. Department of Social Services, 436 U.S. 658 (1978), is a seven-to-two ruling that took three analytical steps, each of which mattered individually and whose combination produced a doctrine more consequential than any of the steps would have suggested in isolation.

The first step was historical.

Justice Brennan, writing for the majority, conducted an exhaustive re-examination of the Civil Rights Act of 1871’s legislative debates, with particular attention to the Sherman Amendment and the reasons for its rejection. The majority concluded that Monroe v. Pape had misread the record: Congress rejected the Sherman Amendment because it created new affirmative municipal duties, not because it wished to immunize municipalities from liability for their own existing constitutional violations. Local governments were persons within the meaning of the statute from the beginning. Monroe‘s exclusion of municipalities was overruled.

The second step was structural, and it shaped everything that followed.

Having determined that municipalities could be sued, the Court addressed how they could be sued. Here the majority made a choice that the historical analysis did not compel but that the majority presented as its natural implication. Because Congress had rejected the Sherman Amendment, which would have imposed something resembling respondeat superior liability on municipalities, the majority concluded that Congress had rejected respondeat superior itself. A city could not be held liable merely because it employed someone who violated the Constitution. Liability required a connection to an official policy or custom of the municipality, something the institution as an entity had decided or persistently permitted. The intake form now had a required field: show the institutional policy or custom that was the moving force behind the violation.

The third step was definitional, and it opened the doctrine to its subsequent expansion and its subsequent narrowing simultaneously.

The Court defined official policy to include formal written directives adopted through official channels, single decisions by officials with final policymaking authority, and governmental customs, patterns of practice so persistent and widespread that the institution could be said to have chosen them. The definition was capacious by design, intended to capture the variety of ways an institution can direct, sanction, or systematically tolerate constitutional violations without putting a policy in writing. What the definition did not specify was how a plaintiff would prove any of these things before reaching the discovery that would supply the proof.

Justice Rehnquist dissented, joined by Chief Justice Burger. The dissent did not argue for continued immunity; the historical question was, Rehnquist acknowledged, genuinely close. The dissent argued that a seventeen-year-old statutory interpretation should be corrected by Congress, not by the Court’s re-examination of the same legislative record the earlier Court had examined and read differently. The dissent was prescient in one respect—the majority’s historical methodology produced a doctrine whose boundaries were unclear from the day of its announcement.

The following five decades of litigation over what policy or custom meant, who qualified as a final policymaker, and what deliberate indifference required for failure-to-train claims have confirmed that the majority created as much uncertainty as it resolved. The case settled. New York City eventually paid eleven million dollars to the affected women, a decade after the original complaint was filed.

Revival

The doctrine Monell created was incomplete on the day it was issued, and the incompleteness invited a series of subsequent decisions that defined its edges in ways that progressively narrowed the distance between the promise of institutional accountability and its practical availability.

The first significant clarification came in 1986, when the Court held that even a single decision by an official with final policymaking authority could constitute official policy sufficient to trigger municipal liability. The holding mattered because it eliminated the requirement that plaintiffs demonstrate a pattern of prior violations. If the right official made a single unconstitutional decision, the municipality bore liability for its consequences.

The difficulty the subsequent cases revealed was that identifying the right official requires jurisdiction-by-jurisdiction analysis of state and local government structure. The identity of the final policymaker on any given question is determined by state law rather than federal constitutional doctrine. A city council may be the final policymaker on some questions, a police chief on others, a mayor on others still, and the answer varies by state, by city charter, and by subject matter.

Getting it wrong means losing the case before the merits are examined.

The most consequential subsequent development was the 1989 failure-to-train ruling, which held that a municipality’s inadequate training of its employees could constitute the policy or custom required for Monell liability, but only when the failure amounted to deliberate indifference to the rights of persons with whom the untrained employees would interact. The deliberate-indifference standard arrived from the Eighth Amendment‘s prohibition on cruel and unusual punishment, where the term described prison officials who knowingly ignored serious medical needs. Imported into the failure-to-train context, it required plaintiffs to show that the need for training was obvious and that the municipality’s failure to provide it reflected a conscious disregard of predictable consequences. The standard is higher than negligence; it requires something approaching knowing indifference assessed at the institutional level. It also requires proof that the need for different training was apparent from prior incidents, which means the first plaintiff in any new category of institutional failure cannot satisfy it, because the prior incidents that would have put the institution on notice do not yet exist. The emergency room’s intake form now required the patient to demonstrate that prior patients had presented with the same injury.

The 1997 single-incident hiring decision tightened the doctrine further still. A plaintiff who sought to establish municipal liability by pointing to a single negligent hiring decision faced the requirement of showing not merely that the employee turned out to be unfit but that the inadequacy of the hiring decision was so obvious, and the constitutional violation so plainly foreseeable at the moment of hiring, that the decision itself constituted deliberate indifference. The standard was difficult to meet because it required courts to assess foreseeable risk at the moment of a hiring decision rather than in retrospect, which is a counterfactual inquiry that benefits defendants with institutional resources and disadvantages plaintiffs who can demonstrate only what happened, not what was predictable before it happened.

The circuit split that has persisted to the present involves the standalone Monell claim: whether a plaintiff can pursue institutional liability when the underlying claim against the individual officer fails because the officer was dismissed, is immune, or prevailed at summary judgment. Both the Fourth Circuit and the Fifth Circuit require a concurrently viable individual claim; other circuits permit the institutional claim to proceed independently. In jurisdictions requiring a viable individual claim, a municipality can eliminate institutional accountability by successfully defending its officers, which creates an incentive to contest individual liability not because the facts favor the defense but because an individual victory forecloses the institutional question. In those jurisdictions, the intake form can be completed correctly in every respect and still rejected because the person who caused the injury was cleared at a prior window.

Exposure

Every emergency room has an intake process. Every intake process has a governing logic that determines which patients receive treatment and which remain in the waiting area. The Monell intake process is governed by a logic the majority presented as a natural consequence of legislative history. That is, on examination, a set of choices about whose injuries the emergency room was calibrated to treat.

The respondeat-superior bar is the intake process’s most consequential rule, and it was not compelled by the Constitution or by the statute’s text. The text says persons are liable; it specifies no causal theory connecting a municipality’s liability to an employee’s act. The majority derived the bar from Congress’s rejection of the Sherman Amendment, but that rejection was grounded in federalism concerns about creating new affirmative duties, not in a commitment to any particular theory of institutional causation. The majority’s decision to prohibit respondeat-superior was a policy judgment, that cities should not bear automatic liability for everything their employees do in the performance of their duties, dressed in historical reasoning the history did not fully support. The intake form’s required field, show the institutional policy or custom, arrived not from the statute but from a choice the majority made about what the statute should mean.

The practical consequence is that the framework simultaneously answers and refuses to answer the most intuitive question a civil rights plaintiff can ask: if the city’s officer violated my constitutional rights, is the city responsible? The answer is that it depends on whether the violation traces to a policy or custom, who the final policymaker was, whether training was deliberately inadequate, and whether the pattern of violations is documented. None of these requirements tracks the severity of the violation, the completeness of the proof that it occurred, or the extent of the harm. A plaintiff who establishes a single devastating constitutional violation by a city employee, committed without a documented prior pattern or a written authorizing policy, may complete the intake form correctly in every other respect and still be turned away at the treatment window. The form was not designed for that patient.

The intake form’s hidden logic is that the proof it requires is held by the institution whose conduct is being assessed.

Complaint logs, training records, personnel files, prior investigation reports, the documents that would establish a pattern of violations or demonstrate deliberate indifference, are institutional records. They exist inside the emergency room. The patient in the waiting area cannot access them without a court order compelling production, cannot obtain that order without surviving a motion to dismiss, and cannot survive the motion to dismiss without alleging facts that are inside the records they cannot access. The intake form requires the patient to describe their prior treatment history at a facility whose records are sealed to them.

The discovery that would unlock the records is conditioned on the pleading that requires the records. The loop is not a malfunction. It is the intake procedure operating as designed.

This gap between the individual violation and the institutional claim was not an oversight. The majority built it deliberately, under the influence of genuine concern that automatic institutional liability would produce municipal insolvency, deterrence of public service, and political backlash against an expanded civil rights regime. Those concerns were real in 1978, when the implications of fully opening municipal treasuries to civil rights judgments were genuinely unknown. They produced a doctrine whose conservatism at the moment of its creation became progressively more visible as the category of cases it was designed to reach, systematic police misconduct, institutional indifference, official policies of exclusion, turned out to require exactly the tools the majority had declined to provide.

The exposure Monell produces, stripped of its doctrinal complexity, is the disclosure that institutional accountability in American civil rights law is an aspiration encoded in requirements so demanding that most victims of institutional violations cannot satisfy them, not because their violations are not real, not because the institution did not produce them, but because the doctrine that was supposed to reach institutions was calibrated, from its inception, to require more proof than most plaintiffs can access.

Deviation

The emergency room that Monell opened has, in the fifty years since the ruling, developed a waiting area substantially larger than its treatment capacity, and the patients who reach treatment are not a representative sample of the patients who present at the door.

Plaintiffs’ lawyers who specialize in civil rights litigation have adapted by concentrating Monell claims in cases with documented institutional histories: departments with citizen-complaint records, use-of-force incidents on video, civil rights investigations by advocacy organizations, and prior lawsuits that establish the pattern the deliberate-indifference standard requires. These are the cases where the institutional records are already partly public, where the pattern has been documented by someone other than the plaintiff, and where the intake form can be completed without first gaining access to the records behind the intake window. Individual plaintiffs whose cases lack this context, the single incident, the first incident, the incident involving a department that has not previously been sued, face an intake process that is formally available and practically impassable.

Municipal defendants have adapted by contesting the institutional claim at the discovery stage, resisting production of complaint logs, training records, personnel files, and internal investigation reports on relevance, privilege, or burden grounds, knowing that a plaintiff who cannot reach the institutional records cannot establish the institutional pattern. Courts have been inconsistent in managing this resistance. Some permit early discovery of institutional records as necessary to plead the Monell claim adequately. Others permit the defendant to challenge the pleading’s adequacy before discovery proceeds, which places the plaintiff in the position of being required to plead facts they cannot obtain without the discovery they cannot access until they have pled the facts. The waiting area fills. The treatment capacity does not expand.

The indemnification structure of municipal employment completes the circuit and produces its most perverse outcome.

Individual officers who violate constitutional rights face no personal financial consequence because their employers indemnify them for civil-rights judgments entered against them in their individual capacities. The employers, meanwhile, face no institutional consequence because Monell requires proof of policy or custom that individual violations, standing alone, do not establish. The officer is made whole by the institution. The institution is insulated from the officer’s conduct by the doctrine. The public bears the cost of indemnification through the taxes that fund it and receives nothing that requires the institution to change the practices that generated the violation. The civil rights statute imposes theoretical liability on both actors while practical accountability falls on neither. The emergency room bills the patient’s insurance and discharges them without changing the procedure that caused the injury.

The most recent scholarly argument centers on whether Monell liability should extend to municipalities that fail to investigate civil-rights allegations surfaced in their own litigation files. A city that receives repeated civil suits alleging the same category of constitutional violation, defends each individually, settles the most evidenced cases under confidentiality agreements, and never investigates the pattern has demonstrated, through sustained institutional inaction, the deliberate indifference the doctrine requires. The theory is analytically coherent, has not yet commanded a circuit consensus, and would expand Monell‘s practical reach precisely in the domain where the current doctrine most conspicuously fails: the department that reads its own settlement history as a management cost rather than a constitutional signal, and that has learned, over fifty years of litigation, exactly how little documentation it needs to avoid producing before the intake form becomes impossible to complete.

Remedy

The calibration the majority set in 1978 was a policy choice, not a constitutional discovery, and the choice was made with full awareness that it would leave some institutional violations unreachable. The majority said so. The Remedy is not to fault the candor but to name what the choice has produced across five decades of application … and what would need to change to produce something different.

The respondeat-superior bar, the policy-or-custom requirement, the deliberate-indifference standard, and the final-policymaker analysis together create an accountability system in which institutional liability tracks not the harm the institution’s employees inflicted but the institution’s own documented decision-making process, specifically whether the institution chose the violation, permitted it through deliberate inaction, or failed to train against it with knowledge of the likely consequences. This is a coherent theory of institutional moral responsibility. It is also a theory that requires the plaintiff to prove the institution’s internal deliberative record, which is held by the institution, contested at discovery, and obscured by a settlement practice whose confidentiality the doctrine neither requires nor prevents.

A different theory was available and was rejected. Respondeat-superior would have imposed automatic institutional liability for unconstitutional employee conduct within the scope of employment, without requiring proof of policy, pattern, or deliberate indifference. It would have reached genuinely aberrational violations by officers acting against institutional policy, which seems intuitively unfair. It would also have reached the violations that are the predictable product of institutional choices about training, supervision, culture, and hiring that the institution made without formally acknowledging their likely consequences.

The majority rejected it because, in 1978, the full fiscal and political implications of that reach were genuinely unknown, and the caution felt proportionate to the uncertainty.

The standard requires plaintiffs to prove that the institution made a knowing choice to leave a gap it recognized would produce violations. The proof of that knowing choice is institutional: training curricula, supervision records, prior complaint investigations, internal communications. The institution controls it all. The discovery process can compel production, but only after the plaintiff has survived a motion to dismiss by alleging facts the institution has not yet been compelled to produce.

The standard is not impossible to meet; it is met regularly in cases where the institutional record has already been made public by journalism, prior litigation, or civil rights investigation. In those cases, the emergency room treats the patient. In cases where the record has not been made public, the intake form cannot be completed, and the patient waits.

The 1871 statute was written by people who had watched institutions organize constitutional violations while individual actors performed the visible acts. They wrote a statute to reach both. The emergency room they built has treated many patients in the nearly century and a half since. The ones it has not treated are not, in the main, the ones whose injuries were less real or less severe. They are the ones whose injuries arrived without a prior public record of the institution that caused them, without the resources to compel that record’s production before the statute of limitations ran, and without the political visibility that makes an institution’s liability something the institution cannot manage by settling quietly and reforming nothing. The emergency room is not broken. The triage protocol was written by the same institution that built it, and it has been functioning exactly as calibrated for fifty years.

For Further Examination . . .

What Monell v. Department of Social Services forecloses and what the law leaves open are not identical. The gap between them is where the next case will be born. Whatever specific circumstances generate it, the logos of any case has three components, each operating simultaneously as a field condition. Institutional Inertia (+) is already in motion, and Socioeconomic Pressure (-) defines the boundary conditions of that motion. These collide within a Cultural Justice Assumption (0) whose content varies but whose function does not.

Institutional Inertia is the affirming force.

Municipal governments operating civil rights defense programs have internal incentives that run consistently in one direction: defend each case individually, settle the ones with the strongest evidence under confidentiality agreements, and never conduct the internal investigation that would convert a series of individual settlements into the documented pattern the deliberate-indifference standard requires. The incentive does not require malice. It requires only that the institution’s legal department treat each case as a discrete liability event rather than as a data point in a pattern the doctrine is waiting to see assembled, and that the settlement structure preserve the confidentiality that prevents the assembly. The path of least resistance runs toward the waiting area, and it has run that direction for fifty years.

Socioeconomic Pressure is the denying force.

The post-civil-rights era expansion of § 1983 litigation produced a federal judiciary and a municipal defense bar that both understood, by the late 1970s, that full institutional accountability under respondeat-superior would expose city treasuries to liability at a scale that had no precedent in American municipal finance. The fiscal concern was real. It shaped the doctrine at the moment of its creation and has been invoked to justify progressively narrower applications of the standard ever since, in a legal culture where the populations most likely to be injured by systematic institutional violations are the populations least represented in the constituent calculus of the legislators who could correct the doctrine by statute and have not. This force does not push toward any particular outcome. It defines the range of responses the institutional inertia already in motion will be permitted to produce, and it raises the stakes of every other force operating in the field without generating directional momentum of its own.

The Cultural Justice Assumption is the reconciling force.

Institutions are morally responsible for what they choose, not for what they accidentally permit, and the law should require proof of institutional choice before imposing institutional liability. The policy-or-custom requirement operationalizes this assumption precisely: a city that chose a violation, or deliberately disregarded an obvious risk of one, is responsible. A city whose employee committed a violation the city did not authorize is not. This is coherent as a theory of moral responsibility. Its application produces a legal regime in which the institutions most responsible for systematic violations, those that defend each case individually while the pattern accumulates invisibly, are precisely the ones whose conduct most easily satisfies the no-written-policy defense the assumption creates. The assumption is neither validated nor invalidated by the outcome it produces. It is the medium through which institutional inertia and socioeconomic pressure interact, and what makes that interaction legible as justice rather than mechanics.

These forces do not finally resolve with a given case, but reconstitute around the next one. However, a reader who can identify all three next time will find that the following questions transfer. These are the questions this case leaves open rather than the ones it answers.

If a police department receives fifty civil-rights complaints alleging the same category of unconstitutional conduct over five years, defends each individually, settles the most evidenced cases under confidentiality agreements, and produces no written policy authorizing the conduct, has the department demonstrated the deliberate indifference that Monell requires, and if not, what additional fact would need to be true before the doctrine reaches an institution whose settlement pattern is itself the evidence of the pattern the doctrine demands?

The Monell doctrine bars respondeat-superior on the basis of a historical inference about the Sherman Amendment’s rejection that four subsequent decades of scholarship characterize as a significant overreach. Congress has had nearly fifty years to correct the inference legislatively without doing so. If congressional silence is neither endorsement nor oversight but evidence that the constituencies most harmed by the calibration lack the political power to compel correction, what would need to change in the distribution of that power before the correction the 1871 statute’s authors would have recognized as necessary becomes politically available?

If individual officers are indemnified by municipalities for civil-rights judgments entered against them, and municipalities escape institutional liability because individual violations without a documented pattern do not establish policy or custom, and the practical consequence is that no actor in the system bears personal financial consequence for constitutional violations that are individually provable and institutionally predictable, at what point does the civil rights enforcement structure become an apparatus for producing the appearance of accountability rather than its substance, and which institution, if any, is both positioned and motivated to recognize that threshold before it is crossed rather than after?

Before the Merits—Lee v. City of Los Angeles, 250 F.3d 668 (9th Cir. 2001)

Every legal procedure for challenging a wrongful arrest assumes the arrested person can speak. Kerry Sanders had chronic schizophrenia and could not speak for two years, so he served two years in a New York prison for a crime committed by a man who shared only his last name. The Los Angeles Police Department had been given a fingerprint card for the actual fugitive before the arrest, did not use it, and then told Kerry Sanders’s mother for two years that his whereabouts were unknown. That last detail is not incidental. The constitutional system protects the person who can perform the assumed process and leaves the person who cannot to the mercy of institutions with no obligation to notice the difference. Read on to learn how to identify where that protection collapsed and which institution bore the obligation to substitute its own verification for the voice it was not hearing.

Case Header:

Case Name: Lee v. City of Los Angeles
Citation: 250 F.3d 668 (9th Cir. 2001)
Court / Jurisdiction: United States Court of Appeals for the Ninth Circuit
Date Decided: May 4, 2001 (opinion refiled; originally filed February 14, 2001)
Docket No.: No. 99-56020

Legal Domain(s):

Primary Issue(s): Civil rights / § 1983 claims; municipal liability; wrongful arrest and extradition; judicial notice at the pleading stage; notice pleading under Rule 12(b)(6)
Practice Area: Federal civil rights litigation; police misconduct; constitutional torts; disability rights (ADA Title II)
Procedural Posture: Appeal from dismissal of federal claims for failure to state a claim under Fed. R. Civ. P. 12(b)(6); dismissal of state-law claims for lack of subject-matter jurisdiction

Holding (One Sentence Rule):

A § 1983 complaint alleging municipal liability for wrongful arrest, extradition, and two-year imprisonment of a mentally disabled man adequately states constitutional claims under the First, Fourth, and Fourteenth Amendments and survives Rule 12(b)(6) where it alleges a specific policy or custom of identity-verification failures amounting to deliberate indifference, and where the district court erred in relying on extrinsic evidence and taking judicial notice of disputed facts outside the pleadings.

Precedent

The extradition system’s identity-verification architecture was designed on a diagnostic assumption that separates a functional mind from a disordered one: that the signals coming in correspond to the reality outside. When they do, the system processes them accurately and delivers a correct result. When they do not, the system has no internal mechanism for detecting the mismatch, because the mechanism for detection is the signal itself. The protection functions for the person whose signal is legible. For the person whose signal is not, the protection does not function; it simply continues.

The extradition clause of the Constitution requires that a person charged with a crime in one state and found in another shall, on demand of the executive authority of the charging state, be delivered to the demanding state.

The clause was written to prevent accused persons from escaping accountability by crossing state lines. For most of American history, its procedure assumed that the person being extradited was the person charged. The assumption was reasonable in an era of small communities, face-to-face identification, and short distances. It became progressively less reasonable as the population grew, as criminal databases expanded to cover hundreds of millions of records, and as the frequency of common surnames multiplied in proportion to the population itself.

The federal statutes implementing extradition procedure, and the state laws operating alongside them, impose identity-verification requirements, but the requirements are procedural rather than substantive. The demanding state must send documentation; the surrendering state must hold a hearing; the identified person may challenge the identification before extradition is completed. These protections function adequately when the identified person can mount that challenge, which requires saying clearly and convincingly, “I am not the person in this warrant,” producing identification, naming witnesses, and engaging the hearing process in a way that creates a record requiring response.

They function almost not at all when the identified person cannot do any of this.

A person with severe mental illness, with cognitive disability, with limited language, with a history that has severed most institutional connections, enters the extradition process without the tools the process assumes all participants possess. The hearing that is supposed to protect the wrongly identified becomes, in practice, a proceeding at which the person’s inability to mount a coherent defense is read as absence of one. The silence or incoherence of the mistakenly identified person looks, from the outside, like the resigned silence of someone who knows they are caught. There is a more precise version of this failure: a man with schizophrenia, under questioning by officers who have already assigned him a name, may confirm that name, not because he is conceding guilt but because the condition that prevents him from correcting the identification also prevents him from resisting the institutional framing that surrounds the questioning. The signal the process receives is the one it created.

The legal architecture built around this process—the civil rights statutes, the constitutional doctrines, the pleading standards governing how a wronged person can reach a court—was developed without particular attention to the category of person most likely to be victimized by its failure. The result is a body of law theoretically adequate to reach institutional misconduct of this kind and practically organized to dismiss it before the evidence can be examined. What the law had not resolved, and what this case would force into view, was whether any institution in the sequence bore an obligation to substitute its own verification for the signal it was not receiving.

Doctrine

The legal system’s first response to a two-year wrongful imprisonment was a question about paperwork: had the complaint been properly drafted? That question, and the procedural machinery it triggers, is the Doctrine section’s subject. The merits of the imprisonment, whether Kerry Sanders should have been arrested at all, had not yet become the case. They would not become the case until the paperwork question was answered, and the system that answered it had no mechanism for asking the underlying question first.

Lee v. City of Los Angeles, 250 F.3d 668 (9th Cir. 2001), arrived before the Ninth Circuit as an appeal from a dismissal under Rule 12(b)(6), which permits a court to terminate a complaint before discovery on the ground that even if the plaintiff’s allegations are true, no legal claim has been stated.

Kerry Sanders was a mentally disabled resident of Los Angeles, a man with chronic schizophrenia and hallucinations, who was arrested by the Los Angeles Police Department in October 1993. He was misidentified as Robert Sanders, a fugitive who had absconded from a New York state prison work-release program. The two men merely shared a surname. The New York State Division of Criminal Justice Services had provided the LAPD with an identification packet that included fingerprints and physical characteristics of Robert Sanders. The LAPD did not compare Kerry Sanders’s fingerprints to those in the packet before arresting him and initiating extradition. Kerry Sanders was extradited to New York and imprisoned at Greenhaven Correctional Center for approximately two years. His mother, Mary Sanders Lee, contacted the LAPD repeatedly during those two years to ask about her son’s whereabouts and was told, each time, that his location was unknown. In October 1995, federal drug agents arrested the actual Robert Sanders in another jurisdiction. Kerry Sanders was released. He had served two years for someone else’s crime, in a state he had not entered voluntarily, while his mother was told he could not be found.

Mary Sanders Lee, as conservator for her son, filed a federal complaint against the City of Los Angeles, individual LAPD officers, and New York state officials, alleging constitutional violations under 42 U.S.C. § 1983 and separately alleging violations of the Americans with Disabilities Act. The district court dismissed all federal constitutional claims by relying on materials outside the complaint: court records and official documents, including a declaration from the New York defendants asserting that Kerry Sanders had himself told the officers he was Robert Sanders. The court took judicial notice of these materials without giving the plaintiffs an opportunity to contest their use, converting what was formally a pleading-stage ruling into a factual determination made on evidence the plaintiffs had never been permitted to challenge.

The Ninth Circuit reversed in part. The court identified two distinct errors, each compounding the other. The district court had taken judicial notice of disputed facts drawn from extrinsic materials that the pleading stage did not permit, and it had used that unauthorized factual determination to defeat claims that, read on the face of the complaint alone, adequately alleged constitutional violations.

On the Fourth Amendment claim, arrest without probable cause, the court held that the plaintiffs had adequately alleged that no reasonable officer could have believed Kerry Sanders was Robert Sanders, given his obvious mental incapacity and the mismatched identifying characteristics the LAPD had been provided before the arrest. On the Fourteenth Amendment claim, the court held that the plaintiffs had adequately alleged that the LAPD maintained a policy or custom of extraditing individuals without proper identity verification, which, if proven, could constitute the deliberate indifference to constitutional rights that Monell liability requires. On the familial-association claim, belonging to the mother rather than the son, the court held that the LAPD’s alleged practice of concealing Kerry Sanders’s whereabouts while telling his mother he was unknown to them adequately alleged a constitutional interference with a protected family relationship.

The opinion contains a fourth holding that the draft’s original version omitted and that is, in the long view, the most consequential of the four.

The Ninth Circuit reversed the district court’s dismissal of the ADA claim with prejudice and reinstated it for amendment, holding that Title II of the Americans with Disabilities Act covers law enforcement activities, including arrests and extraditions, not merely the delivery of services in accessible facilities. The accommodation theory the court recognized is precise: because Kerry Sanders’s schizophrenia prevented him from communicating his identity or navigating the extradition process, the accommodation the law required was not a ramp or an interpreter but a fingerprint comparison, a step the institution would have been obligated to perform for a person who could not advocate for himself, though it might have skipped for a person who could. The court also recognized a distinct discrimination-by-attribution theory. An institution that interprets a disabled person’s symptomatic behavior as criminal conduct, rather than as a signal requiring a different response, has not merely made an error; it has discriminated against the person on the basis of their disability by treating the disability’s manifestation as evidence of guilt.

The opinion additionally holds that the New York state officials who requested the extradition, sent identification materials into California, and traveled to Los Angeles to take custody were subject to California personal jurisdiction, because they had deliberately used California’s criminal justice system as the instrument of the harm rather than merely foreseen that harm would be felt there.

Revival

Lee‘s three holdings have aged at different rates, and the rate of aging tracks a single variable: whether subsequent doctrine required more from the plaintiff or less from the institution.

The procedural track, the ruling that a district court cannot take judicial notice of disputed extrinsic facts when evaluating a Rule 12(b)(6) motion, has become the most durable portion of the opinion. Courts confronted with defendants who attempt to paper a motion to dismiss with official records, agency documents, or court files the plaintiff has not incorporated by reference cite Lee for the rule that such materials, if their accuracy is disputed, cannot be used at the threshold stage to defeat the plaintiff’s well-pleaded factual allegations. The rule is not unique to Lee, reflecting a principle embedded in the Federal Rules of Civil Procedure, but the opinion’s application of it to a municipal civil rights case involving official misconduct has made it a frequently invoked authority in exactly the context where the temptation to use extrinsic materials is strongest: cases where the defendant is a government entity with access to official records the plaintiff has never seen.

The ADA track has settled in one direction and remained unresolved in another. The holding that Title II of the Americans with Disabilities Act covers law enforcement activities, including arrests and extraditions, has been broadly followed and is now settled doctrine in the Ninth Circuit and most other circuits. No serious argument remains that police functions fall outside Title II’s scope. The holding’s more specific applications have fared differently. The failure-to-accommodate theory in the identity-verification context, the proposition that an institution must perform an independent check when a subject cannot advocate for themselves, has been litigated with mixed results and has not produced a clear circuit consensus on what accommodation is required, when it is triggered, and whose failure to perform it generates liability. The discrimination-by-attribution theory has been applied inconsistently across circuits; some courts treat the institutional misreading of disability symptoms as criminal behavior as independently actionable, while others require proof of intentional discrimination before the theory generates liability. The door Lee opened in 2001 remains open. It has not been walked through with confidence.

The substantive Monell track has aged differently, shaped by subsequent Supreme Court decisions that tightened the pleading standards governing civil rights complaints. After 2007 and 2009, a complaint must plead enough facts to make a claim plausible, not merely conceivable, and formulaic recitation of the elements of a cause of action no longer suffices. Lee‘s holding that the plaintiffs’ policy-and-custom allegations were adequate under the pleading rules of its era does not translate automatically into adequacy under current standards, and a lawyer who cites Lee for the proposition that a pattern-and-practice allegation survives a motion to dismiss must account for the intervening evolution.

The tightening did not eliminate the Monell theory; it raised the price of entry.

The jurisdictional holding has been specifically narrowed by Walden v. Fiore. The requirement now is that personal jurisdiction over an out-of-state official must rest on the official’s purposeful contacts with the forum state itself, not merely on foreseeable harm to a forum-state resident. Lee‘s jurisdictional analysis survives Walden in its specific factual configuration, because the New York officials did far more than foresee harm; they affirmatively engaged California’s institutions as instruments of the result. The extension of that analysis to less factually dense configurations is foreclosed, and what remains is a holding narrowed to the precise facts that generated it, potent within those facts and inert outside them. The question the narrowing raises (without answering) is whether an institution that causes a two-year wrongful imprisonment by deliberately using a forum state’s criminal machinery, but does so in a way that a future court finds insufficiently contact-rich, has placed itself beyond the reach of that forum’s courts entirely.

Exposure

Three institutions processed Kerry Sanders’s two-year wrongful imprisonment, and none of them malfunctioned. Each operated within its own procedural logic, received the outputs of the prior institution as confirmed inputs, and produced results that were, by its own internal standards, correct. The criminal justice system identified a match and initiated extradition. The disability support system was told its subject could not be located. The legal remedy system received a complaint and evaluated it against its pleading standards. The error did not propagate because any institution failed to do its job, but because each institution’s job was defined without any mechanism for detecting whether the adjacent institution’s output was accurate.

No single system bore the obligation to check—the failure lived in the space between them.

The criminal justice system’s identity-verification procedures assume that the subject is capable of participating in their correction. When an officer compares a person’s appearance to a physical description and finds approximate matches, the system expects the subject to offer a countervailing account: this is not my name, here is my identification, check these records. Kerry Sanders could not do this in a way the system recognized as a claim requiring investigation. His schizophrenia manifested in ways that officers and officials could read as the resigned presentation of someone who knew they were caught, and the system has no internal mechanism for distinguishing “I cannot explain myself coherently” from “I have no coherent explanation.” The filter that would have caught the difference, fingerprint comparison against the packet from New York, was not used. The backup the system provided for exactly this category of error was available and ignored, and the system recorded the result as a successful identification.

The disability support system fails at a different point in the sequence. Mary Sanders Lee, as conservator for her son, is the person best positioned to identify the error and demand correction. She attempted to do exactly that, repeatedly, for two years. The LAPD’s alleged response, that her son’s whereabouts were unknown, was not merely unhelpful; it was, if the plaintiffs’ allegations were true, a lie. Kerry Sanders was in LAPD’s own records as a person the department had extradited to New York on its own initiative. The institution that had placed him on a plane possessed his location and denied that possession to the one person the disability support system had positioned to advocate for him. When an institution removes the conservator’s information, it does not merely obstruct the conservator; it eliminates the safeguard the disability system had deployed as its substitute for the person who could not speak.

The legal remedy system’s exposure runs through the pleading architecture itself. To reach discovery, the plaintiffs had first to survive a motion to dismiss. To obtain the institutional records that would confirm or deny the factual allegations they had already made, they had first to survive the motion to dismiss. The district court’s use of extrinsic materials to dismiss the complaint before discovery was precisely the move the Ninth Circuit’s reversal corrected. The correction cost years of litigation the family bore entirely. The system contains an ultimate correction; it does not contain a mechanism for bearing the correction’s cost on behalf of the party who was already wronged.

What the legal remedy system produced, at the end of years of appellate process, was the right to start again.

The case’s hidden gift is the one the constitutional claims cannot carry and the ADA was designed to reach. The accommodation theory the Ninth Circuit recognized does not require proof of what the institution knew or intended, only proof that the institution failed to respond to what was obvious. Kerry Sanders’s schizophrenia was, by the plaintiffs’ allegations, obvious; an institution exercising reasonable attention would have recognized that his presentation required a different response than the one it gave.

The accommodation the theory requires is not elaborate: a fingerprint comparison the institution already possessed the materials to perform. The discrimination-by-attribution theory the court also recognized goes further still, holding that an institution which reads disability symptoms as criminal behavior has not merely made a factual error but has discriminated against the person on the basis of the condition that produced the symptoms. Together, these theories offer a cause of action that the constitutional framework, which requires proof of deliberate indifference or an established pattern, cannot provide for the single, isolated, preventable failure.

The ADA track is the legal system’s one instrument calibrated for the case where the failure was not a policy, not a custom, not a pattern, but a single institution’s failure to notice what was in front of it. That instrument remains available. It remains unapplied with consistency.

The gap between availability and application is where the next Kerry Sanders enters the system.

Deviation

The legal framework Lee created operates at the wrong point in the sequence. It intervenes after the two-year imprisonment, after the settlement negotiation, after the confidential resolution, and before none of the institutional decisions that produced the error. The case established the doctrinal basis for seeking damages, but did not establish a mechanism for interrupting the process before the damages accumulate.

The case that produced the Ninth Circuit’s reversal and remand was never publicly resolved.

Civil rights cases of this structural type rarely are. The litigation continues past the appellate correction, reaches the settlement table, and ends under a confidentiality agreement that prevents the institutional record from becoming public. The institution pays without admitting that its practice was the moving force behind the violation; no official is named in a public verdict; the policy or custom that produced the error is reformed quietly or not at all, and the next instance produces the next complaint without the prior settlement serving as any kind of public acknowledgment that the structural condition persists. The settlement functions, from the institution’s perspective, as a cost of doing business rather than a corrective signal. The assumption that the error was exceptional rather than structural is preserved intact by the agreement that prevents its examination.

The pattern the subsequent cases reveal is not linear improvement. Identity-verification failures in the extradition context continued after 2001 in forms that differ from Kerry Sanders’s case in detail but not in form. In 2022, a woman was arrested at Los Angeles International Airport and held for thirteen days on a warrant for a fugitive she did not resemble. Officers did not compare her driver’s license or fingerprints against the warrant despite the physical differences. The database produced a name match; the institution treated the match as sufficient; the subject had no way to correct the error through the channels the process assumed were available to her.

The Lee framework gave her a theory of recovery. It gave her nothing that interrupted the thirteen days. The gap between the legal framework’s operation and the moment the error begins is not a gap that post-error damages litigation was designed to close.

Twenty years of cases have not closed it.

The ADA track’s post-Lee development is where the structural gap becomes most visible. The ADA’s accommodation theory is the only legal instrument that operates, in principle, at the moment of the error rather than after it. An institution that fails to perform a fingerprint comparison when the subject’s obvious disability prevents the subject from requesting one has not waited for a pattern to develop; it has discriminated in a single encounter that a single plaintiff can challenge without proving institutional history.

That theory is available. Its consistent application to wrongful extradition has not materialized. Courts that have addressed the failure-to-accommodate question in law enforcement contexts have divided on what accommodation is required before the accommodation obligation is triggered, what counts as “obvious” disability in an arrest context, and whether the discrimination-by-attribution theory requires proof of intentional misattribution or can be established by demonstrating that the institution’s response to symptomatic behavior was objectively unreasonable. The biometric verification technology that would make the accommodation trivially easy to perform has existed and improved continuously since Lee was decided. The legal obligation to perform it has not developed at the same pace.

The difference between the tool’s availability and its legal compulsion is a choice about whose interests the ADA protects at the threshold moment, and the series of cases that have declined to resolve it have left the answer to the next plaintiff who can afford to litigate past the motion to dismiss.

Remedy

The constitutional system protects the person who can perform the process as assumed and leaves the person who cannot to the mercy of institutions that have no obligation to notice the difference. That gap is not an accident, but the design. Can anything in the current legal architecture close it before the next Kerry Sanders enters?

The answer is NO, and not for lack of technology.

The technological infrastructure for rapid biometric verification has improved beyond recognition since 1993. Fingerprint databases that once required days to search are now searched in minutes. Facial recognition systems exist that can, in principle, compare a detention subject against a warrant photograph in real time. The tools an officer needs to verify identity before committing to an extradition process are better than at any point in the procedure’s history, and they were sufficient for the purpose in 1993, when LAPD possessed the fingerprint card and did not use it. The problem the technology does not solve is the institutional imperative to use it, because the legal consequence of skipping the step is neither swift nor certain enough to change the decision at the moment it is made.

The qualified-immunity standard protects officers who made identification errors unless the right to accurate pre-extradition identification was clearly established at the level of specificity the doctrine requires. The right cannot be clearly established without judgments that reach the merits and, again, judgments cannot reach the merits when the cases that would establish the right are dismissed or settled before judgment. The municipal-liability standard requires proof of a pattern or policy, which means one wrongful identification, however devastating, is not enough. Only after a pattern emerges can the municipality’s failure to prevent it constitute deliberate indifference. The combination produces an institutional environment in which the legal cost of individual errors is low enough to be absorbed without generating systemic change,.

The precondition-versus-best-practice distinction is not a technical legal question but a structural decision with known beneficiaries and known costs.

Several states have enacted specific identity-verification requirements in their extradition statutes, requiring biometric comparison before proceedings are initiated, and these requirements carry automatic legal consequence for their violation rather than liability contingent on proving institutional pattern and deliberate indifference. Federal law has not followed. The variation across jurisdictions does not track technological capacity. It tracks political will, which in this domain is shaped by the distribution of visibility. The people most likely to be wrongly extradited, people with severe mental illness, with cognitive disability, with limited institutional connections, are also the people whose wrongful imprisonment is least likely to generate the kind of sustained public attention that makes political will available.

A mind that recovers from a psychotic episode cannot retrieve the time the episode consumed; it can only understand, in retrospect, what that time contained. The legal instrument the Lee opinion created has the same orientation. It faces back toward what has already occurred and measures the institutional failure against a damages standard. The measurement is accurate, but it faces the wrong direction. What would need to change to reorient it is a decision, already made in several state legislatures and available to be made federally, that identity verification before extradition is not a courtesy extended to the identified person but an obligation the institution owes independent of whether the identified person can demand it.

The courts have not made that decision. The legislature has not made it nationally. The cost of the refusal is paid, quietly and repeatedly, by the person who could not speak.

For Further Examination . . .

What Lee v. City of Los Angeles forecloses and what the law leaves open are not identical. The gap between them is where the next case will be born. Whatever specific circumstances generate it, the logos of any case has three components, each operating simultaneously as a field condition. Institutional Inertia (+) is already in motion, and Socioeconomic Pressure (-) defines the boundary conditions of that motion. These collide within a Cultural Justice Assumption (0) whose content varies but whose function does not.

Institutional Inertia is the affirming force (+).

Law enforcement institutions calibrated for the median case process thousands of extradition requests against standardized procedures whose internal validity check is the procedure itself. The LAPD’s alleged failure to compare a fingerprint card it already possessed was not, if the allegations were accurate, a product of hostility toward Kerry Sanders; it was a product of institutional economies in which verification is a step that seems redundant until it is not, and the cost of skipping it falls on the subject rather than the institution. The inertia does not require malice. It requires only that the path of least resistance run consistently toward the next case, and that the cost of the error remain invisible at the level where the procedure is designed.

Socioeconomic Pressure is the denying force (-).

The networking of state criminal databases during the late 1980s and early 1990s produced efficiency gains whose false-positive rate was low enough to be invisible at the institutional level and catastrophic at the individual level. An era that treated the database match as sufficient identification created a category of wrongly identified persons whose only mechanism for correction was their own coherent voice, precisely at the moment when the expansion of the database system made coherence the rarest resource among those most likely to be mismatched. This force does not push toward any particular outcome. It defines the range of responses the institutional inertia already in motion will be permitted to produce, and it raises the stakes of every other force operating in the field without generating directional momentum of its own.

The Cultural Justice Assumption is the reconciling force (0).

Criminal justice processing is accurate by default and wrong by exception, and the person in custody who cannot explain themselves coherently is probably there correctly. The district court treated this as procedurally obvious; the Ninth Circuit treated it as an assumption requiring evidence before it could defeat a well-pleaded complaint. The fault line runs precisely through the phrase “obviously cannot be the fugitive,” and what the case decided is not which side of the line Kerry Sanders stood on but whether the institution bearing the fingerprint card had an obligation to look before drawing the line. The assumption is neither validated nor invalidated by the legal outcome it produces. It is the medium through which institutional inertia and socioeconomic pressure interact, and what makes that interaction legible as justice rather than mechanics.

If a person with severe schizophrenia cannot coherently assert their own identity during an extradition hearing, and the fingerprint card that would have demonstrated the error was available but unused, and the institution that possessed both the card and the person’s location told the person’s conservator for two years that the person was unknown to them, which institution in that sequence bore the constitutional obligation to substitute its own verification for the voice it was not hearing, and what would a future plaintiff need to prove, that no plaintiff has yet been required to prove, to convert that obligation from a moral claim into a legal one?

The Cultural Justice Assumption that criminal processing is accurate by default has been most effectively challenged not by litigation but by the proliferation of biometric verification technology that makes independent confirmation trivially inexpensive; if that technology continues to develop until a jurisdiction’s failure to deploy it before extradition becomes indefensible as a resource claim, what would the institutional inertia already in motion be forced to produce, and would the legal doctrine governing wrongful extradition evolve to meet the new factual baseline or remain calibrated to the era when verification required resources the institution could plausibly claim not to have?

If the ADA’s failure-to-accommodate theory were applied consistently to require independent identity verification whenever a subject’s obvious disability prevents them from performing the process the process assumes, the Cultural Justice Assumption would shift from “probably correct” to “cannot be assumed correct without verification,” and the institutional inertia that currently treats verification as a best practice would be required to treat it as a precondition; what socioeconomic pressure would need to be operating, and in what form, for the courts to be willing to impose that structural obligation on institutions whose incentive architecture has been designed to absorb the cost of the occasional error rather than to eliminate the error’s preconditions?

The Leading Indicator

Tag: Municipal Liability