by Austin Upshaw *
Since Congress passed the Prison Litigation Reform Act (PLRA) in 1996, courts have struggled to harmonize one of its key provisions—the ‘three-strikes rule’—with the Heck bar, a rule derived from Heck v. Humphrey, which prevents district courts from reaching the merits of Section 1983 claims that attack the validity of a conviction before that conviction has been overturned on appeal or through collateral proceedings. The PLRA’s three-strikes rule provides that if an incarcerated person has had three or more of their suits dismissed for failing to state a claim, they can no longer proceed in forma pauperis. Denial of in forma pauperis effectively prevents indigent incarcerated people from accessing relief in federal court. The crux of the Heck issue is a procedural one: is every Heck dismissal equivalent to a dismissal for failure to state a claim under Federal Rule of Civil Procedure 12(b)(6)? The current circuit split mainly concerns whether Heck created a pleading requirement Section 1983 plaintiffs must plead and prove, or whether it is a waivable affirmative defense. But that focus is misplaced because a Heck-barred complaint can be properly dismissed under 12(b)(6) regardless of whether it is a pleading requirement or an affirmative defense. The better way to resolve the issue is on pragmatic, policy-driven grounds. The Heck bar is best treated as a ripeness doctrine that district courts can bypass at their sound discretion. If Heck is treated as a tool for judicial traffic control, rather than a rigid jurisdictional bar, district courts can ensure meritorious claims are not wrongfully treated as ‘strikes’ under the PLRA.
In Heck v. Humphrey,1 the U.S. Supreme Court instructed lower courts to dismiss civil rights suits brought under 42 U.S.C. § 19832 where a plaintiff’s claim necessarily challenges the validity of a conviction they have yet to reverse on appeal or through a collateral proceeding. The rule’s purpose is to promote consistent and final judicial outcomes. The Heck court worried that if one court awards an incarcerated person damages for a wrongful conviction, but another court denies the same person’s appeal or habeas suit seeking reversal of the same conviction, that would undermine consistency and finality in the legal system.3 In addition to the Heck bar, incarcerated people filing in federal court also have to contend with the Prison Litigation Reform Act (PLRA),4 perhaps the most ubiquitous statute arising in prison litigation. It was thus only a matter of time before courts needed to harmonize the Heck bar with provisions of the PLRA.
The PLRA instructs courts to deny in forma pauperis status—which allows indigent litigants to file lawsuits without paying court fees—to any incarcerated person bringing a civil action who has had three or more suits dismissed for being “frivolous, malicious, or fail[ing] to state a claim.”5 This is referred to as the PLRA’s ‘three-strikes rule.’ Since the PLRA was passed in 1996, the federal circuits have divided over whether a Heck dismissal counts as a PLRA strike. More precisely, courts have divided over the third criteria for a strike: whether Heck-barred suits all categorically fail to state a claim.6 If the Supreme Court decides to take up this question, it should answer in the negative: Heck dismissals do not automatically count as strikes under the PLRA because not every Heck-barred suit categorically fails to state a claim.
The reason Heck dismissals ought not count as PLRA strikes is a pragmatic one. In some instances, the best way for an incarcerated person to ensure the statute of limitations will not run on their potential Section 1983 claims is to file suit before they have exhausted every collateral route to reverse their conviction. In those instances, a district court presented with a Heck-barred complaint should not hastily rule on the merits of the Section 1983 claim—that could undermine Heck’s core holding by getting ahead of the separate proceedings challenging the conviction’s validity. But the court should at least be able to decline to dismiss the Section 1983 suit unless and until the collateral proceedings are resolved with finality. The Heck bar thus ought to be construed as a waivable bar to relief that district courts can use based on their discretion, provided they do not get ahead of courts tasked with ruling on the validity of the conviction. And if the Heck bar is waivable, then not every Heck-barred complaint necessarily fails to state a claim at the time the complaint is filed.
The federal circuits are divided on the question of whether Heck dismissals count as strikes under the PLRA. The core issue behind the circuit split is whether Heck created a new pleading requirement that Section 1983 plaintiffs must include in their complaints (which makes Heck dismissals count as strikes), or whether Heck created an affirmative defense (which makes Heck dismissals not count as strikes).
The Third, Fifth, Tenth, and D.C. Circuits have held that Heck dismissals always count as strikes.7 Heck explains that favorable termination is an “element” that “must be alleged and proved” in a malicious prosecution action, and that “[t]he common-law cause of action for malicious prosecution provides the closest analogy” to a Section 1983 claim alleging an invalid conviction.8 Based on the closeness of this analogy, the Third, Tenth, and D.C. Circuits have all reasoned that favorable termination must be an essential element of a Section 1983 claim challenging the validity of a conviction just as it is an essential element in the malicious prosecution context.9 These courts reason that because favorable termination is an essential element of a Section 1983 claim challenging the validity of a conviction, every Heck-barred complaint necessarily fails to state a claim because they fall short of the pleading requirements.10 Accordingly, Heck dismissals count as PLRA strikes.
The Ninth Circuit instead treats Heck dismissals as affirmative defenses. In Washington v. Los Angeles County Sheriff’s Department,11 the court explained that favorable termination is not a “necessary element of a civil damages claim under [Section] 1983” because Section 1983 “merely requires that a litigant allege a deprivation of any rights, privileges, or immunities secured by the Constitution and law, and that the challenged conduct transpire under color of state law.”12 Favorable termination is thus “not an element” of any Section 1983 claim, so failure to allege favorable termination is not failure to state a claim even though it can be grounds for dismissal of the suit.13 Instead, Washington reads Heck as creating an affirmative defense by analogizing to administrative exhaustion. It explains that “compliance with Heck most closely resembles the mandatory administrative exhaustion of PLRA claims, which constitutes an affirmative defense and not a pleading requirement.”14 Washington cites Jones v. Bock,15 which separately held that administrative exhaustion—the requirement that incarcerated people attempt all possible administrative remedies before bringing suit in federal court—is an affirmative defense. The Ninth Circuit based its analogy to Jones on the fact that, like dismissals for lack of administrative exhaustion, “Heck dismissals do not reflect a final determination on the underlying merits of the case.”16 On this view, the Heck bar seems closer to a ripeness doctrine a defendant can raise in response to a Heck-barred complaint, rather than a facial deficiency with the plaintiff’s pleadings. As a result, Heck dismissals are not always dismissals for failure to state a claim and thus do not always count as strikes.
The Third Circuit in Garrett considered the Ninth Circuit’s approach and rejected it, citing language from Heck that “[w]e do not engraft an exhaustion requirement upon [Section] 1983, but rather deny the existence of a cause of action.”17 Because Heck disclaims the exhaustion analogy, Garrett reasons that “[t]he more apt analogy is the one Heck itself makes: a malicious-prosecution claim, which requires alleging and showing favorable termination to state a claim for relief.”18
There is plenty to say about which side has the better argument and which analogy is stronger. On the one hand, the Garrett court seems to channel the letter of the Heck decision more faithfully. If the ball game is to determine which analogy is closer to the Heck bar—the favorable termination requirement in a malicious prosecution tort, or the administrative exhaustion requirement under the PLRA—then Heck provides an answer. Heck explicitly calls a malicious prosecution tort the “closest analogy” to a Section 1983 suit alleging an invalid conviction.19 But Heck was decided decades ago, so this particular claim the court makes may be worth revisiting. In a new decision resolving the strike-counting issue, the Supreme Court could add new gloss to Heck by deciding whether it is a pleading requirement or affirmative defense.
There is, however, a problem looming over this divide in the federal circuits that makes resolving this particular issue—about pleading requirements versus affirmative defenses—seem much less important. The problem comes from Jones, the case the Ninth Circuit cited in Washington for the proposition that administrative exhaustion is an affirmative defense. In Jones, the Supreme Court grappled with whether administrative exhaustion was a pleading requirement or an affirmative defense, not whether unexhausted claims “fail[s] to state a claim”20 under Rule 12(b)(6).21 As Jones notes, unexhausted claims can be dismissed under 12(b)(6) for failure to state a claim.22 If administrative exhaustion is simultaneously an affirmative defense and a proper ground for 12(b)(6) dismissal, then, extending the analogy to Heck, why should it matter whether Heck is an affirmative defense? If Heck is an affirmative defense but Heck-barred claims can still be dismissed for failure to state a claim anyway, treating Heck as an affirmative defense does not resolve the PLRA strike-counting issue.
The Jones court grappled with whether administrative exhaustion was a pleading requirement or an affirmative defense because the Sixth Circuit had treated it as a pleading requirement and, as a result, developed a screening process that required incarcerated persons to allege and demonstrate exhaustion in their complaints.23 Jones’s reversal of the Sixth Circuit put an end to that practice, but it did not create a new ground for dismissal of unexhausted claims outside of the familiar 12(b)(6) ‘failure to state a claim’ territory. Jones even said as much explicitly, clarifying that “[w]hether a particular ground for opposing a claim may be the basis for dismissal for failure to state a claim depends on whether the allegations in the complaint suffice to establish that ground, not on the nature of the ground in the abstract.”24 So the basis for a Heck dismissal can be ‘failure to state a claim’ regardless of whether Heck is a pleading requirement or an affirmative defense.25 To Washington’s credit, the court notes that even though it concludes Heck is an affirmative defense, “a court may properly dismiss a Heck-barred claim under Rule 12(b)(6) if there exists an obvious bar to securing relief on the face of the complaint.”26 But that caveat undercuts the court’s effort to make the strike-counting issue turn on whether Heck is a pleading requirement or an affirmative defense. Resolution of the issue therefore ought to turn on other decisional grounds, because counter-analogizing between malicious prosecution torts and administrative exhaustion under the PLRA cannot definitively resolve whether Heck dismissals count as strikes.
If Washington’s administrative exhaustion argument cannot resolve the issue definitively, as it cannot in light of Jones, then it might seem like the Third, Fifth, Tenth, and D.C. Circuits have the better of the argument. But that would be getting ahead of things too quickly. The crux of the strike-counting analysis should not be whether the Heck bar, as an abstract procedural matter, is a pleading requirement or an affirmative defense because a complaint can be dismissed for failing to state a claim either way. But it is a non-sequitur to conclude that every Heck-barred claim must necessarily fail to state a claim in all cases just because a Heck-barred complaint could be dismissed under 12(b)(6) regardless of Heck’s status as a pleading requirement or affirmative defense.
Not every Heck-barred complaint should be treated as automatically failing to state a claim. Jones provides that “[a] complaint is subject to dismissal for failure to state a claim if the allegations, taken as true, show the plaintiff is not entitled to relief.”27 It is true that in some situations, Heck-barred claims will meet that standard for dismissal. If, for example, an incarcerated person has exhausted every option available to them to challenge their conviction collaterally or on appeal, and has lost at every turn, then they cannot survive 12(b)(6) dismissal in a Section 1983 suit alleging the invalidity of the conviction. Dismissal of those Heck-barred claims ought to count as PLRA strikes.
However, there is another category of Heck-barred claims that are not as clearly dismissible under 12(b)(6): those in which an incarcerated person files a Section 1983 suit alleging an invalid conviction but has not yet exhausted all the options available to them to reverse their conviction. Polzin v. Gage28 provides a good example. In that case, Gerald Polzin filed a Section 1983 claim for false arrest. The claim attacked the validity of his conviction, but at the time he filed it in federal court, he had a suit pending in state court collaterally challenging the validity of the conviction. When Polzin filed the Section 1983 suit, he asked the district court to stay the proceedings because he knew the Section 1983 claim was Heck-barred while his state court proceeding was pending.29
Why did he file an unripe claim? Because the statute of limitations for his false arrest claim under Section 1983 began running from the time that his false arrest was legalized through process—when he was arraigned—so the clock was ticking.30 And because, as the Polzin court explains, “the rule in Heck does not affect the date when a claim for false arrest under [Section] 1983 accrues or when its statute of limitations is tolled,”31 Polzin’s decision to file the still Heck-barred Section 1983 claim when he did was entirely reasonable. Wallace v. Kato even suggests judges should do exactly what Polzin asked the district court to do in this situation: “[i]f a plaintiff files . . . any . . . claim related to rulings that will likely be made in a pending or anticipated criminal trial[], it is within the power of the district court, and in accord with common practice, to stay the civil action until the criminal case . . . is ended.”32
If Polzin went on to lose his suit in state court collaterally challenging the validity of his conviction, the federal district court would then be required to dismiss the Section 1983 civil action pursuant to Heck. But does it also make sense to count that dismissal as a strike under the PLRA? At the time Polzin filed the unripe false arrest claim, it was not clear that taking the allegations as true would show that he was not entitled to relief, which is the standard for dismissal for failing to state a claim under 12(b)(6).33
The Polzin court itself did not grapple with the strike-counting issue under the PLRA because that issue was not presented to it. It did hold that Heck was waivable though, which should bear on the strike-counting analysis.34 In Polzin, the district court below had declined to stay the proceedings as Polzin requested, and instead dismissed the Section 1983 suit for being Heck-barred while his collateral suit challenging his conviction was pending.35 The Seventh Circuit reversed in part, holding that the Heck bar did not require the district court to dismiss the case.36 It held instead that “[b]ecause [the Heck bar] is not jurisdictional,” it “is subject to waiver” and “district courts may bypass the impediment of the Heck doctrine and address the merits of the case.”37 But Polzin did not remand the case in its entirety to the district court for adjudication on the merits. The Seventh Circuit instead went on to address the merits of the plaintiff’s claims and held that most of them must fail. There was only one claim the court left intact, and it gave the following instruction to the district court on remand: “[i]f the district court decides that dismissal [of the outstanding claim] on the Heck doctrine alone is appropriate, it should dismiss that portion of Mr. Polzin’s complaint without prejudice.”38
Such an instruction complicates the PLRA strike-counting analysis. Despite the reasonableness of Polzin’s decision to file an unripe claim, even a dismissal against him without prejudice counts as a strike under the PLRA.39 There is no good reason for the district court to dismiss Polzin’s Heck-barred claim, even without prejudice, when he in the first instance acknowledged the action was not yet ripe, but filed the claim to preserve it against a statute of limitations bar. Moreover, filing the complaint with a request to stay the proceedings was consistent with the Supreme Court’s guidance in Wallace. The Polzin decision is thus somewhat strange in that it begins to grapple with the impossibility of the situation, but stops short of a true remedy, instead issuing an instruction to the district court that undermined Polzin’s ability to vindicate his rights in the future. Thus, while Polzin is a very instructive case, it does not provide a rule that can be uniformly adopted to solve the puzzle created by the intersection of Heck, Wallace, and the PLRA’s three-strikes rule.
Instead, if the Supreme Court takes up this question, it should confront head-on the impossible situation some incarcerated people face. The Court should fashion a decision that provides clear instructions to district courts for ensuring they do not issue PLRA strikes against litigants seeking to preserve Section 1983 claims that are not yet ripe due to the Heck bar. The Ninth Circuit’s Washington decision offers a helpful insight for how to do this. It explains that in practice, “Heck dismissals reflect a matter of judicial traffic control” to prevent civil actions from collaterally attacking existing criminal judgments.40 This view of Heck, which is along the same lines as Polzin’s construal of Heck as a waivable bar a district court may bypass, reflects confidence in district judges to know they ought not get ahead of other courts tasked with reviewing the validity of a conviction. By conceiving of Heck as a tool of judicial traffic control, rather than a rigid jurisdictional bar, courts are empowered to safeguard people’s rights against untenable situations like the one Polzin was placed in.
The purpose of the PLRA is to “reduce the quantity and improve the quality of prisoner suits.”41 But by counting all Heck dismissals as strikes, including Heck-barred complaints filed in good faith only to preserve the claim against a statute of limitations bar, the interpretation of 28 U.S.C. § 1915(g) urged by the Third, Fifth, Tenth and D.C. Circuits accomplishes only the former. It reduces the quantity of prisoner suits without improving their quality, unduly filtering meritorious claims into strikes against prisoners.
The best way to solve this problem is to give district courts more discretion when applying the Heck bar to control traffic. If a district court misuses that discretion by awarding civil damages for a Section 1983 plaintiff based on a finding of an invalid conviction before parallel state court litigation can reach that result, then that decision should be reversed. But if, as has happened in some cases since Heck was decided, judges manage their dockets in ways that respect parallel state court litigation while also preserving meritorious claims, those judges ought to be lauded for their efforts instead of reversed. Treating every Heck dismissal as a PLRA strike, regardless of the underlying context, undermines both the PLRA’s goal of improving the quality of prisoner suits and the legal system’s “commit[ment] to guaranteeing that prisoner claims . . . are fairly handled according to law.”42 Not all Heck dismissals should automatically count as PLRA strikes. The Supreme Court should reverse Garrett, Colvin, Smith, and In re Jones by issuing a decision embracing the “judicial traffic control” conception of Heck articulated in Washington.
* Austin Upshaw is a J.D. Candidate (2025) at New York University School of Law. This Contribution is a commentary on the problem at the 2025 William B. Spong, Jr. Constitutional Law Moot Court Tournament, hosted by William & Mary Law School. One of the questions presented was whether suits dismissed pursuant to Heck v. Humphrey count as ‘strikes’ against incarcerated people under the PLRA’s ‘three-strikes’ rule. This Contribution distills one side of the argument, and the views expressed herein do not necessarily represent the author’s views.
1. 512 U.S. 477 (1994).
2. 42 U.S.C. § 1983.
3. Heck, 512 U.S. at 484–85.
4. Prison Litigation Reform Act of 1995, 42 U.S.C. § 1997e.
5. 28 U.S.C. § 1915(g) (1996).
6. The Fifth Circuit has also held that all Heck dismissals are categorically frivolous. See Hamilton v. Lyons, 74 F.3d 99, 102 (5th Cir. 1996). The Fifth Circuit is alone in this holding, and thus is not the focus of this Contribution.
7. See Garrett v. Murphy, 17 F.4th 419 (3d Cir. 2021); Colvin v. LeBlanc, 2 F.4th 494 (5th Cir. 2021); Smith v. Veterans Admin., 636 F.3d 1306 (10th Cir. 2011); In re Jones, 652 F.3d 36 (D.C. Cir. 2011).
8. Heck, 512 U.S. at 484.
9. See Garrett, 17 F.4th at 428 (holding that favorable termination is a pleading requirement for Section 1983 claims alleging invalid convictions because “favorable termination [is] an element of malicious prosecution claims”); In re Jones, 652 F.3d at 38 (holding the same because “one element that must be alleged and proved in a malicious prosecution action is” favorable termination (quoting Heck, 512 U.S. at 484)); Smith, 636 F.3d at 1312 (holding the same because favorable termination is an “essential element” of Section 1983 claims in this context). The Fifth Circuit has reached the same result, but it did not as explicitly ground its reasoning in the malicious prosecution analogy. See Colvin, 2 F.4th at 498–99 (“Heck implicates a plaintiff’s ability to state a claim” because “[i]t based its holding on the hoary principle that civil tort actions are not appropriate vehicles for challenging the validity of outstanding criminal judgments” (internal quotations omitted)).
10. See Garrett, 17 F.4th at 427; In re Jones, 652 F. 3d at 38; Smith, 636 F. 3d at 1312-14.
11. 833 F.3d 1048 (9th Cir. 2016).
12. Id. at 1056 (internal quotations omitted).
13. Id.
14. Id.
15. 549 U.S. 199, 203, 212 (2007) (finding that administrative exhaustion is an affirmative defense rather than a pleading requirement because creating a new pleading requirement would “exceed[] the proper limits on the judicial role” and contravene Rule 8(a)’s mandate of a “short and plain statement of the claim”) (citing Fed. R. Civ. P. 8(a)).
16. Washington, 833 F.3d at 1065.
17. Garrett, 17 F.4th at 429 (citing Heck, 512 U.S. at 489).
18. Id.
19. Heck, 512 U.S. at 484 (emphasis added).
20. Fed. R. Civ. P. 12(b)(6).
21. Jones, 549 U.S. at 211.
22. Id. at 216.
23. Id. at 203.
24. Jones, 549 U.S. at 215.
25. Id. (“A complaint may be subject to dismissal under Rule 12(b)(6) when an affirmative defense . . . appears on its face” (internal quotations omitted)).
26. Washington, 833 F.3d at 1056 (internal quotations omitted).
27. Jones, 549 U.S. at 215.
28. 636 F.3d 834 (7th Cir. 2011).
29. Polzin, 636 F.3d at 836.
30. See Wallace v. Kato, 549 U.S. 384, 391 (2007).
31. Polzin, 636 F.3d at 837 (summarizing the Supreme Court’s holding in Wallace, 549 U.S. 384).
32. Wallace, 549 U.S. at 393–94.
33. Fed. R. Civ. P. 12(b)(6).
34. Polzin, 636 F. 3d at 837–38.
35. Id. at 837.
36. Id. at 837–38.
37. Id.
38. Id. at 839.
39. See Lomax v. Ortiz-Marquez, 140 S. Ct. 1721, 1727 (2020) (“A dismissal of a suit for failure to state a claim counts as a strike, whether or not with prejudice.”).
40. Washington, 833 F.3d at 1056.
41. Porter v. Nussle, 534 U.S. 516, 524 (2002).
42. Jones, 549 U.S. at 203.