dissenting:
More than thirty-five years ago, the Supreme Court recognized that it is “established beyond doubt that prisoners have a constitutional right of access to the courts.” Bounds v. Smith, 430 U.S. 817, 821, 97 S.Ct. 1491, 52 L.Ed.2d 72 (1977). Certainly, that right is not without limits. The three-strikes rule imposes a notable limit on a prisoner’s ability to proceed in forma pauperis (“IFP”). 28 U.S.C. § 1915(g). But this limit must have its own limits and Congress recognized as much. Only by disregarding the clear statutory language of § 1915(g) can the majority hold that a grant of summary judgment constitutes a “dismissal” for purposes of the statute. In doing so, the majority improperly restricts access to the courts well beyond Congress’ intent. With respect, I dissent.
I.
The theory offered by the majority for its holding is belied by the unambiguous language of § 1915(g) itself and the Supreme Court’s instruction as to proper statutory construction.
*625Section 1915(g) expressly provides that a prisoner may not proceed IFP “if the prisoner has, on 3 or more prior occasions, while incarcerated or detained in any facility, brought an action or appeal ... that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted.” Id. (emphasis added). The plain language of § 1915(g) thus states that only cases that are “dismissed” can constitute strikes. This language makes clear that cases resolved by a procedure other than dismissal — like summary judgment, as in this case — are not strikes.
Time and again, the Supreme Court has directed us to defer to “the language employed by Congress” and adopt “the assumption that the ordinary meaning of that language accurately expresses the legislative purpose.” Engine Mfrs. Ass’n v. S. Coast Air Quality Mgmt. Dist., 541 U.S. 246, 252, 124 S.Ct. 1756, 158 L.Ed.2d 529 (2004) (internal quotation marks and citation omitted). Staying true to that directive requires rejection of the majority’s theory.
In Tolbert v. Stevenson, 635 F.3d 646, 654 (4th Cir.2011), a unanimous panel of this court correctly recognized as much, holding that “a grant of summary judgment ... is not one of the grounds listed in § 1915(g), and therefore ... does not count as a strike.” Our Tolbert rule comports with the position of all other courts of appeals to have reached the question. As the District of Columbia Circuit has explained, “if the court dismisses an unex-hausted complaint on a Rule 12(b)(6) motion or if it dismisses the complaint sua sponte and expressly declares that the complaint fails to state a claim, the dismissal counts as a strike.” Thompson v. DEA 492 F.3d 428, 438 (D.C.Cir.2007). “But if the court dismisses the complaint on some other procedural mechanism, such as ... a motion for summary judgment, the dismissal will not count as a strike.” Id.; accord Taylor v. First Med. Mgmt., 508 Fed.Appx. 488, 494 (6th Cir.2012); Stallings v. Kempker, 109 Fed.Appx. 832, 832-33 (8th Cir.2004) (per curiam); Angelle v. Gibson, No. 00-50675, 2001 WL 498763, at *1 (5th Cir. Apr. 12, 2001) (per curiam).
Indeed, that a case resolved on summary judgment is not “dismissed” and thus does not constitute a strike for § 1915(g) purposes is so obvious that at least three circuits, and numerous district courts, simply have assumed as much in unpublished opinions. See Taylor, 508 Fed.Appx. at 494 (“The plain language seemingly limits the application of a strike to dismissals by only speaking of dismissals.”); Stallings, 109 Fed.Appx. at 832-33 (“Because the district court resolved the case through summary judgment, the dismissal does not constitute a ‘strike’.... ”); Angelle, No. 00-50675, 2001 WL 498763, at *1 (“Because the district court’s dismissal for failure to state a claim acted as a grant of summary judgment, the district court’s judgment does not count as a ‘strike.’ ”); see also A Jailhouse Lawyer’s Manual ch. 14 n. 85 (8th ed.2009) (collecting cases).
Regrettably, today the majority rejects the rule adopted in Tolbert and by our sister circuits and concludes that cases resolved on summary judgment count as strikes. Neither of the rationales offered for doing so is persuasive.
A.
The majority’s initial — and extraordinary — rationale for this theory is that the plain language of § 1915(g) somehow permits a grant of summary judgment to count as a strike. The majority relies on the fact that in “common usage,” disposition on summary judgment is sometimes referred to as “dismissal” and that the dictionary definition of “dismiss” is “to ter-*626mínate (an action or claim) without further hearing, esp. before the trial of the issues involved.” Black’s Law Dictionary 482 (7th ed.1999). But neither imprecise common usage nor an overbroad dictionary definition can eliminate a lesson learned in the first year of law school: dismissal and summary judgment differ in important respects. Both can terminate an action, but a case resolved by summary judgment is not “dismissed.”
In the very context of the three-strikes rule, we have emphasized that “[w]hen Congress directly incorporates language with an established legal meaning into a statute, we may infer that Congress intended the language to take on its established meaning.” McLean v. United States, 566 F.3d 391, 396 (4th Cir.2009); accord Miles v. Apex Marine Corp., 498 U.S. 19, 32, 111 S.Ct. 317, 112 L.Ed.2d 275 (1990) (“We assume that Congress is aware of existing law when it passes legislation.”).
The word “dismissed” in § 1915(g), particularly “coupled with the words ‘[for] failure] to state a claim upon which relief may be granted,’ ... has a well-established legal meaning.” See McLean, 566 F.3d at 396. And that well-established legal meaning obviously differs from the equally well-established legal meaning of summary judgment. Compare Fed. R.Civ.P. 56 (defining the distinct basis for summary judgment) with Fed.R.Civ.P. 12(b)(6) (listing grounds for dismissal including failure to state a claim upon which relief can be granted) and 28 U.S.C. § 1915(e)(2)(listing grounds for dismissal including frivolity, maliciousness, and failure to state a claim); id. § 1915A(b) (same); 42 U.S.C. § 1997e(c) (same).1
I find perplexing the majority’s repeated assertion that a dismissal and a grant of summary judgment differ only in their “procedural posture.” See ante at 609, 612, 614. Even if this were so, this is not an insignificant difference. Rather, “procedural posture” may make all the difference. See, e.g., Lucas v. S.C. Coastal Council, 505 U.S. 1003, 1013 n. 3, 112 S.Ct. 2886, 120 L.Ed.2d 798 (1992). In any event, a dismissal and a grant of summary judgment differ from each other in far more than “procedural posture.” See Bradley Scott. Shannon, A Summary Judgment Is Not a Dismissal!, 56 Drake L.Rev. 1, 7 (2007) (summarizing differences in moving party, timing of motion, ability to waive, determining propriety of jurisdiction, nature of relevant evidence, appealability, preclusive effect, etc.). And, contrary to the majority’s suggestion, these differences are well recognized. See, e.g., Lujan v. Nat’l Wildlife Fed’n, 497 U.S. 871, 889, 110 S.Ct. 3177, 111 L.Ed.2d 695 (1990); Celotex Corp. v. Catrett, 477 U.S. 317, 327, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986).
To be sure, if Congress had wanted § 1915(g) to cover more than actual dismissals, it could have said so. The fact that it did not speaks volumes, as we and other courts have previously explained. See, e.g., Green v. Young, 454 F.3d 405, 409 (4th Cir.2006) (“The ... three-strikes provision by its terms applies only if a prisoner has had three prior actions dismissed as ‘frivolous, malicious, or [for] failfure] to state a claim.... ’ Because a dismissal for *627failure to exhaust is not listed in 1915(g), it would be improper for us to read it into the statute.”); Butler v. Dep’t of Justice, 492 F.3d 440, 444 (D.C.Cir.2007) (“Had Congress wanted to include dismissals for failure to prosecute among the strikes listed in § 1915(g), it could have done so. If we were to adopt the government’s approach, we would be effectively writing another category of strikes into the [statute]. We have neither the authority nor inclination to substitute our policy judgment for that of Congress.”) (internal citation omitted).
In sum, the plain language of § 1915(g) most certainly does not permit a court to treat a case resolved by summary judgment as “dismissed” and so count it as a strike for purposes of § 1915(g).
B.
The majority also attempts to rely on Congress’ “intent” in drafting the three-strikes rule. It maintains that there is no indication Congress had in mind at the time the “academic” distinction between dismissal and summary judgment. It emphasizes that an overriding purpose of the Prison Litigation Reform Act (“PLRA”), Pub.L. No. 104-134, 110 Stat. 1321 (1996), which established the three-strikes rule , in § 1915(g), was to limit frivolous prisoner suits as much as possible. See, e.g., 141 Cong. Rec. S14,418 (daily ed. Sept. 27, 1995) (statement of Sen. Orrin Hatch) (“Our legislation ... addresses the flood of frivolous lawsuits brought by inmates.”).
The majority ignores the fact, however, that reducing frivolous suits was but a means to an end: by enacting the PLRA, Congress intended most fundamentally to reduce the burden on overworked courts. See id. (“The crushing burden of these frivolous suits makes it difficult for courts to consider meritorious claims.”); 141 Cong. Rec. S7526 (daily ed. May 25, 1995) (statement of Sen. Jon Kyi) (noting that frivolous suits were “draining precious judicial resources” and that § 1915(g) would “free up judicial resources for claims with merit by both prisoners and nonprison-ers”); . id. at S7524 (statement of Sen. Robert Dole) (“Frivolous lawsuits ... waste valuable judicial and legal resources, and affect the quality of justice enjoyed by the law-abiding population.”).
Counting summary judgments as strikes is wholly.out-of-step with this intent. Doing so would require courts to engage in a time-intensive, individualized inquiry to determine whether, in each of a plaintiffs prior cases, a court had granted summary judgment on the basis of a specific statutory criterion not required for the grant of summary judgment, i.e., frivolousness, maliciousness, or failure to state a claim. In Tolbert, we emphasized this very point, noting that “[t]o require district courts to so parse summary judgment orders and their supporting documents would even further increase their workload, again straying far afield of the purpose of the PLRA.” 635 F.3d at 653 n. 7.
Other courts, too, have recognized the need for bright-line rules for identifying strikes to avoid increasing courts’ workloads in contravention of the PLRA’s purpose. See, e.g., Byrd v. Shannon, 715 F.3d 117, 125 (3d Cir.2013) (stating that rules “reducing litigation on whether a particular dismissal constitutes a strike” serve the PLRA’s overriding purpose); Thompson, 492 F.3d at 438 (“In addition to our obligation to adhere to section 1915(g)’s text, we are mindful that a driving purpose of the PLRA is to preserve the resources of both the courts and the defendants in prisoner litigation. Here, all agree that purpose is best accomplished by a bright-line rule that avoids the need to relitigate past cases.”).
In an attempt to avoid the reality that their new rule will require this time-inten*628sive parsing — wreaking the precise waste of judicial resources that Congress sought to avoid — the majority offers a limitation on this new rule. My colleagues would treat a case disposed of by summary judgment as “dismissed” for purposes of § 1915(g) only when the district court has made explicit that it believed the case was frivolous, malicious, or failed to state a claim. This asserted limitation, however, runs into a different problem. It requires an appellate court inappropriately to acquiesce in a district court’s determination of an issue that the parties may not have had an opportunity to address and that is totally unnecessary to the court’s grant of summary judgment. See Fed.R.Civ.P. 56 (permitting summary judgment on the basis that “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law”). The majority’s view thus forces appellate courts into an untenable catch-22: rubber-stamp district court decisions on issues not strictly before them or expend time and energy to decide these issues independently-
Limiting strikes to actual dismissals avoids this conundrum. The reviewing court would not- need to bind itself to statements a lower court may have made in passing and without briefing by the parties. Moreover, a reviewing court would not need to engage in a searching inquiry of the district court’s decision to decide the issues of frivolity or maliciousness for itself. Thus, the straightforward rule that a case resolved on summary judgment is never “dismissed” for purposes of § 1915(g) is both dictated by the statutory language and the only workable rule consistent with the statutory purpose.
II.
Perhaps recognizing the defects in the majority’s rationale, a group of my colleagues join a long concurrence offering a second theory for denying Blakely IFP status. The concurrence, of course, is only dicta, without precedential effect, because á majority of the court does not embrace its theory. There is good reason for this.2
The concurrence contends that 28 U.S.C. § 1915(a)(1) permits a court to deny IFP status to a prisoner who does not have three strikes, but who has, in the court’s view, previously abused the IFP system, whatever the merits of his current case. This assertion — that we have broad discretion to deny IFP status to a prisoner who has not had three cases dismissed, on the basis of other aspects of his filing history — is deeply flawed. Use of such discretion would defy the clear and unambiguous legislative history and the very purpose of the PLRA’s three-strikes scheme, a result Congress, surely did not intend.
Under 28 U.S.C. § 1915(a)(1), “any court of the United States may authorize the commencement, prosecution or defense of any suit, action or proceeding, civil or criminal, or appeal therein, without prepayment of fees.” The concurrence posits that this language — that a court “may authorize” a party to proceed IFP — also implies an inherent authority not to authorize a prisoner to proceed IFP if he has ever previously abused the IFP right, regardless of whether the prisoner has three strikes.
Try as it might, the concurrence cannot reconcile its theory with 'the telling legislative history of § 1915(a)(1) and § 1915(g). Nearly half a century after the general grant of discretionary authority in § 1915(a)(1) had been on the books, and *629more than a century after its predecessor had, Congress established a more specific scheme for limiting prisoners’ abuse of the IFP system: the three-strikes rule of § 1915(g). See 62 Stat. 954 (1948); 27 Stat. 252 (1892); see also Rivera v. Allin, 144 F.3d 719, 722 (11th Cir.1998) (noting that “[o]n April 26, 1996, [§ 1915(a)] changed when the President signed into law the PLRA”), abrogated on other grounds by Jones v. Bock, 549 U.S. 199, 127 S.Ct. 910, 166 L.Ed.2d 798 (2007).
It is “a commonplace of statutory construction that the specific governs the general,” and this command applies with particular force where, as here, the general clause (§ 1915(a)(1)) is a “relic” of an earlier (pre-§ 1915(g)) reign. Morales v. Trans World Airlines, Inc., 504 U.S. 374, 384, 385, 112 S.Ct. 2031, 119 L.Ed.2d 157 (1992). As the Supreme Court explained in FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 143, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000): ;
The classic judicial task of reconciling many laws enacted over time ... necessarily assumes that the implications of a statute may be altered by the implications of a later statute. This is particularly so where the scope of the earlier statute is broad but the subsequent statute!] more specifically address[es] the topic at hand.... [A] specific policy embodied in a later federal statute should control our construction of the [earlier] statute, even though it ha[s] not been expressly amended.
(emphasis added) (internal citation and quotation marks omitted). The coneur-rence simply ignores these long-established principles. Undoubtedly, this is because, if properly applied, those principles lead to but one conclusion: the PLRA’s specific three-strikes scheme for abusive prisoner-litigants must govern any more general authority that might otherwise apply.3
Although the concurrence protests to the contrary, its theory also cannot be reconciled with another long-established statutory construction principle: a court must read statutory provisions in light of the whole statute and the objects and policy of that statute. See, e.g., id. at 133, 120 S.Ct. 1291 (“It is a fundamental canon of statutory construction that the words of a statute must be read in .their context and with a view to their place in the overall statutory scheme.” (internal quotation marks omitted)); Babbitt v. Sweet Home Chapter of Cmtys. for a Great Or., 515 U.S. 687, 698-700, 115 S.Ct. 2407, 132 L.Ed.2d 597 (1995) (emphasizing that we must read a statute in light of its underlying purpose). >
Congress enacted the three-strikes statute to eliminate waste of judicial resources by setting forth a single, clear rule for denying IFP status to abusive prisoner-litigants. This rule burdens judicial resources far less than the two-tiered system advocated by the concurrence, for a two-tiered system requires courts to assess both the number of strikes and whether there is some other discretionary reason for denying IFP status. The simpler rule is preferable especially given the *630amorphous nature of the multi-factor test propounded by the concurrence for determining when an exercise of discretionary authority is appropriate. This multi-fac-tored test would only “further increase [courts’] workload, ... straying far afield of the purpose of the PLRA.” See Tolbert, 685 F.3d at 653 n. 7. It defies reason to engage in such a time-consuming inquiry — in the name of judicial economy — simply to avoid reaching the merits.4
Finally, contrary to the concurrence’s suggestion, Supreme Court authority hardly supports its view that the lower courts have discretion to deny IFP status to a prisoner who has not accumulated three strikes but has, in the court’s view, previously abused the IFP system, regardless of the merit of his current case. First, the Supreme Court does not invoke its own discretionary authority to deny IFP status prospectively without first determining that the petitioner’s present case is frivolous. See In re Amendment to Rule 39, 500 U.S. 13, 14, 111 S.Ct. 1572, 114 L.Ed.2d 15 (1991) (per curiam). Thus, the Supreme Court rule does not countenance, let alone support, denial of IFP status to Blakely here solely on the basis of his previous cases. Moreover, the Supreme Court has never denied anyone the right to appeal IFP, i.e., to be granted review on the merits as of right. Rather, in every case — including those relied on by the concurrence — it has applied this discretionary authority only when a litigant petitions for relief by writ. Obviously, this is a far more limited burden on access to the courts. Finally, the Supreme Court has never suggested that the courts of appeals should exercise discretionary authority to deny IFP status to prisoners appealing as of right. Accordingly, Supreme Court precedent offers precious little support for the concurrence’s expansive view of the discretion granted in § 1915(a)(1).
In sum, the concurrence refuses to recognize that the PLRA’s specific three-strikes rule displaced any general discretionary authority set forth in § 1915(a)(1) with respect to prisoner “action[s] or appeals],” 28 U.S.C. § 1915(g). Instead it attempts to extend general discretionary authority well beyond anywhere the Supreme Court has taken it.
III.
By ignoring the plain language limiting § 1915(g) to dismissals, (and in some cases reading § 1915(a)(1) to swallow § 1915(g)), my colleagues defy the express will of Congress. Worse yet, in doing so, they undermine the most fundamental promise of our legal system; equal access to justice. Their theories fly in the face of our obligation to construe narrowly any limitation on a litigant’s constitutional right of access to the courts. See, e.g., Chambers v. Balt. & Ohio R.R. Co., 207 U.S. 142, 148, 28 S.Ct. 34, 52 L.Ed. 143 (1907) (“The right to sue and defend in courts is ... the right conservative of all other rights, and lies at the foundation of orderly government.”); accord Cromer v. Kraft Foods N. Am., Inc., 390 F.3d 812, 817-18 (4th Cir.2004).
As the Supreme Court recognized in a case on which the concurrence itself relies, “[p]aupers have been an important — and valued — part of the Court’s docket, see, *631e.g., Gideon v. Wainwright, 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963), and remain so.” In re McDonald, 489 U.S. 180, 184, 109 S.Ct. 993, 103 L.Ed.2d 158 (1989) (per curiam). Regrettably, my colleagues disregard this value. By denying Blakely leave to proceed IFP before even glancing at the merits of his current claim, the majority of the court improperly denies an indigent access to justice, and sets a dangerous course for the future.
I respectfully dissent. Judges King, Davis, and Thacker join in this dissent; Judge Gregory joins in Part I.
. The majority posits that I consider dispositions under Federal Rule of Civil Procedure 12(b)(6) to be the only "actual dismissals” for the purpose of § 1915(g). Not so. As noted in the text above, dismissals pursuant to 28 U.S.C. §§ 1915(e)(2), 1915A(b) and 42 U.S.C. § 1997e(c) also count as strikes because those statutes expressly direct courts to "dismiss” an action if it is frivolous, malicious, or fails to state a claim. My emphasis on Rule 12(b)(6) serves only to illustrate from the well developed law considering Rule 12(b)(6) motions that "dismissal” differs from "summary judgment.”
. I do not "sow differences,” ante at 619, among my colleagues; I merely count votes, a majority of the court does not join in the rationale advocated by the concurring judges.
. Those colleagues who join the concurrence complain that I rewrite § 1915(a), inventing a “limitation” to the statute where none exists. Ante at 622. This criticism is exceedingly odd given that all of these judges also join the majority’s extensive rewrite of § 1915(g) to invent an entirely new category of cases— summary judgments — that will henceforth count as strike "dismissals.” And, of course, the concurrence's criticism is baseless: I do not "rewrite” § 1915(a). Rather, in accord with Supreme Court directives, I simply rely on statutory text, history, structure, and purpose to conclude that, in the limited context of prisoner cases, Congress intended § 1915(a)(1) to be displaced by a new and more specific statute, § 1915(g).
. As the majority properly and repeatedly explains, we requested Blakely only to "address whether certain orders [four summary judgment orders] constitute strikes” and our review is thus "restricted” to those orders. Ante at 610 n. 1; 615; and 617 n. 9. At its outset, the concurrence states it “agrees fully” with the majority. Id. at 618. Thus, the concurrence's criticism of Blakely (and me) for our failure to address other cases brought by Blakely, id. at 620-22, seems inexplicable. This is particularly so given that the record evidence as to the character of these cases is, to put it generously, thin.