Harvard Law Review Harvard Law Review Harvard Law Review

Habeas Corpus

Suspended Justice: The Case Against 28 U.S.C. § 2255’s Statute of Limitations


Habeas corpus has become almost a dead letter in American law. Many claims are still filed, and some are still granted, but after the Burger and Rehnquist Courts’ paring of habeas corpus’s scope and accessibility,1×1. See generally Kathleen Patchel, The New Habeas, 42 Hastings L.J. 939 (1991) (tracing the narrowing of habeas corpus by the Burger and Rehnquist Courts). as well as the introduction of a statute of limitations and restrictions on successive motions in the Antiterrorism and Effective Death Penalty Act of 19962×2. Pub. L. No. 104-132, 110 Stat. 1214 (codified as amended in scattered sections of the U.S. Code). (AEDPA), habeas has lost its luster.

However, there is still hope for habeas to be restored to its place as the ultimate protector of liberty. Although neither Congress nor the Court is likely to remove obstacles to habeas in the near future, there are seeds of a future resurgence. This Note focuses on one green shoot, arguing that AEDPA’s statute of limitations on habeas corpus for federal prisoners should be recognized as a violation of the Suspension Clause.3×3. U.S. Const. art. I, § 9, cl. 2 (“The Privilege of the Writ of Habeas Corpus shall not be suspended, unless when in Cases of Rebellion or Invasion the public Safety may require it.”). Alternatively, should the Supreme Court refuse to find that the writ has been suspended, federal prisoners should still have access to the residual federal habeas corpus right that is protected by the Constitution even after the statute of limitations has expired.

Part I will describe the origins of the statute that governs federal prisoners’ applications for collateral review, AEDPA’s revisions to the previous habeas regime, and the motives that drove AEDPA’s development. Part II argues that the statute of limitations on collateral review of federal prisoners’ detention likely violates the Suspension Clause under the Court’s primary recent cases dealing with suspension issues, Felker v. Turpin4×4. 518 U.S. 651 (1996). and Boumediene v. Bush.5×5. 553 U.S. 723 (2008). Although neither case decides the issue directly, Felker’s concerns and its justifications for finding no Suspension Clause violation do not apply to the statute of limitations under discussion here, while Boumediene’s reasoning suggests that the statute of limitations may rise to the level of a constitutional violation. Finally, even if the Court were not to recognize a Suspension Clause violation, there are still prudential and statutory reasons to find that, even after the statute of limitations under 28 U.S.C. § 2255 has run, prisoners should have access to habeas corpus with a pre-AEDPA scope.

I. Habeas Hobbled

A. Section 2255

The inclusion of the Suspension Clause in the Constitution underscores the Founders’ keen interest in adopting habeas corpus from the English common law. The first Congress wasted no time enshrining the writ in American statutory law, authorizing federal courts to grant it in the Judiciary Act of 1789.6×6. Ch. 20, § 14, 1 Stat. 73, 81. Habeas corpus initially applied only to prisoners held by the federal government until, in 1867, Congress passed a Habeas Corpus Act7×7. Ch. 28, 14 Stat. 385 (1867). that extended the writ to prisoners convicted by a state. For the next eighty years, habeas corpus remained relatively stable.

In 1948, however, Congress passed an act that fundamentally reshaped habeas corpus for federal prisoners. Among other changes, Congress introduced 28 U.S.C. § 2255,8×8. Act of June 25, 1948, Pub. L. No. 80-773, § 2255, 62 Stat. 869, 967. the statute under which prisoners being held on federal convictions apply for collateral review to this day. Initially, § 2255 aimed solely to clean up a jurisdictional mess. A prisoner filing a habeas corpus petition was, at common law, required to file in the jurisdiction he was being confined in. Thus, those districts housing federal prisons received enormous numbers of petitions, usually implicating trial transcripts, witnesses, and evidence housed in distant sentencing courts.9×9. See United States v. Hayman, 342 U.S. 205, 213–14 (1952).

To fix this inefficiency, Congress passed § 2255, which required that petitioners seek collateral relief in the jurisdiction where they were sentenced, but which “was not intended to change the substantive rights of federal prisoners to seek habeas review.”10×10. Sarah French Russell, Reluctance to Resentence: Courts, Congress, and Collateral Review, 91 N.C. L. Rev. 79, 95 (2012). The Supreme Court contemporaneously acknowledged that § 2255 was created for this reason. See Hayman, 342 U.S. at 219 (“[T]he sole purpose [of enacting § 2255] was to minimize the difficulties encountered in habeas corpus hearings by affording the same rights in another and more convenient forum.”); see also id. at 217 (quoting legislative history as indicating that § 2255 was meant to be “as broad as habeas corpus”). It allowed prisoners to challenge their detention on the same grounds as habeas at common law11×11. See Act of June 25, 1948, § 2255, 62 Stat. at 967 (“A prisoner in custody under sentence of a court of the United States claiming the right to be released upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States, or that the court was without jurisdiction to impose such sentence, or that the sentence was in excess of the maximum authorized by law, or is otherwise subject to collateral attack, may move the court which imposed the sentence to vacate, set aside or correct the sentence.”). and explicitly provided that “[a] motion for such relief may be made at any time.”12×12. Id. Although the statute required federal prisoners to pursue relief under a § 2255 motion instead of through habeas corpus proper, it allowed prisoners recourse to the writ if it appeared “that the remedy by motion is inadequate or ineffective to test the legality of his detention.”13×13. Id. § 2255, 62 Stat. at 968.

Even though § 2255 was designed to, and in fact did, provide protections coextensive with those of habeas corpus, the Supreme Court promptly held that § 2255 was distinct from habeas corpus.14×14. See Hayman, 342 U.S. at 220 (“This is not a habeas corpus proceeding.”). So long as § 2255 had as broad a scope as habeas, and so long as habeas was still available in case a § 2255 motion was inadequate or ineffective, the writ had not been suspended. Section 2255 stood in the place of habeas without quite taking on its identity; should it fail to serve justice, habeas was still waiting in the wings.


In 1996, however, Congress fundamentally rewrote and restructured § 2255 with AEDPA. Although AEDPA describes itself as “[a]n Act [t]o deter terrorism, provide justice for victims, [and] provide for an effective death penalty,”15×15. Pub. L. No. 104-132, 110 Stat. 1214 (1996) (codified as amended in scattered sections of the U.S. Code). its effect on collateral challenges was to greatly restrict access to § 2255 relief and habeas corpus.

1. Statutory Roadmap. — The present state of habeas law is a dizzying maze of interrelated statutes; a brief illustration of the relief available to various classes of prisoners should aid in understanding habeas corpus post-AEDPA. As described above, § 2255 is the primary statute governing collateral relief for prisoners convicted in federal court. Section 2254,16×16. 28 U.S.C. § 2254 (2012). meanwhile, is the statute providing for habeas corpus for prisoners who were convicted by a state court.17×17. Section 2254 is the descendant of the Habeas Corpus Act of 1867, which initially authorized state prisoners to seek habeas corpus. See Ch. 28, 14 Stat. 385. Unlike the remedy in § 2255, the collateral review provided for in § 2254 is itself habeas corpus. Section 2254 applications are additionally governed by the restrictions set forth in 28 U.S.C. § 2244; most importantly, § 2244 contains restrictions on second or successive § 2254 applications and the statute of limitations for § 2254 applications. Notably, although § 2254 and § 2244 contain several restrictions relating to § 2254’s nature as a challenge to a state court ruling,18×18. For instance, 28 U.S.C. § 2254(b)(1) requires that the applicant “exhaust[] the remedies available in the courts of the State” or show that there is no adequate state corrective process. There are additional rules governing res judicata, id. § 2254(d), and factfinding, id. § 2254(e). those areas addressed both by these statutes and by § 2255 are drafted in parallel.19×19. Compare, e.g., id. § 2255(f) (declaring that “[a] 1-year period of limitation shall apply to a motion under this section” and establishing four potential triggering dates), with id. § 2244(d) (declaring that “[a] 1-year period of limitation shall apply to an application for a writ of habeas corpus by a person in custody pursuant to the judgment of a State court” and providing four triggering conditions nearly verbatim to those in § 2255(f)).

Finally, in 28 U.S.C. § 2241, true habeas corpus for federal prisoners has survived. This statute is the direct descendant of the habeas authorization in the Judiciary Act of 1789.20×20. See Ch. 20, § 14, 1 Stat. 81. Section 2241 itself allows for prisoners in custody under a federal sentence to file a habeas corpus petition;21×21. See 28 U.S.C. § 2241(c)(1)–(2). however, even though this general authorization was never repealed, § 2255(e) provides a gatekeeping mechanism. Under this section, if a federal prisoner is eligible to file a § 2255 motion and either fails to do so or loses the motion, then a § 2241 application for habeas corpus proper “shall not be entertained . . . unless it also appears that the remedy by motion is inadequate or ineffective to test the legality of the decision.”22×22. Id. § 2255(e).

2. Habeas After AEDPA. — The first five provisions of AEDPA’s newly fashioned § 2255 carry over portions of the 1948 act verbatim, including the definition of § 2255’s scope23×23. Compare id. § 2255(a), with Act of June 25, 1948, Pub. L. No. 80-773, § 2255, 62 Stat. 869, 967. AEDPA’s § 2255(a) retains the original language verbatim, save for “court established by Act of Congress” instead of “court of the United States.” This minor change stems from an amendment to § 2255 passed in 1949. See Act of May 24, 1949, Pub. L. No. 81-72, § 114, 63 Stat. 89, 105. and its barrier to applying for habeas when a § 2255 motion could be filed.24×24. 28 U.S.C. § 2255(e). AEDPA went further, however, and also created a one-year statute of limitations for § 2255 motions25×25. Id. § 2255(f). and substantially narrowed the availability of successive motions.26×26. Id. § 2255(h). As § 2255 motions could be filed, pre-AEDPA, “at any time,” and as successive motions were prohibited only when they were “for similar relief on behalf of the same prisoner,”27×27. Id. § 2255 (1994) (amended 1996). these two provisions greatly narrow the availability of collateral review.

The concerns driving AEDPA’s revision of habeas corpus law explain these changes. AEDPA itself is a Frankenstein’s monster of a law; it is in essence the result of combining numerous bills dating back as far as 1981 into one overarching reform of the criminal justice system.28×28. See Larry W. Yackle, A Primer on the New Habeas Corpus Statute, 44 Buff. L. Rev. 381, 381 (1996) (“[AEDPA] is not well drafted. It bears the influence of various bills that were fiercely debated for nearly forty years.”). In debating AEDPA, Congress was primarily focused upon problems that arose when state prisoners applied for § 2254 habeas relief.29×29. The three most pressing concerns addressed on the floor were: (1) that endless successive petitions clogged up the court system; (2) that a statute of limitations and ban on successive filing would increase public confidence in the court system, specifically by preventing death row prisoners from indefinitely deferring their executions; and (3) that imposing these limitations would promote finality in the criminal justice system. See Benjamin R. Orye III, Note, The Failure of Words: Habeas Corpus Reform, the Antiterrorism and Effective Death Penalty Act, and When a Judgment of Conviction Becomes Final for the Purposes of 28 U.S.C. § 2255(1), 44 Wm. & Mary L. Rev. 441, 453–55 (2002). Numerous representatives cited to the report prepared by the Ad Hoc Committee on Federal Habeas Corpus in Capital Cases (Powell Report), a committee chaired by retired Justice Lewis Powell to “inquire into ‘the necessity and desirability of legislation directed toward avoiding delay and the lack of finality’ in capital cases in which the prisoner had or had been offered counsel.”30×30. Ad Hoc Comm. on Fed. Habeas Corpus in Capital Cases, Report on Habeas Corpus in Capital Cases, 45 Crim. L. Rep. (BNA) 3239, 3239 (1989). Although the Powell Report was motivated by problems in capital cases, those provisions stemming from findings in the Report are not limited solely to capital contexts. One of the main problems the Powell Report found was that the habeas system led to “unnecessary delay and repetition.”31×31. Id. The bulk of delay and repetition, however, was due to the proceedings being shuttled back and forth for state remedy exhaustion, or pre-habeas presentation of claims to state courts; as § 2255 motions come before a federal court originally, these motions were not directly implicated by the Powell Report.32×32. See Orye, supra note 29, at 456.

Ultimately, although the main justifications for reforming habeas corpus originated with observed problems with federal review of state convictions, parallel restrictions were written into both § 2254 and § 2255.33×33. See, e.g., supra note 19. Congress hardly examined the potential effects of these changes on § 2255 or discussed whether § 2255 needed to be changed at all.34×34. The most direct attention this question received in the debates over AEDPA was in a statement by Representative Robert Kastenmeier decrying the “wholesale” incorporation of the Powell Committee’s suggestions to reform § 2254 practice into the inapposite § 2255 context, which did not present the same problems. See 136 Cong. Rec. 27,521–22 (1990) (statement of Rep. Kastenmeier). Although AEDPA’s changes to habeas were inspired by and tailored to § 2254 habeas and capital cases, Congress inserted a one-year statute of limitations and limited bases for successive motions into both § 2254 habeas from state convictions as well as § 2255 motions.

Notably, Congress did not create a statute of limitations for federal habeas corpus under § 2241 even though it created one for collateral review proceedings for both state and federal prisoners. It may be that Congress believed that imposing a statute of limitations on the category of habeas corpus closest to the common law version protected by the Suspension Clause would have been an unconstitutional suspension,35×35. As noted above, § 2255 is not itself habeas corpus, and so there is a clear argument that restricting access to § 2255 motions does not constitute a suspension so long as habeas continues to exist somewhere in the background. And federal habeas corpus for state prisoners did not exist until 1867. Section 2241 is the descendant of the habeas corpus passed in the Judiciary Act of 1789, and it has been argued that the Suspension Clause prevents Congress only from suspending habeas corpus for federal prisoners. See Abernathy v. Wandes, 713 F.3d 538, 553–55 (10th Cir. 2013) (collecting cases). or that it believed it had the power to impose conditions on habeas corpus so long as it did not fully abolish it. It may even be that Congress was not paying close enough attention to notice. Unfortunately, this question is unanswerable as this issue never arose during any of AEDPA’s legislative history. We can only speculate.

3. The Post-AEDPA Status Quo. — What is clear, though, is that AEDPA and § 2255 have so far been upheld as constitutional. Even beyond the canon of construction that Congress intends to legislate constitutionally,36×36. See Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988) (“The courts will . . . not lightly assume that Congress intended to infringe constitutionally protected liberties or usurp power constitutionally forbidden it.”). § 2255 as it stands has the benefit of more than sixty years of supportive case law. In United States v. Hayman,37×37. 342 U.S. 205 (1952). the Court’s first investigation of § 2255’s constitutionality, the Court indicated that the statute’s savings clause, which allows a petitioner access to habeas corpus proper when the § 2255 remedy is “inadequate or ineffective,” obviated the need to address § 2255’s constitutionality.38×38. Id. at 223 (quoting 28 U.S.C. § 2255(e) (2012)). So long as there was no functional difference in the remedies and the savings clause provided an avenue to habeas when necessary, there was no need to pass upon the statute’s constitutionality.39×39. Id.

The Court treated § 2255’s constitutionality more thoroughly in Swain v. Pressley.40×40. 430 U.S. 372 (1977). Although the statute at issue in Swain was a District of Columbia law and not § 2255 itself, the Court found that it was “deliberately patterned after” § 2255 and governed by the same principles.41×41. Id. at 377. The Court focused its attention on the savings clause again, finding that “the only constitutional question presented is whether the substitution of a new collateral remedy which is both adequate and effective should be regarded as a suspension of the Great Writ.”42×42. Id. at 381. Citing Hayman and its “implicit[]” holding that such a substitution is not a suspension,43×43. Id. and noting that both § 2255 and the District of Columbia statute were exactly commensurate with habeas corpus, the Court held that neither statute amounted to a suspension under the Suspension Clause.44×44. Id. at 381–83 (citing Hill v. United States, 368 U.S. 424, 427 (1962) (holding that § 2255 is the exact equivalent of the pre-existing habeas corpus remedy)).

Hayman and Swain at first blush seem to support the constitutionality of habeas substitutes against Suspension Clause challenges. However, a key aspect of these two cases is that they address only the constitutionality of a habeas substitute as such, and not that of a habeas substitute of any particular dimensions. That is, while these cases show that providing prisoners access to a different remedy than habeas, with only the de minimis difference in venue, does not violate the Suspension Clause, they do not establish that a habeas substitute with consequential additional restrictions is constitutional. This reading is confirmed by Boumediene, which involved a habeas substitute that differed greatly from habeas proper and was found unconstitutional, Hayman and Swain notwithstanding.45×45. See infra pp. 1100–03. For pre-AEDPA § 2255, these cases foreclosed Suspension Clause challenges. For post-AEDPA § 2255, a substantially different beast than habeas proper, they do not.

Beyond Hayman and Swain, the Court has decided very few cases considering the constitutionality of § 2255. As the next Part shows, however, close attention to § 2255’s evolution and to the reasoning underlying the Court’s few Suspension Clause cases suggests that it may be possible to establish that § 2255’s statute of limitations constitutes a suspension.

II. The Writ Resurgent

Originally, § 2255 simply provided an alternative mechanism for federal prisoners to obtain collateral review, one with an identical scope and availability to common law habeas corpus. There could be little doubt that this alternative was constitutional, notwithstanding the Suspension Clause. Section 2255 was not properly habeas corpus, but because it had not in any but the most formalistic of ways restricted access to the same review provided by habeas, its interposition was of little legal significance.46×46. See Swain, 430 U.S. at 381 (“[T]he substitution of a collateral remedy which is neither inadequate nor ineffective to test the legality of a person’s detention does not constitute a suspension of the writ of habeas corpus.”).

Once Congress passed AEDPA, however, the situation changed. With a one-year statute of limitations, § 2255 was no longer coextensive with habeas corpus as it existed at common law, where applications for the writ were never time-barred nor limited in number.47×47. See Mayle v. Felix, 545 U.S. 644, 655 (2005) (“In enacting AEDPA in 1996, Congress imposed for the first time a fixed time limit for collateral attacks in federal court on a judgment of conviction.”). Instead, § 2255 shrank the circle of those with access to collateral review from all those with colorable claims to all those with colorable claims pursued within one year. Just like the old British Empire, habeas used to stretch across the entire expanse of one’s sentence; now, having lost many of the jewels in its crown, habeas has been restricted to a fairly small set of islands.

One could easily conjure sympathetic cases where the statute of limitations seems particularly unjust. For instance, a prisoner may have been given a life sentence based on aggravating factors, only to have these aggravating factors overturned by a later decision.48×48. See, e.g., United States v. Surratt, 797 F.3d 240 (4th Cir. 2015) (holding that § 2255(e) does not allow a § 2241 petition when the sentencing factors and not the conviction are undermined by subsequent circuit decision and § 2255 procedurally bars the claim). Or a prisoner may be illiterate and unable to afford a lawyer until more than one year has passed, making a collateral challenge effectively impossible. And in fact, courts have used their equitable powers to make some allowances for particularly sympathetic classes of prisoners. Actual innocence of the crime of conviction, if proven, will suffice to bypass otherwise ironclad limitations on habeas appeals.49×49. See McQuiggin v. Perkins, 133 S. Ct. 1924, 1928 (2013) (“We hold that actual innocence, if proved, serves as a gateway through which a petitioner may pass whether the impediment is a procedural bar . . . or . . . expiration of the statute of limitations.”). Less dramatically, but more commonly, courts have been receptive to equitable tolling of the limitations period,50×50. See, e.g., Ramos-Martínez v. United States, 638 F.3d 315, 320–22 (1st Cir. 2011); United States v. Martin, 408 F.3d 1089, 1093–95 (8th Cir. 2005). and the contours of this equitable accommodation have attracted great academic interest.51×51. See, e.g., Lisa L. Bellamy, Playing for Time: The Need for Equitable Tolling of the Habeas Corpus Statute of Limitations, 32 Am. J. Crim. L. 1 (2004); Brandon Segal, Habeas Corpus, Equitable Tolling, and AEDPA’s Statute of Limitations: Why the Schlup v. Delo Gateway Standard for Claims of Actual Innocence Fails to Alleviate the Plight of Wrongfully Convicted Americans, 31 U. Haw. L. Rev. 225 (2008).

But access to habeas corpus should not be premised on whether a case is particularly sympathetic. The Suspension Clause does not limit its protection to those with meritorious claims; it preserves habeas generally so that courts may grant those petitions grounded in the law.52×52. This broad reading of the Suspension Clause may seem at odds with the more restrained conception of habeas on display in Teague v. Lane, 489 U.S. 288 (1989), and similar cases. In Teague, the Court made sure to point out that it “never has defined the scope of the writ simply by reference to a perceived need to assure that an individual accused of crime is afforded a trial free of constitutional error.” Id. at 308 (quoting Kuhlmann v. Wilson, 477 U.S. 436, 447 (1986)). Instead, the Court observed, “interests of comity and finality must also be considered in determining the proper scope of habeas review.” Id. These countervailing values are inapposite here, however. Section 2255 motions appeal from a federal court to a federal court, sidestepping any comity concerns, and delayed first motions do not endanger finality in the same way that allowing endless successive motions would. See infra pp. 1107–08. While nearly all constitutional rights yield to some extent in the face of compelling state interests, when the state interest is slight and the promised right is broad, such a Faustian bargain is not necessary. The benchmark protections of criminal procedure exist primarily to protect the innocent, but they also apply to the manifestly guilty. Similarly, when prisoners are granted access to habeas outside of the statute of limitations only when they fall into a sympathetic class, this discretionary access does not vitiate the right as such. Bypassing the statute of limitations to hear actual innocence claims, or equitably tolling the statute of limitations, is better than not doing so. But if the imposition of the statute of limitations on cases outside these limited classes rises to the level of a Suspension Clause violation, these half measures do not cure the defect.

However, all may not be lost. The principles espoused in the Court’s Suspension Clause cases suggest that AEDPA constitutes an unconstitutional suspension of habeas corpus to prisoners convicted under federal law. If courts decline to fully embrace this constitutional argument, then they should at least read AEDPA as allowing access to habeas proper once the statute of limitations has expired in order to avoid the serious constitutional issue.

A. Supreme Court Treatment

In the past twenty years, the Court has heard two major cases concerning the Suspension Clause in the context of habeas corpus, Felker and Boumediene.53×53. The Court also heard INS v. St. Cyr, 533 U.S. 289 (2001), which addressed whether another section of AEDPA or the Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, div. C, 110 Stat. 3009-546 (codified as amended in scattered sections of the U.S. Code), violated the Suspension Clause by withholding access to habeas from immigrants held by the INS. Insofar as it concerned the Suspension Clause, however, this case largely restricted itself to showing that habeas corpus, as it existed in 1789, encompassed legal challenges to executive detention even for noncitizens. St. Cyr also explored whether these statutes, in repealing a previous statute confirming access to habeas in immigration contexts, withdrew habeas; the Court found that “its repeal cannot be sufficient to eliminate what it did not originally grant — namely, habeas jurisdiction pursuant to 28 U.S.C. § 2241,” St. Cyr, 533 U.S. at 310, and, invoking the constitutional avoidance canon, it distinguished other purported statutory attempts to restrict access to habeas as lacking “a clear, unambiguous, and express statement of congressional intent to preclude judicial consideration on habeas,” id. at 314. As the savings clause of § 2255 does explicitly mention § 2241 and places undeniable limits on access to it, see 28 U.S.C. § 2255(e) (2012), St. Cyr does not speak directly to the constitutionality of § 2255. Felker is inapposite to the statute-of-limitations issue. However, many of the arguments made in the majority decision of Boumediene, the Court’s most extensive discourse on the Suspension Clause, could provide support for recognizing a residual habeas right after the expiration of the statute of limitations.

1. Felker. — Felker dealt with AEDPA’s § 2244(b), which limits state prisoners from filing subsequent habeas petitions: claims presented in the previous petition must be dismissed, while those that were not raised in the first petition cannot be considered unless they fall under several delineated exceptions.54×54. 28 U.S.C. § 2244(b)(1)–(2).

Concerning the Suspension Clause issue, the Court observed that “the power to award the writ by any of the courts of the United States, must be given by written law”55×55. Felker v. Turpin, 518 U.S. 651, 664 (1996) (quoting Ex parte Bollman, 8 U.S. (4 Cranch) 75, 94 (1807)). and “judgments about the proper scope of the writ are ‘normally for Congress to make.’”56×56. Id. (quoting Lonchar v. Thomas, 517 U.S. 314, 323 (1996)). More specifically, the Court held that the limitations on successive petitions largely codified the existing limits on abuse of the writ that had developed over the history of habeas. Describing abuse of the writ as “a complex and evolving body of equitable principles informed and controlled by historical usage, statutory developments, and judicial decisions,”57×57. Id. (quoting McCleskey v. Zant, 499 U.S. 467, 489 (1991)). the Court found that § 2244(b)’s added restrictions were “well within the compass of this evolutionary process”58×58. Id. and thus did not constitute a Suspension Clause violation.59×59. The Court also held that the restriction on second or successive motions did not unconstitutionally limit the Court’s jurisdiction to hear habeas petitions filed under the Court’s original jurisdiction. Id. at 660–62. Importantly here, since federal prisoners cannot file a § 2255 motion originally in the Supreme Court, and the Court’s original jurisdiction is through § 2241, the limitations on § 2255 do not implicate the original jurisdiction issue.

Felker can be distinguished, as the statute of limitations is a novel addition to habeas corpus restrictions. Felker noted that some of the limits codified in § 2244(b) already applied to successive petitions.60×60. Id. at 664. Those limitations that did not previously apply to successive petitions were found to be “within the compass of this evolutionary process” of the abuse-of-the-writ doctrine.61×61. Id. Before AEDPA, however, there was never a statute of limitations on habeas petitions. There were rules that let judges dismiss a habeas petition for equitable reasons, such as unjustified delay that caused actual prejudice to a state in responding to the petition,62×62. See Lonchar v. Thomas, 517 U.S. 314, 316 (1996) (citing Rules Governing Section 2254 Cases in the United States District Courts, 28 U.S.C. § 2254, Rule 9(a) (1994) (amended 2004)). but no strict time limit existed and not without a showing of prejudice. Unlike § 2244(b)’s limitations on successive motions, which were largely commensurate with the existing abuse-of-the-writ doctrine and included some related additions, the statute of limitations is a new beast, transforming an equitable consideration of whether the petition is prejudicially delayed into a firm procedural bar.63×63. It is, of course, possible that the Court would decide that the statute of limitations, or § 2255(e)’s gatekeeping restriction, does not restrict the Court from hearing those habeas petitions it wishes to hear. Cf. Felker, 518 U.S. at 663 (“Whether or not we are bound by [the limits on second or successive habeas applications], they certainly inform our consideration of original habeas petitions.”). And if it did, this decision would foreclose the argument that the statute of limitations together with § 2255(e) constitutes a Suspension Clause violation, because there would still be access to habeas outside of the AEDPA statutory structure. But such a decision would do so by removing the potential violation rather than denying it. While not conclusive, a strong case can be made, then, that Felker’s logic compels the conclusion that AEDPA’s novel § 2255 statute of limitations constitutes a violation of the Suspension Clause.

2. Boumediene. The reasoning of Boumediene, the Court’s most searching engagement with the Suspension Clause, strongly suggests that § 2255’s statute of limitations is unconstitutional. Although Boumediene focused on national security in considering whether the Detainee Treatment Act of 200564×64. Pub. L. No. 109-148, 119 Stat. 2739 (codified as amended in scattered sections of the U.S. Code). and the Military Commissions Act of 200665×65. Pub. L. No. 109-366, § 7, 120 Stat. 2600, 2635–36 (codified as amended at 28 U.S.C. § 2241 (2012)), invalidated by Boumediene v. Bush, 553 U.S. 723 (2008). gave Guantanamo Bay prisoners an adequate alternative to habeas,66×66. See Boumediene, 553 U.S. at 769–70. the principles underlying its finding of a Suspension Clause violation can be directly applied to § 2255’s statute of limitations.

First, the Court suggested that the Suspension Clause implications of a statute reforming habeas proceedings may depend on whether the statute restricts access to habeas.67×67. Id. at 774. In Hayman and Swain, the “two leading cases addressing habeas substitutes,”68×68. Id. the statutes in question created habeas substitutes that differed in only administrative ways from habeas proper without changing the substance of the remedy, and the Boumediene Court thus found that these cases did not govern the Suspension Clause issue under consideration.69×69. Id. The Court acknowledged that AEDPA had added some restrictions to habeas, noting Felker’s approval of AEDPA’s restriction on successive motions.70×70. Id. However, the Court observed that the provisions at issue in Felker “did not constitute a substantial departure from common-law habeas procedures,” as they “codified the longstanding abuse-of-the-writ doctrine.”71×71. Id. Further distinguishing Felker, the Court noted that AEDPA cases occur after full criminal proceedings have been carried out, whereas in Boumediene there had been no trial. Id. As these cases underpin many of the subsequent lower court decisions that have upheld AEDPA’s statute of limitations and restricted subsequent access to § 2241 writs,72×72. See, e.g., cases cited infra note 127. the fact that the Court found them to be quite distinct from the case in Boumediene is significant. Common law habeas had no statute of limitations and creating one certainly restricted habeas rather than strengthening it. The Court’s reasoning thus suggests that AEDPA’s statute of limitations may be closer to Boumediene’s Suspension Clause violation than to the earlier cases’ de minimis and ultimately permissible tweakings.

Second, the Court looked to Congress’s intention in creating the new procedure. It noted that “the legislative history confirms what the plain text strongly suggests: In passing the DTA Congress did not intend to create a process that differs from traditional habeas corpus process in name only. It intended to create a more limited procedure.”73×73. Boumediene, 553 U.S. at 778. Similarly, in enacting AEDPA’s addition of a statute of limitations, Congress intended to go far beyond the kind of changes “in name only” presented by earlier § 2255 cases. A review of AEDPA’s legislative history reveals that its changes were intended not to make habeas proceedings more efficient, but instead to reduce the number of petitions filed in district courts and to ensure that, with vanishingly few exceptions, a prisoner could not challenge even the most unjust of convictions after the statute of limitations had run.74×74. See supra section I.B.2, pp. 1093–95. One can twist these motivations to view them through the filter of efficiency: a statute of limitations motivates prisoners to file all their challenges at once, while the record is still fresh, which allows the court system to reap the benefits of judicial efficiency and sentence finality. However, one can just as easily perform this trick on the facts in Boumediene: by centralizing review in the D.C. Circuit, Congress attempted to ensure that a court experienced with national security issues and state secrets would handle all Guantanamo cases. The Court, though, made clear that it is more concerned with the substantive effects of a change than its form. Creating a habeas substitute is not necessarily a Suspension Clause violation; creating one that substantially deviates from habeas’s common law protections at the very least “test[s] the limits of the Suspension Clause in ways that Hayman and Swain did not.”75×75. Boumediene, 553 U.S. at 779.

Third, the Court referred several times to the fact that the statutes in Hayman and Swain contain a “saving[s] clause” — a provision that allows for recourse to habeas proper when the alternative process is inadequate and ineffective — as a mark in their favor.76×76. Id. at 776. The Court was clear that Hayman and Swain “placed explicit reliance upon these provisions in upholding the statutes against constitutional challenges.”77×77. Id. And when the availability and substance of the remedy is exactly the same, as it was in those two cases, a savings clause does render moot any constitutional objections to the statutes at issue. It is almost certain that the result of a § 2255 motion will be identical whether it is considered in the district of detention or sentencing, but just in case an unforeseen complication causes this precise and minor difference to render the remedy inadequate, recourse can be had to § 2241 habeas. In a post-AEDPA world, though, § 2255 motions and § 2241 applications are no longer congruent. One has a time limit, and the other does not. The savings clause does not provide any safety from the statute of limitations. Most likely, the statute of limitations will serve as a per se barrier to any remedy. As described in more detail below, most circuits have developed some test to determine when the savings clause is invoked, each concerning the substantive merits of the motion.78×78. See infra note 127. The statute of limitations provides no discretion to the reviewing judge to consider its effect on the adequacy of the remedy. Once the statute runs, one may not file a § 2255 motion: the application for § 2241 habeas “shall not be entertained” if one has “failed to apply for relief” under § 2255 “unless it also appears that the remedy by motion is inadequate or ineffective to test the legality of [one’s] detention.”79×79. 28 U.S.C. § 2255(e) (2012) (emphasis added). As in Abernathy v. Wandes,80×80. 713 F.3d 538 (10th Cir. 2013). in which the Tenth Circuit declined to find § 2255 inadequate and ineffective where the petitioner was legally foreclosed from pursuing the claim when a § 2255 motion would have been timely,81×81. Id. at 556–57. courts are likely to divorce the procedural from the substantive, evaluating the adequacy of the remedy based on its adequacy if all procedural requirements were fulfilled. Where the procedural requirements are themselves the problem, the underlying remedy is considered intact. But this approach fails to account for the incongruity in remedies identified above. Because § 2255 thus categorically deprives some petitioners of the habeas relief promised them by the Constitution, it should get more serious scrutiny.

Finally, the Court addressed questions of process, observing that whether a habeas substitute is adequate and effective might be affected by the amount of process that preceded the application for a writ.82×82. See Boumediene, 553 U.S. at 783. In Boumediene, there had been no initial proceedings, while criminal trials, “the most rigorous proceedings imaginable,”83×83. Id. at 785. precede most § 2255 motions. However, the Court conceded that “the Suspension Clause remains applicable and the writ relevant . . . even where the prisoner is detained after a criminal trial conducted in full accordance with the protections of the Bill of Rights,” observing that if this were not so, the Court would not have needed to evaluate the adequacy of the habeas substitutes in Hayman and Swain.84×84. Id. While the amount of process at earlier stages of litigation is relevant, then, it is not dispositive.

With the Court’s reasoning in mind, the imposition of a set-in-stone statute of limitations, without allowing for differing circumstances or judicial discretion to hear or dismiss § 2255 motions based on their timeliness, is closer to Boumediene than to Hayman and Swain.85×85. As noted above, Felker dealt with the successive-motions section, which codified elements of the common law’s abuse-of-the-writ doctrine. As such, the Court described Felker as being closer to a Hayman case, where the form of the remedy changes while its substance remains untouched. See id. at 774. To be sure, the context of Boumediene distinguishes it from § 2255. Most significantly, Boumediene implicates executive detention, a context in which habeas has always been available and which forms the core of habeas as understood when the Suspension Clause was written.86×86. See INS v. St. Cyr, 533 U.S. 289, 301 (2001) (“At its historical core, the writ of habeas corpus has served as a means of reviewing the legality of Executive detention, and it is in that context that its protections have been strongest.”). By contrast, § 2255 covers collateral review of convictions, which is a role that habeas grew to fill largely over the last century.87×87. See Felker v. Turpin, 518 U.S. 651, 663 (1996) (“[I]t was not until well into [the twentieth] century that this Court interpreted [the Act of 1867] to allow a final judgment of a conviction in a state court to be collaterally attacked on habeas.”). The argument that the Suspension Clause applies only to cases covered by habeas as originally conceived, however, does not prevent Boumediene from informing an interpretation of § 2255. First, the Court has studiously avoided deciding this issue: in Felker, the Court assumed for the purposes of the decision that the clause extended to modern-day habeas;88×88. See id. at 663–64. in INS v. St. Cyr,89×89. 533 U.S. 289. the Court held that the clause protected the original understanding of habeas “at the absolute minimum”;90×90. Id. at 301. and in Boumediene, the Court noted that it “has been careful not to foreclose the possibility that the protections of the Suspension Clause have expanded along with post-1789 developments that define the present scope of the writ.”91×91. Boumediene, 553 U.S. at 746. This “careful” forbearance suggests that the two contexts may be similarly covered.92×92. See Daniel J. Meltzer, Habeas Corpus, Suspension, and Guantánamo: The Boumediene Decision, 2008 Sup. Ct. Rev. 1, 23 (arguing that the Suspension Clause protected a common law writ that had been developing for centuries, and that interpreting this clause as enshrining a “snapshot” of the 1789 writ “would be a departure from the history of the writ with which the Founders were familiar”). When the original understanding is sufficient to decide a case, the Court does not proceed further, as in St. Cyr and Boumediene. When it is not, as in Felker, the Court considers the Suspension Clause issue despite the lack of certainty in that clause’s reach. Second, even if the difference between executive detention and collateral review is significant enough to distinguish Boumediene from § 2255, Boumediene is one of the few cases that has deeply engaged with the Suspension Clause in detail and demonstrated what factors are important in finding a suspension. Because the Court discussed the Suspension Clause generally, without cabining its discussion to the limited context of executive detention, and because there are so few other cases to inform Suspension Clause evaluations in other contexts, Boumediene’s reasoning should be seen as persuasive even in the context of collateral review.

Another potential difference between Boumediene and § 2255 is that the statute at issue in Boumediene, the Military Commissions Act, eliminated the remedy of habeas for enemy-combatant detainees, while § 2255 only limits the availability of the habeas substitute beyond one year. It is undeniable that removing habeas access entirely is more severe than providing access to a time-limited remedy. While this means that Boumediene does not conclusively show that the statute of limitations is a suspension, however, it does not mean that Boumediene fails to speak to the issue. Ultimately, we have two strong data points in Felker and Boumediene by which to evaluate the Suspension Clause’s application to habeas limitations. In Felker, the added limitation was not extremely burdensome because it was consistent with the existing abuse-of-the-writ doctrine and represented an organic evolution of existing habeas. In Boumediene, the statute did heavily burden the habeas right because it cut off access to habeas entirely. Trying to answer whether the statute of limitations rises to the level of a suspension violation using these two conclusions alone is futile; the statute of limitations is a novel, inorganic, and highly restrictive limitation on habeas, more burdensome than the limitation in Felker. However, the statute of limitations still provides limited access to the writ, and is thus less restrictive than the limitation in Boumediene. Falling in the middle, the statute of limitations is directly governed by neither. After examining the reasoning and factors employed in these two decisions, however, the statute of limitations seems to fall under Boumediene rather than Felker, for the reasons described above. This is admittedly not a clear, indisputable method for determining whether the statute of limitations constitutes a Suspension Clause violation. In the absence of a Supreme Court decision directly on point, however, it is the best method available.

B. Circuit Court Treatment

Since the passage of AEDPA, courts of appeals have rejected arguments that § 2255(f)’s statute of limitations constitutes a Suspension Clause violation and that prisoners have access to § 2241 petitions once the § 2255 statute of limitations has run.

Some cases deal specifically with the statute of limitations. In Miller v. Marr,93×93. 141 F.3d 976 (10th Cir. 1998). the Tenth Circuit noted that “[w]hether [§ 2255(f)] violates the Suspension Clause depends upon whether the limitation period renders the habeas remedy ‘inadequate or ineffective’ to test the legality of detention.”94×94. Id. at 977 (citing Swain v. Pressley, 430 U.S. 372, 381 (1977)). Appealing to Felker, the court observed that the Supreme Court had held that restrictions on filing second or successive appeals did not suspend habeas, and pointed out that “the Court expressed a clear deference to the rules that Congress has fashioned concerning habeas.”95×95. Id. at 978; see also id. at 977–78. Referencing the “serious constitutional issue” test of Triestman v. United States,96×96. 124 F.3d 361 (2d Cir. 1997). Triestman established the “serious constitutional issue” test for § 2255, suggesting that district courts should “find that habeas corpus may be sought whenever situations arise in which a petitioner’s inability to obtain collateral relief would raise serious questions as to § 2255’s constitutionality.” Id. at 377. The court explicitly avoided asking whether a serious Suspension Clause issue was in play and decided Triestman on due process and Eighth Amendment grounds. Id. at 378–79. the court stated simply that “[a]fter considering the claims Mr. Miller desires to raise . . . , we are satisfied that such circumstances are not implicated here.”97×97. Miller, 141 F.3d at 978. Miller’s treatment is consistent with other cases directly addressing the Suspension Clause’s application to § 2255: a court finds summarily that no Suspension Clause issue exists by relying primarily on Felker’s holding that limiting habeas does not rise to a Suspension Clause violation98×98. But see supra section II.A.1, 1098–100. and Congress’s assumed purpose in creating a statute of limitations.99×99. But see supra section I.B.2, pp. 1093–95.

More commonly, the question of whether § 2255’s procedural limitations render it inadequate and ineffective, opening the door to a § 2241 petition, arises in cases challenging the second-or-successive-motion limitation. One recent case, United States v. Surratt,100×100. 797 F.3d 240 (4th Cir. 2015). falls into this category. The Fourth Circuit opened by discussing the “narrow gateway to § 2241 relief for certain prisoners found actually innocent of their offenses of conviction,”101×101. Id. at 247. a gateway opened by many other circuits. Finding that Surratt presented no claim of innocence, the court proceeded to statutory interpretation of the savings clause.

The court’s arguments cover the same basic ground as those in other circuits. First, the court noted that the savings clause triggers when the petitioner has no opportunity to “test” the legality of his detention.102×102. Id. at 251; see also Prost v. Anderson, 636 F.3d 578, 584 (10th Cir. 2011) (“[T]he [savings] clause is concerned with process — ensuring the petitioner an opportunity to bring his argument — not with substance — guaranteeing nothing about what the opportunity promised will ultimately yield in terms of relief.”). That is, it operates only when there has been no chance to be heard, no “unobstructed procedural shot,” and not when the petitioner could have tested the legality of his detention in a proper § 2255 motion.103×103. Surratt, 797 F.3d at 252 (quoting Rice v. Rivera, 617 F.3d 802, 807 (4th Cir. 2010)). Notably, this reading denies subsequent relief when a petitioner’s claim was foreclosed by contrary precedent at the proper time of a motion, which precedent later changed. Id. at 252–53; see also supra notes 80–81 and accompanying text. Calling “Congress’s deliberate use of the word ‘test,’ rather than a more expansive term like ‘guarantee’ or ‘ensure,’ very meaningful,” the court ruled that the substantive inability to raise the claim had no role in whether § 2255 was inadequate or ineffective.104×104. Id. at 251; see also id. at 251–52.

This focus on formal opportunity to be heard, even if there was no actual chance of relief when that opportunity occurred, is based on a misguided interpretation of terms divorced from their context. Quite simply, AEDPA did not revise the language of the savings clause at all. This language was included in § 2255 before AEDPA’s restrictions were added — that is, when § 2255 and § 2241 promised equivalent remedies.105×105. See Act of June 25, 1948, Pub. L. No. 80-773, § 2255, 62 Stat. 869, 968. To place great significance on the savings clause’s use of “test” or “remedy” instead of “ensure” and “relief” is to imbue the drafting choices of 1948 with the context of 1996. While it is possible that Congress drafted AEDPA’s habeas restrictions, looked to the savings clause’s wording, and made a reasoned decision that the words already used captured Congress’s precise intent in light of the divergence it was creating between § 2255 and § 2241, there is no indication of this. Instead, the clause’s diction is likely a historical accident without any particular guidance concerning AEDPA’s subsequent additions.

Similarly, the court posited that the savings clause’s use of “detention,” rather than “sentence” or “conviction,” “draws a line . . . between permissible challenges to executive acts and impermissible challenges to the acts of other branches of government.”106×106. Surratt, 797 F.3d at 257. The focus on “the act of physically confining or restraining an individual” rather than on the legal basis of such restraint indicates that § 2241 is reserved for challenges to executive actions, such as those of the Bureau of Prisons, rather than judicial actions.107×107. Id.

This argument is, on its face, internally consistent and convincing, but it neglects those few Supreme Court cases interpreting the savings clause. In Hayman, the Court found that retaining habeas as a last resort made it unnecessary to “reach constitutional questions.”108×108. United States v. Hayman, 342 U.S. 205, 223 (1952). In Swain, the Court held that the savings clause “avoid[ed] any serious question about the constitutionality of the statute.”109×109. Swain v. Pressley, 430 U.S. 372, 381 (1977). The Court viewed the clause, then, as an expansive backstop to § 2255, not as an alternative procedure used to partition different classes of cases into different motions.110×110. See Hayman, 342 U.S. at 223. When Congress passed AEDPA and left the savings clause unchanged, the presumption is that it did so knowing and accepting existing case law on the provision’s meaning.111×111. See Pierce v. Underwood, 487 U.S. 552, 567 (1988) (noting that when “Congress reenact[s] a statute that had . . . been given a consistent judicial interpretation . . . [, s]uch a reenactment, of course, generally includes the settled judicial interpretation”). While this argument — that the savings clause allows only challenges to the conditions of confinement to invoke § 2241 — is appealing in a vacuum, the Court’s established, contrary interpretation and Congress’s ratification of this interpretation complicate the narrative.

Finally, the court in Surratt appealed to Congress’s intention in passing AEDPA. Allowing petitioners access to § 2241 when their claims are not allowed under AEDPA’s restrictions would “thwart almost every one of the careful limits that Congress placed on post-conviction challenges.”112×112. Surratt, 797 F.3d at 259. Allowing any prisoner who does not meet § 2255’s requirements to use a different process would empty these restrictions of any effect.113×113. Id. at 260–61. A parade of horribles follows: a prisoner could file endless collateral attacks,114×114. See id. at 260. he could forum shop by using the motion putting him in the most favorable district,115×115. Id. at 260–61. and finality, “the foremost AEDPA goal,” would be defeated.116×116. Id. at 262.

While these concerns certainly apply to the second-or-successive-motions issue in Surratt, they are not seriously implicated by the statute-of-limitations question. First, as discussed above, it is not clear that Congress carefully placed the statute of limitations on collateral review by federal prisoners; the legislative history reveals nearly no mention of the one-year limit outside its application to habeas corpus for state prisoners.117×117. See supra p. 1094. Second, allowing access to § 2241 would not necessarily leave the § 2255 statute of limitations without any effect. The one-year period for § 2255 could function as a safe harbor during which a prisoner would be unquestionably entitled to move for collateral relief. After proceeding to § 2241, the common law abuse-of-the-writ doctrine would apply, leaving it to the judge to decide whether the delay was reasonable.118×118. See McCleskey v. Zant, 499 U.S. 467, 498 (1991) (discussing the abuse-of-the-writ doctrine). Once the safe harbor of § 2255 expired, a prisoner would have to rely on the judge’s evaluation of his timeliness, an evaluation that surely would be informed by the preference for claims brought within one year.119×119. Cf. Felker v. Turpin, 518 U.S. 651, 663 (1996) (“Whether or not we are bound by these restrictions [on second or successive petitions], they certainly inform our consideration of original habeas petitions.”). Regarding finality, late motions do not pose the same dangers as repeated motions. Constantly reopening a case certainly does require courts to “continually marshal resources in order to keep in prison defendants” who have been properly convicted,120×120. Surratt, 797 F.3d at 263 (quoting Whiteside v. United States, 775 F.3d 180, 186 (4th Cir. 2014)). and limiting prisoners to one collateral attack certainly does encourage them to “exercise greater diligence and invoke whatever rights they may have early on.”121×121. Id. (citing Wainwright v. Sykes, 433 U.S. 72, 90 (1977)). But a late petition that does not run afoul of the abuse-of-the-writ doctrine does not impose substantially greater burdens on the judicial system. Furthermore, prisoners do not lack the motivation to challenge their incarceration early; they are already in prison and, presumably, would prefer to serve less time on a sentence they believe to be vulnerable.

It is beyond dispute that lower courts have upheld § 2255’s statute of limitations and found that the Constitution does not require access to § 2241 once the statute of limitations has expired. As this Note has shown, however, the reasoning leading to those conclusions is not required by existing Supreme Court precedent.

C. A Statutory Approach

Even so, the Supreme Court may not embrace the constitutional arguments presented above to find that the statute of limitations violates the Suspension Clause. If the Court declines to reach this holding directly, it should invoke the canon of constitutional avoidance to interpret § 2255 as allowing access to § 2241 once the statute of limitations has expired. This canon holds that, when the validity of a congressional statute is called into question, “it is a cardinal principle that [courts] will first ascertain whether a construction of the statute is fairly possible by which the question may be avoided.”122×122. Ashwander v. Tenn. Valley Auth., 297 U.S. 288, 348 (1936) (Brandeis, J., concurring) (quoting Crowell v. Benson, 285 U.S. 22, 62 (1932)). In light of the serious reasons presented above to doubt the constitutionality of the statute of limitations, the Court should adopt the available interpretation of § 2255 that avoids the constitutional question.

By its own terms, § 2255 applies a statute of limitations only to “a motion under this section.”123×123. 28 U.S.C. § 2255(f) (2012). By comparison, § 2244’s statute of limitations for federal review of state convictions applies to “an application for a writ of habeas corpus by a person in custody pursuant to the judgment of a State court.”124×124. Id. § 2244(d)(1). Section 2255, then, does not purport to place a statute of limitations on federal habeas corpus petitions for federal prisoners, as § 2244 does for state prisoners, but instead to limit the availability of § 2255 motions themselves. Even though § 2255 has essentially stepped in for federal habeas corpus, it is not itself habeas, nor does it eliminate habeas from the picture. Section 2241 still provides for habeas corpus for prisoners “in custody under or by color of the authority of the United States” and those “in custody in violation of the Constitution or laws or treaties of the United States.”125×125. Id. § 2241(c). Because § 2255 applies to that very class of prisoners — anyone “in custody under sentence of a court established by Act of Congress” (so long as he has an appropriate claim)126×126. Id. § 2255(a). — it would appear that habeas relief is available in these cases, without a time bar, under § 2241.

If a prisoner is equally entitled by statute to apply for either habeas corpus or § 2255 relief, then if one is independently restricted, the other remains available.127×127. Although the circuit courts have confirmed that § 2241 habeas remains available for some prisoners unable to proceed under § 2255, the conditions for satisfying the “inadequate and ineffective” test vary among circuits. See, e.g., United States v. Tyler, 732 F.3d 241, 252–53 (3d Cir. 2013) (allowing recourse to § 2241 habeas when an intervening change in law made conduct noncriminal); Abernathy v. Wandes, 713 F.3d 538, 541 (10th Cir. 2013) (holding that § 2255 is inadequate and ineffective only if the petitioner could not have tested the claim in a § 2255 motion); Wooten v. Cauley, 677 F.3d 303, 307 (6th Cir. 2012) (requiring a claim of actual innocence to sustain a § 2241 application); Darby v. Hawk-Sawyer, 405 F.3d 942, 945 (11th Cir. 2005) (holding that § 2255(e) allows § 2241 applications only when the claim is based on a Supreme Court decision that can be retroactively applied, when that decision shows the offense was nonexistent, and when circuit precedent squarely foreclosed the claim when it otherwise should have been made). The only obstacle to filing a § 2241 petition once the § 2255 statute of limitations has run is § 2255(e)’s savings clause. The initial purpose of this clause was to ensure that the administrative benefits of the original, habeas-equivalent § 2255 over § 2241 habeas would be realized,128×128. See supra section I.A, pp. 1091–92. while still allowing for access to § 2241 habeas when necessary. After AEDPA, its effect shifted from preserving effective habeas in the face of statutory tweaks to ensuring that prisoners could not access traditional habeas without confronting increased statutory limitations on § 2255. Although all circuit courts to confront the question have held that a procedural barrier to filing a § 2255 motion, such as the statute of limitations, would not suffice to make § 2255 inadequate and ineffective,129×129. See, e.g., supra note 127. the Supreme Court has not squarely decided this issue.

Despite appearances, instating such residual habeas protection would not hollow out Congress’s intention of promoting prompt and efficient collateral review. This residual habeas, although it would have no strict statute of limitations, would still be subject to the equitable doctrine of laches, which warrants dismissal of petitions brought with unreasonable delay and causing prejudice to the government’s ability to respond.130×130. See Brecht v. Abrahamson, 507 U.S. 619, 637 (1993) (“[T]here is no statute of limitations governing federal habeas, and the only laches recognized is that which affects the State’s ability to defend against the claims raised on habeas . . . .”); Baxter v. Estelle, 614 F.2d 1030, 1035 (5th Cir. 1980) (“Laches requires not only unreasonable delay but also that the delay work to the detriment of the other party.”). Section 2255 motions would be accepted and evaluated, as a matter of right, within one year; unreasonably delayed § 2241 petitions could be summarily dismissed. In the absence of a reason requiring a late petition, a prudent prisoner would file a § 2255 motion to avoid the danger that his delay be deemed unreasonable. When there is a good reason, even if it is not sufficiently “extraordinary” to trigger equitable tolling, the petitioner may still exercise his right to one bite at the apple of habeas corpus.

For one example of a case where this interpretation would make a difference, imagine an illiterate prisoner who cannot afford to hire a lawyer until more than one year after his conviction becomes final. In most jurisdictions, illiteracy is not sufficient to invoke equitable tolling of the statute of limitations.131×131. See, e.g., DeLeon v. Fla. Dep’t of Corr., 470 F. App’x 732, 734 (11th Cir. 2012) (“An inability to understand English does not constitute extraordinary circumstances justifying equitable tolling.”); Felder v. Johnson, 204 F.3d 168, 171–72 (5th Cir. 2000) (finding ignorance of the law, lack of knowledge of filing deadlines, pro se status, illiteracy, deafness, and lack of legal training do not support equitable tolling of statute of limitations). Nor is there a right to counsel in collateral proceedings.132×132. See Johnson v. Avery, 393 U.S. 483, 488 (1969) (“It has not been held that there is any general obligation of the courts, state or federal, to appoint counsel for prisoners who indicate, without more, that they wish to seek post-conviction relief.”); Hooks v. Wainwright, 775 F.2d 1433, 1438 (11th Cir. 1985) (“[T]here is no automatic constitutional right to representation in a federal habeas corpus proceeding.”). Currently, this prisoner, however diligent he may be, will not be able to meaningfully access habeas corpus. His late petition will be denied as time-barred. With residual habeas, this unfortunate prisoner would be able to file his single § 2241 petition as soon as possible, while a less sympathetic prisoner who delayed his petition to wait for a prosecution witness to die would have his petition dismissed. This result would serve Congress’s aims — preventing gamesmanship and unnecessary delay in reaching finality — without unnecessarily restricting access to habeas in appropriate cases. Where the canon of constitutional avoidance intersects with serving the ends of justice, such an approach should be preferred.


It is settled law that § 2255 is not itself habeas corpus, but that federal prisoners must proceed through § 2255 motions unless they can show that the remedy is inadequate and ineffective. It is clear from the statutes that § 2254 provides a statute of limitations for “habeas corpus” applications for state prisoners and that § 2255 provides a statute of limitations for “motion[s] under this section.” It is also clear that habeas corpus for federal prisoners, both at common law and in § 2241, does not have any statute of limitations.

It is not clear, however, whether AEDPA’s statute of limitations is constitutional, as the Court has never addressed this question. Nor is it clear whether federal prisoners may have recourse to § 2241 habeas after their § 2255 statute-of-limitations period has run. However, based on the text of § 2255, which appears to carefully avoid placing a statute of limitations on habeas corpus for federal prisoners to avoid potential Suspension Clause issues, and based on the principles expressed in Boumediene, the Supreme Court’s most probing exploration of Suspension Clause issues, it appears that either federal habeas is lurking behind § 2255, ready to protect those whose statutes have run, or AEDPA’s statute of limitations is unconstitutional.

It is unlikely that the Court will soon declare that prisoners have a right to habeas corpus despite any statutes of limitations, and it seems farfetched that the Court will soon take a case in which it finds AEDPA’s statute of limitations to be a constitutional violation. Crime is too political an issue, courts are too concerned about the state of their dockets, and the only people who stand to benefit from more habeas corpus can’t vote in most states. Regardless, if federal habeas corpus truly does have some life outside of the suffocating grasp of AEDPA, there is hope for the future. Perhaps habeas corpus is simply in one of its inevitable historical backswings, biding its time until the forces of liberty march forward and restore it to its status as the Great Writ rather than as dead law.