|

Qualified Immunity, Quantified: How America’s Most Controversial Legal Doctrine Thwarts Civil Rights Plaintiffs

by , , and

Vol. 110 No. 1 (2026) | Judges, Social Media & ‘We the People’ | Download PDF Version of Article
Illustration of a police officer and other government officials inside a large shield, facing a group of citizens, with abstract charts and data graphics in the background.

Drawing on the largest study of qualified immunity appeals to date, the authors argue that the doctrine extends far beyond split-second policing decisions and poses substantial obstacles for civil rights plaintiffs.


When the Minnesota Department of Transportation (MnDOT) awarded a contract to Central Specialties Inc. (CSI) for work on a state highway, Mahnomen County road engineer Jonathan Large was not happy.

CSI wanted to use a Mahnomen County road for empty trucks returning from the worksite. Though the road was open to trucks, and the empty trucks would be well below the posted weight limit, Large told CSI he did not want it using the road. CSI appealed to MnDOT, which essentially told the two parties to work it out.

Large would not budge, so CSI decided to proceed as planned. And Large decided to lay a trap.

The very morning that CSI was to start using the contested road, Large persuaded the county to reduce the road’s weight limit such that the company’s empty trucks would exceed it. CSI was notified of the change less than an hour before it needed the road.

Betting that CSI wouldn’t have time to plot a new course, Large drove out to the road and lay in wait. He stopped and detained the company’s trucks and their drivers for more than three hours until he could get law enforcement on the scene to weigh the trucks. In the meantime, obviously overweight trucks belonging to other companies rumbled past without being stopped.

As Large himself admitted, he had no lawful authority to stop or detain CSI’s trucks and drivers.1 Yet when the company sued him under Section 1983 for unreasonably seizing its trucks and singling them out for the stop, Large was able to escape accountability thanks to a trump card uniquely available to government officials facing civil rights lawsuits: qualified immunity.2 (Our nonprofit, public interest law firm, the Institute for Justice, represented CSI in its appeal to the U.S. Supreme Court. Unfortunately, the high court declined to hear the appeal.3)

America’s Most Controversial Legal Doctrine

Created by the U.S. Supreme Court in 1982, qualified immunity — perhaps the nation’s most controversial legal doctrine4 — blocks civil rights lawsuits against government officials unless plaintiffs can show that officials knew, or should have known, that their actions were unconstitutional based on “clearly established law,” i.e., legal precedent putting the unconstitutionality of the officials’ actions “beyond debate.”5 On occasion, the Court has said that some actions are so egregious that officials can be held liable even without such precedent,6 but it has rarely taken advantage of this “obviousness” exception.7

At the same time, the Court has often directed lower courts “not to define clearly established law at a high level of generality.”8 Little wonder then that courts often interpret the “clearly established” standard strictly, as requiring plaintiffs to show that either the Supreme Court or the same federal circuit in which they are suing has already held virtually identical conduct in virtually identical circumstances to be unconstitutional.9

Indeed, the most minor factual differences can be enough to scuttle a plaintiff’s case. For example, in Latits v. Phillips, a police officer rammed a fleeing motorist’s car and then shot and killed the man as he tried to drive off again.10 The Sixth Circuit held similar conduct unconstitutional in two earlier cases.11 However, in those cases, officers shot motorists who were initiating flight, not ones who were attempting to resume flight, as the motorist in Latits was. This difference was sufficient for the Sixth Circuit to affirm the district court’s grant of qualified immunity in Latits.12

In another case, Sampson v. County of Los Angeles, a social worker sexually harassed and inappropriately touched the guardian of a child whose case he was assigned. The Ninth Circuit granted the social worker qualified immunity because the right to be free from such abuse by public officials was clearly established in workplaces and schools but not in the context of social services.13 The court noted that “the Supreme Court’s exceedingly narrow interpretation of what constitutes a ‘clearly established’ right precludes us from holding what is otherwise obvious to us.”14

Proponents say qualified immunity is needed to give officials leeway to make good faith or reasonable mistakes, especially in tense or dangerous situations requiring split-second decisions. Without this protection, they argue, officials could be chilled from performing their duties, and it could become harder to recruit officials at all.

Meanwhile, detractors argue, among other things, that qualified immunity does too good a job of blocking lawsuits against government defendants, protecting much more than reasonable, heat-of-the-moment mistakes and frustrating civil rights plaintiffs for reasons that have little to do with the merits of their claims.

A study from our firm provides new support for this critique. Titled Unaccountable, the report analyzed the largest and most comprehensive collection of qualified immunity cases ever studied — 5,526 federal appeals from 2010 through 2020 — to answer important questions about the kinds of officials and conduct qualified immunity protects, as well as its impact on civil rights cases.15

Not Just About Police and Excessive Force

A jailer watched as an inmate strangled himself and failed to call for emergency medical assistance when the inmate became unresponsive.16 A university administrator oversaw a sham disciplinary process that resulted in a student’s expulsion.17 A public school gym teacher stuffed a disabled sixth grader into a trash can, leaving the child with post-traumatic stress syndrome.18 Police officers stole over $200,000 in cash and rare coins while executing a search warrant.19 And, of course, a road engineer decided to settle a score by playing traffic cop.20

These claims — all sadly real — are serious. They also look much different from the dominant narrative of qualified immunity, in which the typical government defendant is a law enforcement officer who must make a split-second decision about how much force to use.

The popular focus on police and excessive force might lead one to wonder whether the above examples are outliers in the world of qualified immunity.

Our dataset of federal appeals suggests they are not.21

Kinds of defendants. To be sure, we do see plenty of claims involving police and excessive force in the data. Police were the most common type of official sued, appearing as the only defendants in 50% of appeals.22 Prison officials, whose jobs are also often dangerous, were somewhat common as well; they were the only defendants in 20% of appeals.23 However, fully 21% of qualified immunity cases before the federal appeals courts involved neither police nor prison defendants.24 (See Figure 1.)

Venn diagram showing defendants in federal qualified immunity appeals: law enforcement officials (50%), prison defendants (21%), other defendants (21%), and overlapping categories.

Similarly, claims of excessive force or false arrest25 — the types of claims that seem most likely to arise from tense or dangerous situations — appeared most frequently, but still in only about a quarter of appeals each (some involved both).26 All told, only 23% of the appeals we studied fit the classic mold of a police officer accused of excessive force.27 (See Figure 2.)

Pie chart showing that 23% of federal qualified immunity appeals include excessive force claims against police, while 77% involve other appeals.

If not police or prison officials, who are these “other” defendants in federal qualified immunity appeals? Among the 21% of defendants who were neither police nor prison officials were a wide array of government officials, many of whom rarely, if ever, encounter life-or-death situations in the course of performing their duties. Most often, “other” defendants were mayors or city managers, university or school officials, prosecutors or judges, and child protective services workers. But they also included state ethics commissioners, historic preservation officials, state legislators, zoning board members, horse racing commissioners, and septic system regulators.

Kinds of claims. The constitutional violations at issue likewise often strayed from the dominant narrative. (See Figure 3.) In addition to excessive force and false arrest, discussed above, we frequently saw violations related to care in custody, illegal searches, procedural due process, and parental rights. And fully 18% of appeals alleged a violation of First Amendment rights, making it the third most common rights violation.28

Horizontal bar chart showing alleged constitutional violations in federal qualified immunity appeals. Excessive force is most common, followed by false arrest or malicious prosecution, First Amendment claims, and care in custody.

For the most part, the First Amendment claims in our dataset were not issues like police using force against or arresting unruly protestors or preventing filming of police actions. Instead, a deep dive into a random sample of appeals involving First Amendment claims showed that many of the alleged rights violations did not occur in the heat of the moment at all. In nearly 60% of these appeals, plaintiffs alleged they were victims of premeditated abuse by government officials in retaliation for protected First Amendment activity.29

In the largest share of appeals with premeditated retaliation claims (47%), government workers alleged they suffered reprisals at the hands of their employers for activities such as political association and speech, writing a book as a private citizen, union activity, speaking on matters of public concern, or even declining to speak on matters of public concern.30 For example, in one appeal, the plaintiff was a public school teacher who alleged that school board members decided not to renew her contract after she spoke out in favor of a local bond issue they opposed. 31 In another, the plaintiff was a state university employee who claimed he was fired over his criticism of a U.S. congressman.32

The next-largest share of appeals with premeditated retaliation claims (31%) involved the direct targeting of private citizens for exercising their First Amendment rights,33 including — often — criticizing a government official, speech the Supreme Court has said is “at the very center of the constitutionally protected area of free discussion.”34 The allegations at issue in these appeals often shock the conscience. In one appeal, the plaintiffs were two newspaper executives who alleged county officials orchestrated their late-night arrest and detention in retaliation for a story critical of a grand-jury investigation.35

In another case, a man was shot and killed after police failed to announce their presence as they stormed his business in a raid over a minor reporting violation. The man’s family claimed the raid was retaliation for the man’s speech against a proposed police policy.36 In short, the narrative often used to justify qualified immunity is not the one our data tell. Qualified immunity is a card that can be — and often is — played by any government official or worker for any type of rights violation, including unreasonable, intentional (even premeditated), or malicious conduct.

An Ace Up Defendants’ Sleeves

Not only can any government defendant invoke the doctrine for any violation, but the qualified immunity card also often acts as an ace up the sleeve that gives defendants a winning hand — and one that can beat even meritorious claims.

By the numbers, when qualified immunity is part of a case on appeal, government defendants win more than they lose. Appeals with qualified immunity grants outnumbered denials more than two to one at the appellate level (54% to 26%).37 Likewise, defendants won the appeal at more than twice the rate plaintiffs did (59% to 24%).38 Plaintiffs were especially unlikely to win on appeal when they lost in district court. And even if plaintiffs won the appeal, this merely meant they were one step closer to a trial, not necessarily final victory. Only 73 of the appeals in our dataset — just 1.3% — gave plaintiffs a final, post-trial victory.39

Of course, any number of factors may cause outcomes to favor government defendants over civil rights plaintiffs. Our data point to three such factors: a lack of clearly established law (especially in smaller circuits that publish less), interlocutory appeals, and protracted litigation.

A lack of clearly established law. Pinpointing clearly established law presents a unique challenge for plaintiffs, and this is true in any federal circuit. As Don Willett, judge for the U.S. Court of Appeals for the Fifth Circuit, observed: “Plaintiffs must produce precedent even as fewer courts are producing precedent. Important constitutional questions go unanswered precisely because no one’s answered them before. Courts then rely on that judicial silence to conclude there’s no equivalent case on the books. No precedent = no clearly established law = no liability.”40

However, our data suggest that this challenge is particularly acute in smaller circuits and those that publish less frequently.

Circuits that represent larger populations would be expected to see more appeals and thus to have more opportunities to create clearly established law than circuits that represent smaller populations. And, in fact, the largest circuit, the Ninth, accounts for nearly nine times as many qualified immunity appeals (936) as the smallest, the First (105).41 Unsurprisingly, as Constitutional GPA, another study from our law firm, found, the Ninth Circuit also has substantially more statements of clearly established law.42

Similarly, to the extent that circuits treat only published opinions as setting binding precedent, those that publish more would be expected to have more clearly established law.43 Indeed, even after adjusting for population, higher- publishing circuits tend to have more clearly established law than lower- publishing ones. For example, the Seventh and Eighth Circuits publish more than 75% of their opinions, while the similarly sized Third Circuit publishes only about 25%.44 Overall, the higher-publishing Seventh and Eighth Circuits have more than twice as many statements of clearly established law as the lower-publishing Third Circuit.45

In Willett’s telling, the clearly established standard creates a situation of “[h]eads government wins, tails plaintiff loses.”46 But for plaintiffs in certain circuits, that built-in disadvantage is likely to be compounded by the sheer bad luck of encountering rights violations that had not yet been adjudicated in the places where they occurred.

Interlocutory appeals. Our data suggest another way plaintiffs face an uphill battle in overcoming qualified immunity, regardless of the merits of their claims: interlocutory appeals. Should a plaintiff persuade a district court to reject qualified immunity, government defendants can immediately appeal — and they can do so multiple times in the same lawsuit.47

Defendants are only too happy to exploit this advantage. Interlocutory appeals accounted for nearly 2,000 of the more than 5,500 appeals in our dataset. Indeed, nearly all defendant appeals — more than 96% — were interlocutory.48

Some government defendants may even use interlocutory appeals strategically, filing meritless appeals in an attempt to demoralize plaintiffs and get them to stand down. In our data, 11% of interlocutory appeals were dismissed because they failed to raise an issue within the scope of the appellate court’s authority.49 Such appeals are not designed to be winnable, but they can drag out litigation and wear out even worthy plaintiffs.

Protracted litigation. In addition to giving defendants multiple bites at the apple, these appeals also appear to extend litigation. From filing to an appellate ruling, the median duration of qualified immunity lawsuits in our dataset was three years and two months — 23% longer than the typical civil suit up on appeal.50 Many lawsuits had been open even longer: 29% for more than four years and 8% for more than six.51 We even found ones that had been going on for more than 10 years.52

Such extended litigation eats up substantial judicial resources, and it also risks wearing down all but the most resolute plaintiffs. Few would stick with the fight as long as two Nevada brothers who spent 12 years trying to hold to account a government administrator who stole from their deceased father’s estate. They prevailed at every stage of litigation, defeating motions to dismiss and for summary judgment before being awarded $2.1 million by a jury.

After all three losses, the administrator appealed on qualified immunity grounds, each time extending the case by at least two years but failing to alter the outcome.53

A Deck Stacked Against Plaintiffs

When it created qualified immunity in 1982, the Supreme Court aimed to strike a bargain between the rights of citizens and the ability of officials to vigorously exercise their duties. Four decades later, mounting evidence suggests that the bargain is a lopsided one — that the doctrine stacks the deck against civil rights plaintiffs and in favor of government defendants.

Multiple studies have lent weight to this worry, finding that at the circuit level, qualified immunity favors defendants.54 Our results add to the worry by showing the breadth not only of defendants but also — and even more concerning — of conduct protected by the doctrine. And they help unpack some of the mechanics by which defendants are favored and plaintiffs are disadvantaged.

CSI’s case against Jonathan Large illustrates a number of these concerns. As with many of the government defendants in our dataset, Large is not a law enforcement officer. Instead, he is in a position — road engineer — that should rarely, if ever, put him in the types of tense or dangerous situations where he risks injury or death and might violate someone’s constitutional rights while just trying to do his job.

Likewise, as with many of the violations in our dataset, the conduct at issue was not a reasonable mistake made under difficult circumstances. Rather, what Large did to CSI’s trucks and drivers was intentional and malicious, as well as beyond the scope of his authority. In short, as with many of the federal qualified immunity appeals in our dataset, CSI’s case against Large looks nothing like the good-faith police actions the doctrine is supposedly needed to defend.

This should not be surprising. Despite qualified immunity’s reputation as a police-protection doctrine, its shields have never been confined to police. The doctrine, in fact, stems from a Supreme Court case that had nothing to do with police.55 Moreover, qualified immunity does not demand that conduct be reasonable — only that it not have previously been found unconstitutional. Perversely, and notwithstanding the obviousness exception, qualified immunity’s clearly established standard means the doctrine can — and too often does — protect egregious misconduct that happens to have not come up in litigation before or that has only arisen in a different federal circuit.

With the clearly established standard and given circuit variation, it makes strategic sense for defendants — and especially those who have exhibited the most unreasonable conduct — to claim qualified immunity. It worked for Large, who was able to escape liability precisely because his behavior was so outlandish. Never before in the Eighth Circuit had a government official violated the Constitution in the particular way he did. Never before in the Eighth Circuit had a county road engineer decided to settle a score by playing traffic cop. Notably, he might not have been so lucky in the seven circuits where qualified immunity is available only to defendants acting within the scope of their duties.56

And even if it does not work, and defendants are denied qualified immunity, they have the right of interlocutory appeal. As it happened, Large did not need to invoke that right, but as our data show, other government defendants take advantage of it liberally, and however meritorious some of these appeals may be, many will likely do little more than draw out litigation, clogging up the courts and punishing plaintiffs.

Our research sheds light on the wide variety of government defendants and constitutional violations that can be shielded by qualified immunity. More than that, though, it shows how qualified immunity systematically disadvantages victims of government abuse. In creating the doctrine, the Supreme Court said, “we provide no license to lawless conduct.”57

Unfortunately, qualified immunity appears to do just that.


Jason Tiezzi is a data scientist who frequently collaborates with the Institute for Justice. He holds a master’s degree in data science from the University of Virginia.

Elyse Pohl is a research attorney at the Institute for Justice, where she contributes legal research and analysis to the strategic research team’s publications.

Robert McNamara is the deputy litigation director at the Institute for Justice. He litigates cutting-edge constitutional questions in courts nationwide.

Mindy Menjou is an assistant director at the Institute for Justice, where she edits and manages the strategic research team’s original social science research.


  1. All case details are drawn from Petition for a Writ of Certiorari, Cent. Specialties, Inc. v. Large, 143 S. Ct. 369 (2022) (No. 21-1552) (available at https://duke.is/CSIcert).
  2. See Cent. Specialties, Inc. v. Large, 18 F.4th 989 (8th Cir. 2021) (granting the defendant qualified immunity).
  3. See Cent. Specialties, Inc., 143 S. Ct. at 369.
  4. See Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982) (creating the current doctrine of qualified immunity: “We therefore hold that government officials performing discretionary functions generally are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.”). See also, e.g., Scott A. Keller, Qualified and Absolute Immunity at Common Law, 73 Stan. L. Rev. 1337, 1337 (2021) (“Qualified immunity has become one of the Supreme Court’s most controversial doctrines.”).
  5. Ashcroft v. al-Kidd, 563 U.S. 731, 741 (2011).
  6. See Hope v. Pelzer, 536 U.S. 730, 731 (2002) (the Court’s first use of the “obviousness” exception: “The Eighth Amendment violation here is obvious on the facts alleged.”).
  7. Between 2002 and 2020, the Supreme Court used the carveout in just three of the 25 qualified immunity cases that came before it, including Hope. See Patrick Jaicomo & Anya Bidwell, Recalibrating Qualified Immunity: How Tanzin v. Tanvir, Taylor v. Riojas, and McCoy v. Alamu Signal the Supreme Court’s Discomfort with the Doctrine of Qualified Immunity, 112 J. Crim L. & Criminology 105, 129, 141–43 (2022).
  8. See al-Kidd, 563 U.S. at 742. See also, e.g., Kisela v. Hughes, 584 U.S. 100, 104 (2018) (per curiam); City & County of San Francisco v. Sheehan, 575 U.S. 600, 613 (2015); Mullenix v. Luna, 577 U.S. 7, 12, 16 (2015) (per curiam).
  9. See e.g., Frasier v. Evans, 992 F.3d 1003 (10th Cir. 2021) (finding it was not clearly established that a private citizen has a First Amendment right to record law enforcement performing their official duties in public); Zadeh v. Robinson, 928 F.3d 457 (5th Cir. 2019) (finding it was not clearly established that a warrantless search of medical records by the state medical board violated the Fourth Amendment); Jessop v. City of Fresno, 936 F.3d 937 (9th Cir. 2019) (finding it was not clearly established that police theft of $225,000 seized under a search warrant violated the Fourth Amendment).
  10. Latits v. Phillips, 878 F.3d 541, 551–53 (6th Cir. 2017).
  11. Sigley v. City of Parma Heights, 437 F.3d 527 (6th Cir. 2006) and Smith v. Cupp, 430 F.3d 766 (6th Cir. 2005). (The Sixth Circuit in Latits distinguished the prior two cases where the court found the officers’ actions unconstitutional by saying that in those cases, the police approached a nonviolent driver who initiated flight, while, in Latits, the driver was resuming flight after leading police on a car chase. The difference between resuming and initiating flight was the dividing line between a shooting that was a clearly established violation of constitutional rights and action protected by qualified immunity.)
  12. Latits, 878 F.3d at 551–53.
  13. Sampson v. County of Los Angeles ex rel. L.A. Cnty. Dep’t of Child. & Fam. Servs., 974 F.3d 1012, 1023 (9th Cir. 2020) (holding that “the constitutional right not to be sexually harassed by public officials providing social services was not clearly established outside of the workplace or school contexts.”).
  14. Id. at 1025.
  15. Jason Tiezzi, Robert McNamara & Elyse Pohl, Unaccountable: How Qualified Immunity Shields A Wide Range of Government Abuses, Arbitrarily Thwarts Civil Rights, and Fails to Fulfill Its Promises (2024), https://ij.org/report/unaccountable.
  16. Cope v. Cogdill, 3 F.4th 198, 203 (5th Cir. 2021).
  17. J. Endres v. Ne. Ohio Med. Univ., 938 F.3d 281, 288–91 (6th Cir. 2019).
  18. Mahone v. Ben Hill Cnty. Sch. Sys., 377 F. App’x 913, 914, 916 (11th Cir. 2010).
  19. Jessop, 936 F.3d at 939–40.
  20. Cent. Specialties, Inc., 18 F.4th at 994.
  21. We did not use ChatGPT, Claude, CoCounsel, or any other similar AI service for any part of our study. In a nutshell, here is how we built and parsed the dataset: We searched Westlaw for all federal appellate opinions that contained the phrase “qualified immunity” between 2010 and 2020 and downloaded them. This resulted in a universe of 7,173 opinions, the text of which we turned into machine-readable data. We then turned to developing automated coding methods — commonly called algorithms — to assist in coding the opinions. Specifically, we had a small team of attorneys and others with substantial knowledge of legal matters manually code 604 of the opinions using our codebook, developed in consultation with experienced qualified immunity attorneys. We then used those hand-coded opinions to build algorithms for our various fields of interest. Many of our algorithms involved simple “if . . . then” rules, i.e., if specific language occurred in an opinion, then the opinion was classified a certain way. For example, one of the rules we used to determine whether an opinion was relevant (i.e., whether it really involved qualified immunity) was a search for the court’s recitation of the two-pronged standard of review. (Ultimately, 1,647 of the original 7,173 opinions were not, in fact, relevant, resulting in our final dataset of 5,526 opinions.) Developing the algorithms was an iterative process that relied on our team’s technical knowhow as well as the Institute for Justice’s qualified immunity expertise. When the algorithms were final, we ran them on the remaining opinions, minus 187 we held back for our final performance evaluation, to create our final dataset. We then evaluated the algorithms’ performance by comparing their coding to attorney coding for the 187 opinions we held back. Crucially, the algorithms had never seen these opinions before (i.e., we did not use them to develop the algorithms). The algorithms’ performance was generally very good, with nearly all fields achieving better — often substantially better — performance statistics (“accuracy” or “F1 score,” depending on the field) than those achieved in comparable legal studies. Notably, our algorithms were excellent at identifying excessive force violations (98.1% accuracy) and First Amendment violations (99.4% accuracy) — two of our major focuses in the study. Indeed, we generally confined our analyses of the coded data to fields with strong performance. And humans performed all analyses of the coded data. For transparency and reproducibility, we published online our basic methods, detailed methods, main codebook, final dataset, and the full results of our final performance evaluation at https://ij.org/report/unaccountable/. We also make our programming code available upon request.
  22. Tiezzi, McNamara & Pohl, supra note 15, at 16.
  23. Id.
  24. Id.
  25. Our false arrest category also included malicious prosecution and illegal seizure of a person that did not result in an arrest, such as an illegal traffic stop. Id. at 45.
  26. Id. at 18.
  27. Id. at 23.
  28. Id. at 18.
  29. Id. at 24.
  30. Id.
  31. Utter v. Colclazier, 714 F. App’x 872, 881–82 (10th Cir. 2017).
  32. Cutler v. Stephen F. Austin State Univ., 767 F.3d 462, 466–67 (5th Cir. 2014).
  33. Tiezzi, McNamara & Pohl, supra note 15, at 24.
  34. Rosenblatt v. Baer, 383 U.S. 75, 85 (1966).
  35. Lacey v. Maricopa Cnty., 649 F.3d 1118, 1124–25 (9th Cir. 2011), rev’d in part en banc, 693 F.3d 896 (9th Cir. 2012).
  36. Cass v. City of Abilene, 814 F.3d 721, 727 (5th Cir. 2016).
  37. Tiezzi, McNamara & Pohl, supra note 15, at 21. In the remaining appeals, the courts handed down mixed opinions (i.e., opinions with grants and denials of qualified immunity for different defendants or on different claims) or did not rule on qualified immunity.
  38. Id. at 22. The remaining appeals ended in mixed decisions.
  39. Id. at 26.
  40. Zadeh, 928 F.3d at 479–80 (Willett, J., concurring).
  41. Tiezzi, Mcnamara & Pohl, supra note 15, at 40.
  42. Marie Miller et al., Inst. for Just., Constitutional Gpa: Is Your Government Preventing Accountability? (2022), https://duke.is/constitutionalgpa.
  43. For a discussion of how the federal appellate courts treat unpublished opinions with respect to qualified immunity, see David R. Cleveland, Clear as Mud: How the Uncertain Precedential Status of Unpublished Opinions Muddles Qualified Immunity Determinations, 65 U. Mia. L. Rev. 45, 49–50 (2009) (arguing that federal courts relegate unpublished opinions to “precedential limbo”).
  44. Tiezzi, McNamara & Pohl, supra note 15, at 40.
  45. Marie Miller et al., supra note 42.
  46. Zadeh, 928 F.3d at 480 (Willett, J., concurring).
  47. Bryan Lammon, Rules, Standards, and Experimentation in Appellate Jurisdiction, 74 Ohio St. L.J. 423, 423 (2013) (“Most litigants must therefore wait until the end of proceedings in the district court — when all issues have been decided and all that remains is executing the judgment — before they can appeal.”).
  48. Tiezzi, McNamara & Pohl, supra note 15, at 18.
  49. Id. at 28.
  50. On average from 2010 through 2020, the median length of a civil lawsuit was two years and seven months. For details of how we calculated this figure, see id. at 51 n.61.
  51. Id. at 18.
  52. Id. at 19, figure 5.
  53. Robert Perea, Jury Awards $2.1M to Heirs for Charges Against Glover, Reno Gazette J. (Nov. 16, 2015), https://duke.is/juryawards2mm; Mathis v. County of Lyon, 757 F. App’x 542 (9th Cir. 2018).
  54. See, e.g., Aaron L. Nielson & Christopher J. Walker, The New Qualified Immunity, 89 S. Cal. L. Rev. 1 (2015); Alex Reinert, Qualified Immunity on Appeal: An Empirical Assessment (Cardozo L. Sch. Fac. Rsch. Paper, Paper No. 634, 2021), https://duke.is/qualifiedimmunityappeal; Bryan Lammon, Reforming Qualified-Immunity Appeals, 87 Mo. L. Rev. 1137 (2022).
  55. Harlow, 457 U.S. at 800 (concerning qualified immunity for Nixon administration executives).
  56. See supra note 1.
  57. Harlow, 457 U.S. at 819.