Learn About the Law
Get help with your legal needs
Therl TAYLOR, Plaintiff – Appellant, v. Virginia GRUBBS; Pamala Smith; Ann Hallman; Sherman Anderson, Present Chief; Valerie Jones; Supervisor Randall Williams; John Pate; Allendale Fairfax County-City, Defendants – Appellees.
Therl Taylor, Plaintiff – Appellant, v. John Pate; Randall Williams; Teresa Ramsey; Connie Buehner; LT. Carter, SCDC; SCDC; Jane DOE, Employees; John DOE, Employees; Pamela Smith, AI Grievance Coordinator, Defendants – Appellees.
Therl Taylor, Plaintiff – Appellant, v. Walter Worrick; Virginia Grubbs; Pamela Smith; SCDC LT MR C Hartley; John Pate; Bryan Stirling; Jane Does; John Does, Defendants – Appellees.
Ordinarily, a federal plaintiff lacking sufficient funds may proceed in forma pauperis, that is, without prepaying the filing fees, when filing a complaint or an appeal in federal court. But under the Prison Litigation Reform Act (“PLRA”), a court may not grant in forma pauperis status to a prisoner if he “has, on 3 or more prior occasions, while incarcerated or detained in any facility, brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted.” 28 U.S.C. § 1915(g) (the “three-strikes rule”). This case requires us to decide whether an indigent prisoner may proceed in forma pauperis on appeal from the order assigning his third strike. For the reasons that follow, we conclude that he can and so grant his motions to do so.
Recognizing the importance of providing equal access to federal courts, Congress has provided that indigent prisoners may, before a court rules on the merits of a case, seek to proceed “without prepayment of fees or security therefor.” 28 U.S.C. § 1915(a)(1). Prisoners who do so do not avoid paying the filing fees entirely. Rather, the statute “requires that a prisoner (1) pay an initial partial filing fee based on the funds available in the prisoner's account; and (2) make monthly payments of ‘20 percent of the preceding month's income credited to the prisoner's account.’ ” Tolbert v. Stevenson, 635 F.3d 646, 648 n.1 (4th Cir. 2011) (quoting 28 U.S.C. § 1915(b)(1)–(2)).
Even so, Congress has long recognized the potential for abuse in this system, which reduces the immediate “economic incentive to refrain from filing frivolous, malicious, or repetitive lawsuits.” Coleman v. Tollefson, ––– U.S. ––––, 135 S. Ct. 1759, 1762, 191 L.Ed.2d 803 (2015) (internal quotation marks omitted). Concluding this was especially true for prisoner suits, Congress “enacted a variety of reforms designed to filter out the bad claims and facilitate consideration of the good” through the Prison Litigation Reform Act of 1996 (“PLRA”). Jones v. Bock, 549 U.S. 199, 204, 127 S.Ct. 910, 166 L.Ed.2d 798 (2007); see also Coleman, 135 S. Ct. at 1762.
The “three-strikes rule” is one such reform. It provides that a prisoner may not bring a civil action or an appeal in forma pauperis if he has had three prior actions or appeals dismissed as frivolous, as malicious, or for failure to state a claim upon which relief may be granted. 28 U.S.C. § 1915(g). In this case, we must decide whether an indigent prisoner may proceed in forma pauperis on appeal from the very order assigning his third strike.
Therl Taylor, an indigent state prisoner, filed three pro se civil rights actions in the District of South Carolina against various employees of the South Carolina Department of Corrections and the City of Allendale (collectively, “Appellees”).
In the first action, filed on December 14, 2015, Taylor alleged that seven employees of the South Carolina Department of Corrections and the City of Allendale had denied him access to the prison's mailroom services, interfering with his ability to petition the courts. In the second action, filed against a similar group of corrections officials on June 20, 2016, Taylor alleged that the officials violated his rights by transferring him to a new unit. He also made general allegations of “corruption, drug smuggling, [and] high rates of violence.” Taylor filed a third action on September 8, 2016, again alleging that corrections officials had improperly transferred him to another unit and confiscated his personal belongings. In a set of three orders issued on the same day, the district court dismissed each complaint for failure to state a claim and so assigned Taylor three “strikes” under § 1915(g).
Taylor timely filed a pro se notice of appeal in each case, again on the same day, and moved in each to proceed in forma pauperis, which Appellees opposed. We consolidated the three cases, provisionally granted Taylor in forma pauperis status, and appointed counsel for him to address the following issue: “Whether trial court dismissal only qualifies as a strike for PLRA purposes if it occurred in a different lawsuit.”
We considered the exact issue presented here in Henslee v. Keller, 681 F.3d 538 (4th Cir. 2012). There, as here, a state prisoner appealed the district court's dismissal of his third qualifying complaint for failure to state a claim and moved to proceed in forma pauperis on appeal. Id. at 539. Adopting the view then widely held by other circuits, we concluded that the prisoner could so proceed because “counting the district court's dismissal as the third strike under [§ 1915(g)] would effectively insulate the dismissal from appellate review.” Id. at 539, 541–42.1
To reach this conclusion, we first found the term “occasion” in § 1915(g) to be ambiguous. Id. at 542. We then turned to the statute's history and purpose, which “indicate[d] that Congress's intention was to limit frivolous prisoner litigation without preventing meritorious claims from being heard.” Id. at 543. Against this backdrop, we observed that denying in forma pauperis status in these circumstances would “freeze out meritorious claims or ossify district court errors by effectively preventing the appellate courts from performing their function.” Id. (internal quotation marks omitted). Because the statute's history and purpose evidenced no intent to do so, we held that “the dismissal of the underlying claim” did not “act as a strike to preclude [in forma pauperis] status on its own appeal.” Id.
Three years after we decided Henslee, the Supreme Court considered a related but factually distinct issue: whether a prisoner whose third strike was pending on appeal could file “several additional actions” in forma pauperis. Coleman, 135 S. Ct. at 1761 (emphasis added). In holding that a prisoner could not do so, the Court focused on when a dismissal becomes final under § 1915(g). Id. at 1763–64. The Coleman Court reasoned that because the term “dismiss ․ does not normally include subsequent appellate activity,” a district court's dismissal of an action immediately counts as a strike under § 1915(g) as to all later-filed additional actions. Id. at 1763.
Resisting this result, the prisoner in Coleman had argued, based in part on a portion of our rationale in Henslee, that the phrase “prior occasions” created ambiguity as to the meaning of when an action or appeal “was dismissed.” Id. The Supreme Court rejected this argument — and thus some of our reasoning in Henslee — and held that nothing in the phrase “ ‘prior occasions’ ․ transform[s] a dismissal into a dismissal-plus-appellate-review.” Id. (quoting 28 U.S.C. § 1915(g)).
Notwithstanding its rejection of the prisoner's claim in Coleman, the Supreme Court found it unnecessary to decide the question we faced in Henslee and we face again today: that is, may a prisoner proceed in forma pauperis on “appeal from the trial court's dismissal of his third complaint”? Id. at 1765. The United States, appearing as amicus curiae for the prison officials, offered an answer of its own: it argued that § 1915(g) was best read to afford a prisoner in forma pauperis status on appeal from the third qualifying dismissal. See Brief for the United States as Amicus Curiae Supporting Respondents at 25–27, Coleman, 135 S. Ct. 1759 (No. 13-1333), 2015 WL 272362 [hereinafter “Brief for the United States as Amicus Curiae”]. After explaining the Solicitor General's position, the Supreme Court expressly left for another day resolution of whether that position was correct. Coleman, 135 S. Ct. at 1765.
We now consider again the issue resolved in Henslee but left open by the Court in Coleman.2 Taylor maintains that we should reach the same conclusion here. While conceding that the Coleman Court left this question “unresolved,” Appellees nonetheless maintain that the Coleman Court's rationale “supports” a different result.
We review de novo questions of statutory interpretation, including application of the three-strikes rule. Tolbert, 635 F.3d at 649.
We begin, as always, with the text of the statute, read in “the specific context in which that language is used, and the broader context of the statute as a whole.” Robinson v. Shell Oil Co., 519 U.S. 337, 341, 117 S.Ct. 843, 136 L.Ed.2d 808 (1997). Where the text is unambiguous, our inquiry is complete. Nat'l Ass'n of Mfrs. v. Dep't of Def., ––– U.S. ––––, 138 S. Ct. 617, 631, 199 L.Ed.2d 501 (2018).
Section 1915 provides that a prisoner may not “bring a civil action or appeal a judgment in a civil action ․ if the prisoner has, on 3 or more prior occasions, while incarcerated ․, brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted.” 28 U.S.C. § 1915(g) (emphasis added).3 We must therefore decide whether the district court's dismissal in this case occurred on a “prior occasion.” If it did, Taylor has three strikes and may not proceed in forma pauperis; if not, we must grant his motions.
In the context of a direct appeal, the ordinary meaning of the term “prior” most naturally encompasses dismissals in other actions, but not in the underlying dismissal that is on appeal. In Coleman itself, both the Sixth Circuit and the Solicitor General found it obvious that a district court decision is not a “prior occasion[ ]” in this context. As the Solicitor General put it, “[t]he phrase ‘prior occasions’ is most sensibly read as referring to strikes imposed in prior-filed suits, not to those imposed in an earlier stage of the same suit.” Brief for the United States as Amicus Curiae at 25; see also Coleman v. Tollefson, 733 F.3d 175, 178 (6th Cir. 2013). Indeed, we and other appellate courts regularly (and intuitively) describe proceedings before the district court as part of “this case,” not as a “prior case.” See, e.g., Sierra Club v. Va. Elec. & Power Co., 903 F.3d 403, 409 (4th Cir. 2018). Thus, although a district court's dismissal of an action is surely an “occasion” from the perspective of the court of appeals, it is not a “prior occasion.” Brief for the United States as Amicus Curiae at 26.
This understanding of the term “prior” in § 1915(g) also comports with “the way in which the law ordinarily treats trial court judgments.” Coleman, 135 S. Ct. at 1764. As the Supreme Court explained in Coleman, “a trial court's judgment” ordinarily “takes effect despite a pending appeal,” and “a judgment's preclusive effect is generally immediate.” Id. (citing Clay v. United States, 537 U.S. 522, 527, 123 S.Ct. 1072, 155 L.Ed.2d 88 (2003)). In Coleman, that fact reinforced the conclusion that, as to the filing of a fourth action, a district court's dismissal — and its assignment of a strike under § 1915(g) — was a final judgment. The opposite is true in the present context: because a district court judgment has no preclusive effect on appeal from the ruling itself, it would be anomalous to treat it as a “prior” occasion under § 1915(g). See Richey, 807 F.3d at 1209 (noting “[w]hile judgments are immediately preclusive as to successive suits, ․ they are certainly not preclusive to the panel on appeal”); cf. 18 Charles Alan Wright et al., Federal Practice and Procedure § 4401 (3d ed. 2008) (“Claim preclusion ․ is not appropriate within a single lawsuit so long as it continues to be managed as a single action.”).4
Moreover, to hold otherwise would be to read out the term “prior” and violate the cardinal rule of statutory construction that we are “obliged to give effect, if possible, to every word Congress used.” Nat'l Ass'n of Mfrs., 138 S. Ct. at 632 (internal quotation marks omitted); see also Carroll v. Logan, 735 F.3d 147, 152 (4th Cir. 2013). To see why this is so, consider how the statute works in practice. Section 1915(g) provides that a prisoner may not “bring a civil action or appeal a judgment in a civil action” in forma pauperis if he has three strikes. Courts therefore count a prisoner's strikes as of the date the prisoner files the complaint or appeal. See, e.g., Millhouse v. Heath, 866 F.3d 152, 154 (3d Cir. 2017); Thomas v. Parker, 672 F.3d 1182, 1183 (10th Cir. 2012). Thus, a strike imposed on any occasion will always be “prior” to a court's decision to grant or deny a motion to proceed in forma pauperis. Had Congress intended to achieve the result Appellees advance, there would be no need to use the word “prior” at all. See Brief for the United States as Amicus Curiae at 25–26. By including it, Congress made clear that an appeals court should count a prisoner's strikes not in the same action, but in prior actions.5
The facts of this case illustrate why our reading of the statute is the correct one. The parties have (rightly) presented this case as requiring us to decide the question left open in Coleman: whether an indigent prisoner may proceed in forma pauperis on appeal from a district court's third strike. But because the district court here dismissed each of Taylor's claims on the same day, Appellees’ logic would prevent Taylor from proceeding in forma pauperis on appeal from any of his three dismissals, not just the third. This is so because on Appellees’ reading, from the perspective of each individual appeal, Taylor has received three qualifying dismissals under § 1915(g). Thus, accepting Appellees’ reading of the statute would mean that a district court's docket management practices — over which a prisoner has no control — could effectively curtail a litigant's right to appeal.6 We do not believe Congress intended a scheme that would produce such an arbitrary result.7
Contrary to Appellees’ suggestion, the Supreme Court's rationale in Coleman does not support a different result.
In making this argument, Appellees first attempt to distort the Court's reasoning to suggest that, despite explicitly reserving the question presented here, the Supreme Court somehow resolved it. We must reject this effort. Coleman held that a dismissal is final for the purpose of counting strikes in a different lawsuit. Coleman, 135 S. Ct. at 1763–64. In so holding, the Supreme Court rejected our reasoning in Henslee as to the ambiguity of the term “occasion.” Id. at 1763. But the Court's reasoning did not focus on the meaning of the term “prior,” nor did the Court determine whether a dismissal in the underlying case on appeal counts as a “prior occasion” under § 1915(g).
Appellees also argue that the Coleman Court's statement that “[t]he in forma pauperis statute repeatedly treats the trial and appellate stages of litigation as distinct” resolves the question before us. Coleman, 135 S. Ct. at 1763. Not so. The Supreme Court made this statement in the context of deciding whether a dismissal becomes final when entered or once it is affirmed on appeal. See id. (“Linguistically speaking, we see nothing about the phrase ‘prior occasions’ that would transform a dismissal into a dismissal-plus-appellate-review.”). That statement says nothing about whether the term “prior” refers to an appeal of the very strike at issue in the case. As we have explained, the text and structure of § 1915(g) make clear that for the term “prior” to have meaning, it must refer to a dismissal in a prior action, not the same action.
If anything, the Supreme Court's rationale in Coleman supports the holding we reach. As noted above, our reading accords with how the law ordinarily treats the preclusive effect of trial court judgments. Moreover, the “leaky filter” problem that the Supreme Court identified in Coleman is not present here. Coleman, 135 S. Ct. at 1764. As the Court then explained, “[a]ppeals take time,” and by delaying the effect of a district court dismissal, “a prisoner could file many lawsuits, including additional lawsuits that are frivolous, malicious, or fail to state a claim upon which relief may be granted.” Id. Indeed, that is precisely what happened in Coleman itself. The same is not true here, for our holding grants Taylor in forma pauperis status only on appeal from his third strike, not in any later-filed suits. Thus, Taylor remains unable to proceed in forma pauperis on any lawsuits filed subsequent to the district court's strike, absent a showing that he qualifies for the imminent danger exception.
Even setting aside the foregoing analysis, our prior decision in Henslee dictates the same result. To be sure, the Coleman Court rejected a portion of our analysis in Henslee when resolving a different question. But much of Henslee’s reasoning was left untouched. And Coleman plainly reserved decision on the issue actually resolved in Henslee and presented again here. Coleman, 135 S. Ct. at 1765.
“We do not lightly presume that the law of the circuit has been overturned,” especially “where, as here, the Supreme Court opinion and our precedent can be read harmoniously.” United States v. White, 670 F.3d 498, 516–17 (4th Cir. 2012) (Duncan, J, concurring). Adhering to our longstanding rule that a panel of this court is “bound by prior precedent from other panels in this circuit absent contrary law from an en banc or Supreme Court decision” demands nothing less. United States v. Ruhe, 191 F.3d 376, 388 (4th Cir. 1999).
We have expressly held that when a Supreme Court decision abrogates one portion of our rationale in a prior case but not another, the rationale not abrogated by the Supreme Court nonetheless binds future panels of this court. See, e.g., United States v. Middleton, 883 F.3d 485, 491 (4th Cir. 2018) (holding that although Supreme Court case “abrogated a portion” of our prior precedent, other parts remained “good law”); United States v. Obey, 790 F.3d 545, 549 (4th Cir. 2015) (explaining that prior decision of this court remained controlling precedent where later Supreme Court decision abrogated it in one application, but left open, and “may well [have] endorse[d],” its application in another); see also United States v. Hill, 776 F.3d 243, 247–49 (4th Cir. 2015) (holding circuit precedent “controls” where Supreme Court did not directly contradict our prior holding). We are bound to apply the same rule here.
It would be especially anomalous to jettison Henslee when such a substantial portion of the reasoning that guided our inquiry in that case applies with equal force today. As we recognized there, denying in forma pauperis status in this context would “effectively eliminate our appellate function” and “freeze out meritorious claims or ossify district court errors.” Henslee, 681 F.3d at 543 (internal quotation marks omitted); see also Richey, 807 F.3d at 1209. We also recognized in passing that a “district court dismissal of [a] ‘plaintiff's present action’ d[oes] not constitute [a] ‘prior action[ ]’ and thus do[es] not count as [a] third strike” on its own appeal. Henslee, 681 F.3d at 543 (emphases added) (quoting Pigg, 106 F.3d at 1498). Then and now, we cannot say that Congress meant “to alter the rule of 28 U.S.C. § 1291, which allows appeals as of right for nearly all civil actions,” without clear textual evidence of its intent to do so. Henslee, 681 F.3d at 543.
For the reasons set forth herein, we join the Ninth and Tenth Circuits to reaffirm that a district court's dismissal of a prisoner's complaint does not, in an appeal of that dismissal, qualify as a “prior” dismissal. Accordingly, Taylor's motions to proceed in forma pauperis are
Plaintiff Therl Taylor filed three lawsuits raising a hodgepodge of claims against state prison officials. To give just three examples, he complained that an employee in the prison mailroom put staples in his Christmas cards, that another employee made poor-quality photocopies for him, and that still other prison officials violated the copyright laws by screening movies for inmates. A federal magistrate judge found his claims meritless; the district court agreed and dismissed all three actions. Taylor now seeks to appeal those decisions in forma pauperis—that is, without paying the required filing fee in full. By statute, however, he cannot appeal in forma pauperis if he has, on three or more “prior occasions,” “brought” a civil action that was dismissed on certain specified grounds. 28 U.S.C. § 1915(g).1 The Defendants oppose his request.
Taylor's request turns on a straightforward issue of statutory interpretation: was each district-court action “brought” on a “prior occasion” from these appeals? The plain language of § 1915(g)’s “three-strikes” rule, as interpreted by the Supreme Court, provides an unambiguous answer: yes. That answer may be harsh, but as judges, we must still apply the statute as Congress wrote it. Therefore, I respectfully dissent.
I begin by explaining why we are not bound by our decision in Henslee v. Keller, 681 F.3d 538 (4th Cir. 2012), which would require us to rule for Taylor if it remained in force. The plaintiff in Henslee had previously filed two actions that were dismissed on one of the grounds enumerated in § 1915(g), which made them strikes. The case on appeal had also been dismissed on one of the enumerated grounds. We therefore had to decide whether this third dismissal counted as a strike and thus precluded the inmate from appealing in forma pauperis. We concluded it did not, based our interpretation of the word “occasions.” We reasoned, first, that “the word ‘occasions’ is ambiguous” because it could mean either “an appeal, independent of the underlying action, or ․ the continuing claim, inclusive of both the action and its appeal.” Id. at 542. We then concluded that the interpretation “which adheres most closely to Congress's dual intentions is one that includes both the appeal and its underlying action.” Id. at 543. Accordingly, we held, a district-court dismissal cannot constitute a strike in an appeal from that dismissal. Id. If this holding were still binding, then Taylor would have only two strikes in each of the three cases on appeal, because one of his three dismissals (the one being appealed in that case) would not count.
In Coleman v. Tollefson, ––– U.S. ––––, 135 S. Ct. 1759, 191 L.Ed.2d 803 (2015), the Supreme Court rejected Henslee’s reasoning while considering different facts. The issue in Coleman was whether a district court dismissal in a separate lawsuit—not the lawsuit being appealed—counts as a strike when it is still pending on appeal. In addressing that issue, the Supreme Court quoted the key passage in Henslee identifying “occasions” as ambiguous. Id. at 1763. It then explained that Henslee was wrong, that “occasions” is not ambiguous, and that a district-court dismissal is a separate “occasion” from a subsequent appeal. Id. at 1763–64. The Supreme Court declined, however, to go beyond the facts before it and address the factual scenario in Henslee and these appeals, where one of the dismissals comes from the very case on appeal. Id. at 1764–65.
Prior circuit precedent is not binding if an intervening Supreme Court case has “specifically rejected the reasoning on which [the prior decision] was based.” Qingyun Li v. Holder, 666 F.3d 147, 150 (4th Cir. 2011) (alteration in original) (quoting Etheridge v. Norfolk & W. Ry. Co., 9 F.3d 1087, 1090–91 (4th Cir. 1993)); see also United States v. Winston, 850 F.3d 677, 683 (4th Cir. 2017) (declining to follow prior panel opinion “undermined by later Supreme Court precedent” (citing United States v. Williams, 155 F.3d 418, 421 (4th Cir. 1998))). This is so even when our prior precedent is directly on point. In Winston, for example, we confronted prior circuit precedent holding that Virginia common law robbery was a “violent felony” under the force clause of the Armed Career Criminal Act. 850 F.3d at 683 (citing United States v. Presley, 52 F.3d 64 (4th Cir. 1995)). In an intervening decision, the Supreme Court had held that Florida criminal battery was not a violent felony and, in doing so, adopted an understanding of the force clause different from the one we had applied. See id. at 683–84 (citing Johnson v. United States, 559 U.S. 133, 130 S.Ct. 1265, 176 L.Ed.2d 1 (2010)). We therefore revisited whether Virginia robbery was a violent felony, unbound by our earlier precedent. Id.
This rule of precedent plays an important role in protecting the rule of law. The modern Supreme Court proceeds mainly by writ of certiorari, choosing to hear only a handful of cases each year. If we continued to be bound by circuit precedent until the Supreme Court reached a different outcome on essentially the same facts, we would keep on multiplying our errors for a long time, sometimes for decades. Thus, whenever the Supreme Court has rejected the reasoning underlying our precedent, we must consider the issue anew, faithfully applying the Court's guidance. Every circuit has adopted some version of this rule, albeit with variations on our “specifically rejected” formulation.2
As a result, we are not bound by Henslee: even though Henslee addressed the exact question before us, Coleman explicitly said that Henslee’s reasoning was wrong. And Taylor is wrong when he argues for a different result just because the Supreme Court declined to address Henslee’s facts. The Supreme Court often declines to address hypothetical cases not before it, for the judicial function is to decide cases, not to render advisory opinions. Sometimes it expressly reserves such hypothetical cases; more typically it simply does not address them. But whether such a reservation is explicit or implicit does not matter when the Supreme Court has rejected our reasoning. Whatever the facts before us, we may not rely on a decision whose reasoning we now know is infirm. Taylor's suggested approach would be dangerous, undermining the rule of law by putting our court above the Supreme Court.
For the first six pages of their analysis, my colleagues in the Majority rightly ignore Henslee and analyze the issue before us de novo. Then, at the end, they contend that Coleman abrogated only part of Henslee’s reasoning and that we remain bound by the rest of it. This argument is a non sequitur. Coleman rejected Henslee’s conclusion that the word “occasions” is ambiguous. All that remains is Henslee’s analysis of how to interpret the text if it is ambiguous. I fail to see how this is relevant to my colleagues’ opinion, because they nowhere conclude that the text is ambiguous. Rather, they claim the statute is “clear.” Majority Op. at 617–18. As best I can tell, the Majority basically agrees with me that Henslee is not binding; my colleagues simply want to rely on certain policy arguments from Henslee. No matter what you think of those policy arguments, they are beside the point here.
I now turn to the statute. The issue before us is whether each action below was “brought” on a “prior occasion” from these appeals. This issue can be broken down into two pieces: (1) Is a district-court action “brought” on a separate “occasion” from a subsequent appeal, or on the same “occasion”? (2) If the action and appeal do in fact represent separate occasions, is the district-court action a “prior” occasion?
The Supreme Court supplied the first piece in Coleman, holding that a district-court dismissal is a separate “occasion” from a subsequent appeal in the same case. It relied on the plain meaning of the word “occasion”: “ ‘a particular occurrence,’ a ‘happening,’ or an ‘incident.’ ” Coleman, 135 S. Ct. at 1763 (quoting dictionary). A district-court action belongs to the same lawsuit as a later appeal, but it is plainly a different “happening” within that lawsuit.
The surrounding text reinforces this conclusion. The statute asks whether the inmate “has, on 3 or more prior occasions, while incarcerated or detained in any facility, brought an action or appeal” that was dismissed for a reason that makes it a strike. 28 U.S.C. § 1915(g) (emphases added). As the Coleman Court observed, the text is punctilious in distinguishing between an “action” and an “appeal,” treating them as two different occasions. See 135 S. Ct. at 1763–64. Just as importantly, the text provides that the relevant “occasion” is the bringing of the action or appeal. A district-court action and an appeal are necessarily “brought” on two different occasions.
With this first piece set out, the second fits smoothly into place. If a district-court action and appeal are different “occasions,” then the district-court action can only be a “prior” occasion. The word “prior” means “[p]receding in time or order.” Prior, Black's Law Dictionary (10th ed. 2014). A district-court action precedes an appeal in both time and order: the appeal cannot proceed until the dismissal is final. 28 U.S.C. § 1291.3 The district-court action is therefore “prior” to the appeal in every conceivable sense of the word.
Thus construed, the statute clearly applies to this case. Taylor has, while incarcerated, “brought” a district-court action on three “occasions.” Each occasion was “prior” to these appeals. And each action “was dismissed” for one of the grounds enumerated in the statute. So he cannot proceed in forma pauperis.
Taylor effectively asks us to create an exception from the statutory text where the third strike is the case on appeal. I cannot accede to that request. The statute does contemplate exceptions, just not the one he seeks. Congress provided an exception to § 1915(g)’s “three-strikes” rule where “the prisoner is under imminent danger of serious physical injury.” 28 U.S.C. § 1915(g). Congress easily could, but did not, create another exception providing that an inmate can appeal from a third strike in forma pauperis. “When Congress provides exceptions in a statute, it does not follow that courts have authority to create others. The proper inference ․ is that Congress considered the issue of exceptions and, in the end, limited the statute to the ones set forth.” United States v. Johnson, 529 U.S. 53, 58, 120 S.Ct. 1114, 146 L.Ed.2d 39 (2000).
Undeterred by the lack of a statutory exception supporting Taylor's position, my colleagues write one in. They admit that a district-court action is a “prior occasion” when viewed from the perspective of an appeals court in a different case—Coleman compels that result. But they insist that the phrase “prior occasion” means something different from the perspective of an appellate court in the same case.
I am not entirely sure how this interpretation works. At times, my colleagues suggest that the district-court action and the subsequent appeal are one and the same “occasion.” They point out that we often “describe proceedings before the district court as part of ‘this case,’ not as a ‘prior case.’ ” Majority Op. at 616–17. But the statute doesn't say “case.” It says “occasion.” This attempt to create slippage between “occasion” and “case” ultimately gets the Majority nowhere, because as Coleman held, a district-court action and a subsequent appeal are different “occasions” within the same case. Understandably reluctant to contradict Coleman, my colleagues never explicitly say that a district-court dismissal is the same “occasion” as an ensuing appeal.
Instead, my colleagues try to use the word “prior” to shoehorn their unwritten exception into the statute. A district-court action is a separate occasion, they seem to say, just not a prior occasion from where we sit. To accept this, one would have to posit that the district-court action occurs simultaneously with our review on appeal. But while both we and the district court review the same issues, we do not review them simultaneously: the district court handles the case first, and only once its proceedings are final do we check its work. And we know from Coleman that these two occasions are not simultaneous from the perspective of a different court (or even a different panel of the same court) viewing the same proceedings. I see nothing in the statute to support the idea that two separate occasions can occur at different points in time from the perspective of one panel, but at the same time from the perspective of another.4
Hoping to bolster their interpretation, my colleagues rely on the canon against surplusage. “Prior” is largely redundant in the Defendants’ interpretation of the statute.5 We can give “prior” meaning, so the argument goes, by reading it to say that the dismissal must occur in a “prior” lawsuit. See Brief for the United States as Amicus Curiae at 25–27, Coleman v. Tollefson, 135 S. Ct. 1759 (No. 13-1333), 2015 WL 272362.
Yet the “canon against surplusage is not an absolute rule.” Marx v. Gen. Revenue Corp., 568 U.S. 371, 385, 133 S.Ct. 1166, 185 L.Ed.2d 242 (2013). Many statutes use redundant words: “have and hold,” 16 U.S.C. § 445a, “claim or cause of action,” 28 U.S.C. § 1452, “release and discharge,” 30 U.S.C. § 187a, “perfectly null and void,” 43 U.S.C. § 859—and so on and so forth. As a result, “there are instances in which a court may validly ‘prefer ordinary meaning to an unusual meaning that will avoid surplusage.’ ” Oak Grove Res., LLC v. Director, OWCP, 920 F.3d 1283, 1291 (11th Cir. 2019) (quoting Antonin Scalia & Bryan A. Garner, Reading Law § 26, at 176 (2012)). Certainly, a court should not give a word an entirely fanciful meaning to avoid a minor redundancy.
Here, the surplusage argument distorts “prior” beyond all recognition, assigning that word a meaning it lacks. The word “prior” does not even remotely suggest that a district-court dismissal must have occurred in a different lawsuit to count as a strike. Indeed, this argument is nothing more than an end-run around Coleman: it tries to use the word “prior” to show that the relevant “occasion” is the entire lawsuit (that is, the district-court action and subsequent appeal together), the very interpretation that Coleman rejected.
Instead of misreading “prior,” we should just admit that Congress used a redundant word. That is hardly surprising: redundant uses of “prior” abound. See Redundancies, Grammarist, https://grammarist.com/redundancies/ (last visited July 17, 2019) (noting that the oft-used phrase “prior experience” is usually redundant because almost all experience is “prior”). Even judicial opinions sometimes fall victim, for example by referring to someone's “prior history,” a phrase that is almost always redundant because history necessarily takes place in the past. United States v. Thomas, 669 F.3d 421, 427 (4th Cir. 2012); see also Diamond v. Colonial Life & Accident Ins. Co., 416 F.3d 310, 319 (4th Cir. 2005) (referring to job applicants’ “prior management experience”). Similarly, courts often refer to “prior precedent” even though precedent is, quite literally, that which “precedes.” E.g., United States v. Hargrove, 478 F.3d 195, 204 (4th Cir. 2007).6 The only reasonable conclusion is that Congress committed a similar redundancy here.
The Majority also suggests that we should construe § 1915(g) narrowly to avoid restrictions on our appellate review. This suggestion has no basis in law. True, there is a substantive canon requiring an express statement to oust courts of appellate jurisdiction. See, e.g., United States v. Am. Bell Tel. Co., 159 U.S. 548, 549–50, 16 S.Ct. 69, 40 L.Ed. 255 (1895). But see Scalia & Garner, supra, § 65, at 367–68 (criticizing this rule). Yet this case is not about our appellate jurisdiction, just Taylor's ability to proceed without paying the filing fee in advance. See Blakely v. Wards, 738 F.3d 607, 612 (4th Cir. 2013) (en banc) (“Section 1915(g) is a limitation on prisoners’ rights, not on courts’ authority.”); Lisenby v. Lear, 674 F.3d 259, 263 (4th Cir. 2012) (holding § 1915(g) is “procedural, not jurisdictional, in nature”). There is no recognized canon of construction requiring us to narrowly construe Congress's restrictions on in forma pauperis appeals. See United States v. Kras, 409 U.S. 434, 450, 93 S.Ct. 631, 34 L.Ed.2d 626 (1973) (explaining that there is no “unlimited rule that an indigent at all times and in all cases has the right to relief without the payment of fees”). Any such rule would be inconsistent with Coleman, which did not read § 1915(g) narrowly but rather followed the plain and natural meaning of the text.
In addition, the Majority reasons, this case can be distinguished from Coleman based on “the way in which the law ordinarily treats trial court judgments.” 135 S. Ct. at 1764. A trial-court judgment has preclusive effect on other actions immediately, before it is appealed. By contrast, a trial-court judgment has no preclusive effect on an appeal from that very judgment. Yet this distinction makes no difference. This point was an afterthought in the Coleman Court's analysis, serving merely to confirm its “literal reading” of the statute. Id. Courts often gild the lily in this way, providing policy arguments to confirm a statutory interpretation compelled by the text alone. The simple fact is that nothing in Coleman’s textual reading of § 1915(g) turned on issues of finality and preclusion, because the text says nothing about those issues. So this distinction gives me no reason to change my interpretation of the statute. Even on its own terms, my colleagues’ proposed distinction fails, because it ignores another relevant sense of “finality”: the one provided by 28 U.S.C. § 1291. Each dismissal below was “final” for purposes of appeal, and each action “was dismissed” unless and until we vacate or reverse the district court's order.
Similarly, while this particular case does not implicate Coleman’s concern that § 1915(g) might turn into a “leaky filter,” that provides no reason to deviate from statutory text. While putting a big gash in a filter can make it leaky, so can punching out a bunch of small holes. Thus, this proposed distinction does not hold water: if it did, then judges could riddle the statute with unauthorized exceptions that let through one meritless prisoner lawsuit after another.
Finally, my colleagues “do not believe Congress intended a scheme that would produce such an arbitrary result” as the one in this case. Majority Op. at 618. While I sympathize with my colleagues’ concern that the statute commands a harsh result, I do not see that as a reason to rewrite it.
Consider what Congress was trying to accomplish here. The in forma pauperis statute seeks to balance two competing goals: on the one hand, affording poor people easier access to the courts, and on the other, shielding the courts from the never-ending deluge of meritless prisoner litigation. Congress is uniquely positioned to balance open-ended policy goals of this type, and our role as judges is to discern the balance Congress struck by faithfully interpreting the statutory text. Still, Congress could have left much of the balancing to us, adopting a loose standard with broad room for judicial discretion. Instead, it gave us concrete rules. One of them—designed to address the problem of “serial filers”—is the “three-strikes” rule of § 1915(g).
Rules of this sort are invariably over- and under-inclusive, appearing harsh in some cases and lenient in others. As a result, my colleagues’ concern to avoid an “arbitrary” interpretation misses the point. A “three-strikes” rule is arbitrary almost by definition. Why not four strikes? Or five? Or fifteen? The answer may simply be that many members of Congress are baseball fans. From our perspective, it does not matter. Congress decided to draw the line at three, and our sole job is to count. When we do, it becomes clear that Taylor already has three strikes and may not proceed without paying the filing fee. And even under my colleagues’ interpretation, Taylor would be barred from proceeding in forma pauperis if there were four district-court actions on appeal.7 Somehow, three is unacceptably “arbitrary,” but four is just fine.
The real force of my colleagues’ reasoning is that the statute as written seems too harsh: inmates will have to pay the filing fee, which many of them cannot afford, to get appellate review of whether their third strike was proper. I too have sympathy for inmates in that position. Indeed, the statute is particularly harsh here, because all three dismissals were handed down on the same day, leaving Taylor unable to appeal any of them in forma pauperis. But the statute says what it says, and it is Congress's job—not ours—to rewrite it if it produces unduly harsh results. Taylor asks us to do just that, and my good colleagues in the Majority indulge that request. I would follow the statute and require Taylor to pay the full filing fee. For that reason, I respectfully dissent.
1. Appellees agree that at that time, “there was no significant circuit split” on this question. See, e.g., Henslee, 681 F.3d at 541–42; Thompson v. DEA, 492 F.3d 428, 432–33 (D.C. Cir. 2007); Pigg v. FBI, 106 F.3d 1497, 1498 (10th Cir. 1997) (per curiam); Adepegba v. Hammons, 103 F.3d 383, 387–88 (5th Cir. 1996). Even the Seventh Circuit, which had interpreted the statute in a way that appeared to bar an in forma pauperis appeal of a prisoner's third strike, allowed such an appeal through a motion under Fed. R. App. P. 24(a)(5). See Robinson v. Powell, 297 F.3d 540, 541 (7th Cir. 2002).
2. Since the issuance of Coleman, two circuits have adhered to the prior consensus view that a prisoner may pursue in forma pauperis status on appeal of a third strike. See Richey v. Dahne, 807 F.3d 1202, 1209 (9th Cir. 2015); Dawson v. Coffman, 651 F. App'x 840, 842 n.2 (10th Cir. 2016). Only one has reached a contrary result. Parker v. Montgomery Cty. Corr. Facility/Bus. Office Manager, 870 F.3d 144, 152–53 (3d Cir. 2017).
3. The parties agree that the exception for cases in which “the prisoner is under imminent danger of serious physical injury,” 28 U.S.C. § 1915(g), does not apply here.
4. The dissent suggests that the Coleman Court's reliance on preclusion law “was an afterthought.” We are not free to make such a judgment about the reasoning of a unanimous opinion of the Supreme Court. Even if we could, we must disagree with our colleague's characterization. As in Coleman, we hold only that “[o]ur literal reading of” § 1915(g) is reinforced by black-letter preclusion law. 135 S. Ct. at 1764.
5. In holding to the contrary, the Third Circuit reasoned that “the term ‘prior’ sets a temporal parameter” that instructs courts to count only “strikes accrued earlier in time than the notice of appeal,” excluding “later-accrued strikes.” See Parker, 870 F.3d at 153. But the first clause of § 1915(g), which provides that a prisoner may not “bring a civil action or appeal a judgment in a civil action,” accomplishes this very same goal. Rather than clarifying a rule already set by another part of § 1915(g), the term “prior” more naturally refers to strikes imposed in a prior action, not the same action.
6. For example, imagine that the district court had dismissed each of Taylor's complaints one week apart, and Taylor had noted a timely appeal in his first two dismissals before receiving his third. Because we decide whether to grant in forma pauperis status as of the date the appeal is filed, see supra at 617–18, Appellees’ logic would dictate that we grant Taylor's motions to proceed in forma pauperis in the first two cases but not the third. Yet because the district court in this case dismissed each of Taylor's complaints on the same day and thus prevented him from noting an appeal in any of his cases until all three were dismissed, Appellees would grant Taylor in forma pauperis appeal in none of his three cases. That simply cannot be the rule Congress intended.Of course, we do not doubt the district court's broad authority to manage its own docket. The point is simply that those choices should not affect the neutral operation of § 1915(g). The most sensible result, and the one consistent with the language of the statute, would permit a litigant to proceed in forma pauperis in all three cases in both examples cited above, provided he or she has no other qualifying dismissals.
7. Contrary to our dissenting colleague's suggestion, we doubt whether Congress intended such an arbitrary result — one in which a litigant's appellate rights effectively turn on how a district court manages its docket — not whether that result is a “harsh” one. The latter question surely rests with Congress; the former counsels in favor of our reading of the term “prior” in § 1915(g).
1. 28 U.S.C. § 1915(g):In no event shall a prisoner bring a civil action or appeal a judgment in a civil action or proceeding under this section if the prisoner has, on 3 or more prior occasions, while incarcerated or detained in any facility, brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted, unless the prisoner is under imminent danger of serious physical injury.
2. See, e.g., Mongeau v. City of Marlborough, 492 F.3d 14, 18 (1st Cir. 2007) (“undermined”); United States v. Robbins, 729 F.3d 131, 134 (2d Cir. 2013) (“casts doubt”); United States v. Tann, 577 F.3d 533, 541 (3d Cir. 2009) (“conflicts with”); Martin v. Medtronic, Inc., 254 F.3d 573, 577 (5th Cir. 2001) (“implicitly overrules”); United States v. White, 920 F.3d 1109, 1113 (6th Cir. 2019) (“undercut or abrogated”); United States v. Walton, 255 F.3d 437, 443 (7th Cir. 2001) (“undermined”); Northport Health Servs. of Arkansas, LLC v. Rutherford, 605 F.3d 483, 489 (8th Cir. 2010) (“undermines”); Miller v. Gammie, 335 F.3d 889, 900 (9th Cir. 2003) (en banc) (“undercut”); Auraria Student Hous. at the Regency, LLC v. Campus Vill. Apartments, LLC, 843 F.3d 1225, 1235 (10th Cir. 2016) (“undermines”); United States v. Madden, 733 F.3d 1314, 1319 (11th Cir. 2013) (“substantially undermined”); Davis v. U.S. Sentencing Comm'n, 716 F.3d 660, 664 (D.C. Cir. 2013) (“eviscerated”); Troy v. Samson Mfg. Corp., 758 F.3d 1322, 1326 (Fed. Cir. 2014) (“undercut”). What is true of our own precedent is not true of the Supreme Court's: we follow on-point Supreme Court precedent even when its reasoning has been undermined by a later Supreme Court decision, waiting for the Supreme Court to overrule itself. See, e.g., United States v. Sterling, 724 F.3d 482, 501–02 (4th Cir. 2013) (citing Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989)). This difference reflects the Supreme Court's supremacy over our court.
3. What about interlocutory appeals? We need not concern ourselves with them, for all “strikes” are final orders: the text requires a dismissal of the entire “action” for one of the specified grounds, which is necessarily “final” for purposes of § 1291.
4. According to Einstein's special theory of relativity, two events separated in space can occur at the same time from the perspective of one observer, but at different times from the perspective of an observer in another inertial reference frame. Einstein was right, and many experiments have borne out his predictions. But these relativistic effects are negligible at everyday speeds, so it is hard to think that Congress legislated with them in mind. After all, the Ninth Circuit doesn't observe our court's proceedings from a spaceship zooming past us at half the speed of light. Litigation just doesn't move that fast. Cf. Texas Nat'l Bank v. Sandia Mortg. Corp., 872 F.2d 692, 700 n.4 (5th Cir. 1989) (explaining that the “judicial system is sufficiently lethargic so as to minimize the inaccuracies of Newtonian Mechanics”).
5. I agree with the Majority that the statute provides, independently of the word prior, that each dismissal must occur before the appeal is filed. The statute forbids an inmate to “bring a civil action or appeal a judgment in a civil action or proceeding under this section” if each of three prior actions or appeals “was dismissed” for one of the specified grounds. 28 U.S.C. § 1915(g) (emphases added). In context, the verb “appeal” is most naturally read to refer to the filing of the appeal. So when the statute uses the past tense (“was dismissed”), it tells us that each dismissal must come before filing.
6. Some readers may have noticed that page three of this opinion, above, contains at least four different variations on “prior precedent.” But I suspect many missed it, precisely because this minor redundancy is unremarkable. Those who did notice surely did not conclude that “prior” must have some alternative meaning.
7. This is a consequence of the Supreme Court's holding in Coleman. Suppose that four dismissals had become final on the same day, and that Taylor appealed all four of them the following day. In each appeal, the dismissals in the other three cases would immediately count as strikes under Coleman. The Majority's holding would save only one of the four dismissals from counting as a strike in each appeal.
DIANA GRIBBON MOTZ, Circuit Judge:
Response sent, thank you
Docket No: No. 17-6374, No. 17-6375, No. 17-6376
Decided: July 18, 2019
Court: United States Court of Appeals, Fourth Circuit.
Search our directory by legal issue
Enter information in one or both fields (Required)
FindLaw for Legal Professionals
Search our directory by legal issue
Enter information in one or both fields (Required)