Subjects and lessons to explore.

Law

Criminal Justice & Criminology · Policing

Police Use of Force

14 min read 5 practice questions
9 key termsSourcesReport an error
Want it in plain words first? Jump to Eli explains — the same idea, no jargon.
On this page 9 sections
  1. In 30 seconds
  2. Why this matters
  3. The college version
  4. Eli explains
  5. Worked example
  6. Key takeaway
  7. Quick check
  8. Study tools
  9. Sources & references

In 30 seconds

Police use of force raises three separate questions that students routinely collapse into one: was the force constitutional, was it a crime, and did it break the department's own rules. The Supreme Court set the constitutional minimum in Tennessee v. Garner (1985) and Graham v. Connor (1989). Agency policy usually sits above that minimum, and the United States still has no complete national count of how often force is used.

Why this matters

Use of force is where police authority is most visible and most disputed, which makes it the topic where confident but wrong claims are easiest to make. Knowing the standards lets you read a court opinion, a department policy, and a news report and say which question each one is answering. The same discipline applies to the numbers: knowing what a voluntary federal collection, a household survey, and a journalist-built database can each support keeps you from citing a figure the data cannot carry. Courses, policy work, and research all assume this separation between what the law is, what the evidence shows, and what people argue should change. This lesson is educational material about United States law, not legal advice.

The college version

One incident, three separate questions

When an officer uses force, the encounter can raise three legally distinct questions, and each has its own decision-maker, its own standard, and its own consequence. The first is constitutional: did the force violate the Fourth Amendment? That is usually litigated as a civil damages suit under 42 U.S.C. Sec. 1983, which lets a person sue anyone who, acting under color of state law, deprives them of rights secured by the Constitution. The second is criminal: did the officer commit an offense under state criminal law, or willfully deprive someone of a federal right under 18 U.S.C. Sec. 242? The third is administrative: did the officer break the employing agency's own rules? That is decided inside the agency and can end in retraining, suspension, or dismissal. The same facts can produce three different answers, because the questions differ and so do the people deciding them. Most confusion about use of force comes from treating one of those answers as though it settled the other two.

The constitutional floor, part one: Tennessee v. Garner

Tennessee v. Garner, 471 U.S. 1, decided March 27, 1985, tested a Tennessee statute allowing an officer to use 'all the necessary means' to arrest a suspect who flees or forcibly resists. Justice White wrote for six Justices; Justice O'Connor dissented, joined by Chief Justice Burger and Justice Rehnquist. The Court held that apprehension by is a , so it must satisfy the Fourth Amendment's reasonableness requirement, and that using deadly force to prevent the escape of all felony suspects, whatever the circumstances, is constitutionally unreasonable. The operative rule is narrow and conjunctive: such force 'may not be used unless it is necessary to prevent the escape and the officer has probable cause to believe that the suspect poses a significant threat of death or serious physical injury to the officer or others.' Where feasible, some warning should be given. Two details students often miss. The statute was not struck down on its face: where the officer does have probable cause to believe the suspect threatens serious physical harm, preventing escape by deadly force is not constitutionally unreasonable. And the Court declined to read the Fourth Amendment through the old common-law rule permitting whatever force was necessary to catch a fleeing felon, reasoning that changes in what counts as a felony and in the weapons available had distorted that rule beyond recognition.

The constitutional floor, part two: Graham v. Connor

Four years later, Graham v. Connor, 490 U.S. 386, decided May 15, 1989, settled which constitutional test governs excessive-force claims generally. Chief Justice Rehnquist wrote for the Court, with Justice Blackmun concurring in part and in the judgment for himself and Justices Brennan and Marshall. The holding: all claims that law enforcement officers used excessive force - deadly or not - in the course of an arrest, investigatory stop, or other seizure of a free citizen are analyzed under the Fourth Amendment's standard, not under substantive due process. That displaced the lower courts' test from Johnson v. Glick, which had asked whether force was applied in good faith or instead 'maliciously and sadistically.' Reasonableness, the Court said, requires balancing the intrusion on the individual's Fourth Amendment interests against the governmental interests at stake, with 'careful attention to the facts and circumstances of each particular case, including the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether he is actively resisting arrest or attempting to evade arrest by flight.' Those three considerations are known as the , and the word 'including' matters: the list is illustrative, not exhaustive. Two further instructions do most of the analytic work. Reasonableness 'must be judged from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight,' and the calculus 'must embody allowance for the fact that police officers are often forced to make split-second judgments - in circumstances that are tense, uncertain, and rapidly evolving - about the amount of force that is necessary in a particular situation.'

What the constitutional standard does not decide

Graham's test is objective, and the Court was emphatic about what that excludes. The question is whether the officer's actions were objectively reasonable in light of the facts and circumstances confronting them, 'without regard to their underlying intent or motivation.' An officer's evil intentions will not turn objectively reasonable force into a Fourth Amendment violation, and an officer's good intentions will not rescue objectively unreasonable force. The test is also keyed to the person's legal status at the time: Graham governs force against a free citizen during a seizure, while the Court noted that the Eighth Amendment, which speaks of cruel and unusual punishments and does invite inquiry into subjective state of mind, applies only after the State has secured a formal adjudication of guilt. Most importantly, the standard is a floor: it fixes the minimum the Constitution demands. It does not ask whether the force was wise, whether a different option existed, or whether the officer followed training. Those are real questions; they are simply not the one the Fourth Amendment asks.

Agency policy sits above the floor

Because the Constitution sets a minimum, employers can and often do demand more. The U.S. Department of Justice's own policy, Justice Manual 1-16.000, is a documented example. It adopts Graham as the baseline, then goes further: DOJ officers may use deadly force only when necessary, meaning the officer reasonably believes the subject poses an imminent danger of death or serious physical injury, and never solely to prevent a suspect's escape. The policy prohibits chokeholds and carotid restraints unless the standard of necessity for deadly force is satisfied, a subsection updated in January 2025. Sections updated in July 2022 require training in de-escalation and its use where objectively feasible, a against another officer's excessive force, and a duty to request or render medical aid. Section 1-16.700 then says the policy 'is not intended to, and does not, create any right or benefit, substantive or procedural, enforceable at law or equity.' That sentence is the whole relationship in miniature: a policy can require more than the Constitution while creating no new legal claim, so breaking it is a disciplinary matter rather than automatically a constitutional one. Note the scale of the variation: this policy binds DOJ components, while the Bureau of Justice Statistics counted 17,541 state and local agencies performing law enforcement functions in June 2018, each writing its own policy under its own state's law.

The accountability routes and their different standards

In civil litigation under Sec. 1983, an officer sued personally can raise . Harlow v. Fitzgerald (1982) framed it as shielding government officials performing discretionary functions from civil damages 'insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.' Pearson v. Callahan (2009), a unanimous opinion by Justice Alito, described the two questions - whether the facts make out a violation of a constitutional right, and whether that right was clearly established at the time - and held that courts need not take them in that fixed order. Qualified immunity is a doctrine about damages liability in civil suits; it is not a defense to a criminal charge and has nothing to do with internal discipline. Criminal exposure runs on different tracks. A federal prosecution under Sec. 242 requires that the officer acted 'willfully,' a demanding mental state, and demands proof beyond a reasonable doubt. State prosecutions apply state law, which varies. California permits deadly force only when the officer reasonably believes, on the totality of the circumstances, that it is necessary to defend against an imminent threat of death or serious bodily injury, or to apprehend someone fleeing after a felony that threatened or caused such harm and who the officer reasonably believes will cause it again unless immediately apprehended. Finally, there is a systemic route. Under 34 U.S.C. Sec. 12601, enacted in 1994 and formerly numbered 42 U.S.C. Sec. 14141, the Attorney General may bring a civil action for equitable and declaratory relief to eliminate a of law enforcement conduct that deprives people of federal rights. The Civil Rights Division states that harm to one person or an isolated action is usually not enough, and that the section has no authority over federal law enforcement agencies.

The measurement problem

The United States has no complete national accounting of police use of force, and any lesson that hands you one number is misleading you. The FBI's National Use-of-Force Data Collection gathers incidents in which an officer's force led to death or serious bodily injury, or in which an officer discharged a firearm at or in the direction of a person. Participation is voluntary: the FBI's own Federal Register notice of May 8, 2025 lists the obligation to respond as voluntary and reports 12,861 agencies enrolled as possible respondents. When the Office of Management and Budget approved the collection in September 2018 it set publication thresholds by the share of sworn officers represented - nothing publishable below 40 percent, limited information at 40 to 59 percent, the most frequently reported responses at 60 to 79 percent, and unconditional publication at 80 percent or more. GAO reported in December 2021 that as of that November the FBI had reached 44 percent participation for 2019 and 55 percent for 2020. The Bureau of Justice Statistics measures from the other side. Its Police-Public Contact Survey, a supplement to the National Crime Victimization Survey, asks a nationally representative sample of residents aged 16 or older about contacts with police. In 2022, about 19 percent of residents - 49.2 million people - had contact with police, and 748,800 of them, 1.5 percent, reported the threat or use of nonfatal force, down from 1,045,600 and 1.9 percent in 2020. That is what residents report, not what agencies record, and by construction it does not cover fatal encounters. Independent databases built by journalists and researchers exist because official data are incomplete. A 2021 Lancet study by the GBD 2019 Police Violence US Subnational Collaborators compared the National Vital Statistics System with three such databases, estimated 30,800 deaths from police violence between 1980 and 2018 (95 percent uncertainty interval 30,300 to 31,300), and estimated that the NVSS did not report 55.5 percent of them. That is a modeled estimate with stated uncertainty, not a count, and the authors' interpretation of the racial patterns they report is theirs. The rule for your own writing: name the source, say what it measures, and date the figure.

What the evidence supports about reducing force

De-escalation training is the most frequently recommended reform, and the honest summary is that the evidence is thin but improving. A 2020 systematic review by Engel, McManus, and Herold found 64 de-escalation training evaluations over 40 years, mostly in nursing and psychiatry, reporting few adverse consequences and slight-to-moderate improvements but concluding that almost all the designs were of questionable quality; at that point no study had shown such training reduced force in the field. In 2022, Engel, Corsaro, Isaza, and McManus published a stepped-wedge randomized controlled trial of the Police Executive Research Forum's ICAT training in the Louisville Metro Police Department and reported statistically significant reductions after training: 28.1 percent in use-of-force incidents, 26.3 percent in citizen injuries, and 36.0 percent in officer injuries. That is one agency, one curriculum, one period - encouraging evidence, not a settled finding. Two better-known interventions fare worse on the same test. NIJ's CrimeSolutions rates body-worn cameras Ineffective for their effects on officer use of force, drawing on a 2020 meta-analysis by Lum and colleagues that found no statistically significant impact across 26 studies; that profile was posted in January 2022. CrimeSolutions likewise rates crisis intervention teams Ineffective for reducing trained officers' use of force in encounters involving people with mental illness, based on Taheri's 2016 analysis of five studies, in a profile dating to October 2018. Both pages now carry a notice that they are no longer updated, so read them as ratings of the evidence available then. This lesson is educational material about United States law and research. It is not legal advice, it does not evaluate any particular incident, and law and policy vary by jurisdiction and change over time.

Eli, the EliExplains learning guide

Eli explains

The same idea, in plain words

Explain it like I’m 10

Officers are allowed to use some force to make an arrest, so the legal question is never simply whether force was used. It is whether the amount of force was reasonable given what the officer knew in that moment, judged the way a careful officer standing there would have seen it rather than the way it looks once everyone knows how the story ended. For deadly force against someone running away, the rule is stricter: it takes both a real need to stop the escape and good reason to believe the person is a serious danger to somebody. And one incident gets graded three separate times, by a court, by a prosecutor, and by the officer's own employer, using three different rulebooks.

Picture it like this

Think of a building code. The code sets the minimum a house must meet to be legal. A landlord can add house rules that are much stricter than the code, and a tenant who breaks a house rule has not thereby broken the law. Constitutional standards are the code; department policy is the house rules.

Where the picture stops working

The analogy breaks in two places. A building inspector measures against fixed numbers before anything goes wrong, while reasonableness is judged afterwards by people weighing contested accounts of a fast, frightening event, and there is no tape measure for it. And a house rule and a building code both concern the same building, whereas the criminal question adds something a code inspection never involves: a person's liberty, decided only on proof beyond a reasonable doubt.

Worked example

Work a hypothetical rather than a headline. Officers stop a driver for expired registration. The driver argues, refuses to step out, and pulls away when an officer takes hold of his arm; the officer grips harder and handcuffs him, and the driver's wrist is injured. Constitutional question: run the Graham factors. The underlying offense is minor, which weighs against significant force; any immediate threat has to be assessed from what the officer could perceive at that moment; and there was active resistance to being removed from the car. The court asks whether the force was objectively reasonable to a reasonable officer on the scene, not whether a better option can be identified afterwards, and the officer's irritation is beside the point. Criminal question: a state prosecutor would apply that state's assault or use-of-force statute, and a federal prosecutor under Sec. 242 would have to prove the officer acted willfully, both beyond a reasonable doubt. Administrative question: the agency asks whether its own policy was followed - a policy that may require de-escalation where feasible and may forbid conduct the Constitution would permit. Three questions, three standards, and potentially three different answers on identical facts.

Key takeaway

Graham makes the Fourth Amendment question one of objective reasonableness judged from a reasonable officer's position on the scene, and Garner adds a stricter rule for deadly force against someone fleeing. The constitutional minimum, the criminal law, and the agency's own policy remain three different tests, and no U.S. source counts all police uses of force.

Quick check

3 questions here, of 5 in this lesson’s practice set. Answers stay hidden until you check.

Question 1 of 3foundational

Under Graham v. Connor (1989), which constitutional standard governs a claim that an officer used excessive force while making an arrest or investigatory stop of a free citizen?

Choose an answer, then check it.
Question 2 of 3intermediate

Tennessee v. Garner (1985) permits deadly force against a fleeing suspect only when which condition is satisfied?

Choose an answer, then check it.
Question 3 of 3intermediate

An officer uses force during an arrest while privately hoping to punish the arrestee. A court finds the amount of force objectively reasonable in light of the facts confronting the officer. Under Graham, how does the officer's motive bear on the Fourth Amendment question?

Choose an answer, then check it.
Practice all 5

Keep learning

Ready to build on this? Continue to the next lesson.

Practice this lesson
Study tools & related lessonsYou’ll learn to · Common mistakes · Easily confused · Key vocabulary · Related

You’ll learn to

  • State the holdings of Tennessee v. Garner and Graham v. Connor and apply the Graham factors to a fact pattern.
  • Distinguish the constitutional standard from state and federal criminal standards and from internal agency discipline.
  • Explain qualified immunity and federal pattern-or-practice authority as legal doctrine and enforcement power rather than as outcomes.
  • Evaluate what the FBI's use-of-force collection, the BJS Police-Public Contact Survey, and independent databases can and cannot support.
  • Assess the strength of the evidence behind de-escalation training, body-worn cameras, and crisis intervention teams as force-reduction measures.

Common mistakes

  • Reading 'objectively reasonable' as 'the least force that would have worked, as we can see now.'

    Graham directs courts to judge from the perspective of a reasonable officer on the scene rather than with the 20/20 vision of hindsight, and to allow for split-second judgments in tense, uncertain, and rapidly evolving circumstances. Showing a better option existed does not by itself establish a violation.

  • Arguing that a hostile or a well-meaning officer motive decides the Fourth Amendment question.

    The inquiry is objective and made without regard to underlying intent or motivation. Motive can matter elsewhere - a federal prosecution under Sec. 242 requires willfulness - but it does not decide reasonableness.

  • Treating a policy violation as an automatic constitutional violation, or treating compliance with policy as constitutional immunity.

    Policy generally sits above the constitutional floor. The Justice Department's own policy states that it creates no right or benefit enforceable at law, so breaching it is a disciplinary matter that must still be analyzed separately from the Fourth Amendment question.

  • Reading a decision not to prosecute, or an acquittal, as a finding that the force was constitutional.

    A criminal case asks a different question, of a different decision-maker, on proof beyond a reasonable doubt, and federal charges under Sec. 242 require willfulness. A civil Fourth Amendment claim, an internal investigation, and a prosecution can each come out differently.

  • Citing one database's total as the national number of people killed or injured by police.

    No U.S. source produces a complete national count. The FBI collection is voluntary and has not met its publication thresholds, the BJS survey measures what residents report about contacts, and independent databases and modeled estimates come with their own coverage limits and uncertainty intervals. Name the source, say what it measures, and date the figure.

Easily confused

Tennessee v. Garner (1985) vs. Graham v. Connor (1989)

Garner answers a narrow question - when deadly force may be used to stop a fleeing suspect - and requires both necessity to prevent escape and probable cause of a significant threat of death or serious physical injury. Graham answers the general question and supplies the framework for all excessive-force claims by free citizens: Fourth Amendment objective reasonableness, assessed with the Graham factors. Garner is one application of the balancing Graham later describes.

Constitutional standard (civil suit under Sec. 1983) vs. Criminal standard (state law, or 18 U.S.C. Sec. 242)

The constitutional question asks whether the seizure was objectively reasonable, is decided in civil litigation, and can be met with qualified immunity as a defense. The criminal question asks whether the officer committed an offense defined by a state statute, or willfully deprived someone of a federal right, and requires proof beyond a reasonable doubt. Different standard, different burden, different decision-maker.

Either legal question vs. Internal agency discipline

Discipline measures the officer's conduct against the employer's own policy, which is usually stricter than the Constitution, and is decided inside the agency with employment consequences. A finding of a policy violation is not a court's finding of a constitutional violation, and a lawful use of force can still be disciplined.

FBI National Use-of-Force Data Collection vs. BJS Police-Public Contact Survey

The FBI collection is agency-reported and voluntary, covering death, serious bodily injury, and firearm discharges at or toward a person; participation has fallen short of the thresholds set for publication. The PPCS is a household survey of residents aged 16 or older, so it captures what people say happened to them across all contacts, including non-injurious force, but it cannot capture fatal encounters or agency records.

Qualified immunity vs. The merits of the constitutional claim

Qualified immunity asks whether the right the officer allegedly violated was clearly established at the time, so a plaintiff can lose on immunity without any ruling that the force was reasonable. Pearson v. Callahan lets courts decide that question first and leave the merits undecided, which is why immunity rulings should not be reported as findings that the force was lawful.

Key vocabulary

Seizure
A restraint on a person's freedom of movement brought about by government action, including an arrest, an investigative stop, and apprehension by deadly force.
Deadly force
Any application of force that creates a substantial risk of causing death or serious bodily injury, a definition California's statute states expressly and applies to firearm discharges.
Objective reasonableness
The Fourth Amendment test asking whether an officer's actions were justified by the facts and circumstances confronting them at the time, without regard to intent or motivation.
Graham factors
Three non-exhaustive considerations the Supreme Court named in 1989: severity of the crime at issue, whether the person poses an immediate threat, and whether the person actively resists or flees.
Constitutional floor
The legal minimum below which official conduct becomes unlawful; statutes, agency policies, and training may demand more, but they cannot demand less.
Qualified immunity
A defense in civil damages litigation that shields officials performing discretionary functions unless their conduct violated clearly established rights a reasonable person would have known.
Clearly established law
Legal rules settled enough when the conduct occurred that a reasonable official in that position would have understood the conduct to be unlawful.
Pattern or practice
Systemic conduct by an agency rather than an isolated act, which the Attorney General may challenge in a civil suit seeking equitable and declaratory relief.
Duty to intervene
An affirmative obligation, imposed by some agency policies, requiring an officer to act to prevent or stop another officer's excessive use of force.

Sources & references

  1. Tennessee v. Garner, 471 U.S. 1 (1985) — U.S. Government Publishing Office (govinfo), United States Reports
  2. Graham v. Connor, 490 U.S. 386 (1989) — Supreme Court of the United States (U.S. Reports, digitized by the Library of Congress)
  3. Harlow v. Fitzgerald, 457 U.S. 800 (1982) — United States Reports volume 457, Library of Congress digital collection
  4. Pearson v. Callahan, 555 U.S. 223 (2009) — United States Reports volume 555, Library of Congress digital collection
  5. 42 U.S.C. 1983 - Civil action for deprivation of rights — Office of the Law Revision Counsel, U.S. House of Representatives
  6. 34 U.S.C. Sec. 12601 - Cause of action (police pattern-or-practice authority) — Legal Information Institute, Cornell Law School (mirror of the U.S. Code)
  7. 18 U.S.C. Sec. 242 - Deprivation of rights under color of law — Legal Information Institute, Cornell Law School (mirror of the U.S. Code)
  8. Justice Manual 1-16.000 - Department of Justice Policy on Use of Force — U.S. Department of Justice
  9. Conduct of Law Enforcement Agencies (Special Litigation Section) — U.S. Department of Justice, Civil Rights Division
  10. Law Enforcement: DOJ Can Improve Publication of Use of Force Data and Oversight of Excessive Force Allegations (GAO-22-104456) — U.S. Government Accountability Office
  11. Agency Information Collection Activities; Extension of a Previously Approved Collection; National Use-of-Force Data Collection (90 FR 19534, May 8, 2025) — Federal Bureau of Investigation, U.S. Department of Justice, in the Federal Register
  12. Contacts Between Police and the Public, 2022 (NCJ 308847) — Bureau of Justice Statistics, U.S. Department of Justice
  13. Census of State and Local Law Enforcement Agencies, 2018 - Statistical Tables (NCJ 302187) — Bureau of Justice Statistics (BJS), U.S. Department of Justice
  14. Practice Profile: Body-Worn Cameras' Effects on Police Officer Behavior — CrimeSolutions, National Institute of Justice, U.S. Department of Justice
  15. Practice Profile: Crisis Intervention Teams (CITs) — CrimeSolutions, National Institute of Justice, U.S. Department of Justice
  16. Does de-escalation training work? A systematic review and call for evidence in police use-of-force reform — Engel, R. S., McManus, H. D., & Herold, T. D., Criminology & Public Policy 19(3), 721-759 (2020)
  17. Assessing the impact of de-escalation training on police behavior: Reducing police use of force in the Louisville, KY Metro Police Department — Engel, R. S., Corsaro, N., Isaza, G. T., & McManus, H. D., Criminology & Public Policy 21(2), 199-233 (2022)
  18. Fatal police violence by race and state in the USA, 1980-2019: a network meta-regression — GBD 2019 Police Violence US Subnational Collaborators, The Lancet 398(10307), 1239-1255 (2021)
  19. California Penal Code Sec. 835a - Use of force by peace officers — California Legislative Information, California Legislature

EliExplains lessons are original prose written from the open, credible references above. See Copyright & Licensing.

Researched 2026-08-18

Educational content only. It is not medical, legal or professional advice. Found an error? Tell us.