EAJA Fee-Shifting Applies to § 2241 Immigration-Detention Habeas Petitions as “Any Civil Action”

Case: Adewumi Abioye v. Warden Moshannon Valley Correctional Center (consolidated with Adolph Michelin v. Warden Moshannon Valley Correctional Center)
Court: United States Court of Appeals for the Third Circuit
Date: February 2, 2026
Holding (core): A petition for a writ of habeas corpus under 28 U.S.C. § 2241 challenging civil immigration detention is an EAJA “civil action,” so 28 U.S.C. § 2412(d)(1)(A) waives sovereign immunity for attorneys’ fees and costs (absent substantial justification or special circumstances).

1. Introduction

These consolidated appeals arise from prolonged immigration detentions at the Moshannon Valley facility. Adolph “Lee” Michelin and Adewumi Abioye each filed habeas petitions under 28 U.S.C. § 2241, alleging that lengthy detention without an individualized bond hearing violated the Fifth Amendment’s Due Process Clause. Both prevailed and obtained bond hearings; both then sought attorneys’ fees and costs under the Equal Access to Justice Act (“EAJA”), 28 U.S.C. § 2412(d)(1)(A).

The Government appealed the fee awards, framing a threshold jurisdictional question: because the EAJA is a waiver of sovereign immunity, does its authorization of fees in “any civil action (other than cases sounding in tort)” extend to habeas petitions challenging immigration detention? In Abioye’s case, the Government also challenged the finding that its position was not “substantially justified.”

The Third Circuit’s precedential opinion answers the fee-eligibility question decisively for this Circuit: immigration-detention habeas under § 2241 fits comfortably within the statutory phrase “any civil action,” and the sovereign-immunity canon does not permit narrowing that text where ordinary interpretive tools leave no ambiguity.

2. Summary of the Opinion

  • EAJA coverage: The Court holds that the EAJA unambiguously covers § 2241 habeas challenges to immigration detention because habeas is a civil action historically and doctrinally, and the EAJA’s phrase “any civil action (other than cases sounding in tort)” is categorical.
  • Sovereign immunity methodology: The Court reiterates that sovereign-immunity clear-statement rules operate only after exhausting ordinary textual and contextual interpretation; they do not justify importing implied exclusions into otherwise clear statutory language.
  • Substantial justification (Abioye): The Court affirms the fee award to Abioye, holding the District Court did not abuse its discretion in finding the Government lacked a reasonable basis to oppose an individualized bond hearing after Abioye had been detained over 16 months with no end in sight.

3. Analysis

3.1. Precedents Cited (and How They Shape the Decision)

A. Habeas as a “civil action”: deep history and Supreme Court continuity

The opinion anchors its interpretation in the long-settled characterization of habeas as civil—beginning with common-law framing and continuing through Supreme Court doctrine:

  • Holmes v. Jennison (holding it “too plain for argument” that a habeas petition is a civil action under the Judiciary Act of 1789) and Ex parte Tom Tong (habeas enforces the civil right of personal liberty) are invoked to show that habeas’ “civil” label is not modern convenience but inherited structure.
  • Browder v. Dir., Dep't of Corr. of Ill. and Stafford v. Briggs are pivotal because they are close in time to the EAJA’s enactment and state the proposition in unqualified terms (“well settled”; “Habeas corpus is a civil action.”). They help the Court answer the key “plain meaning in 1980” inquiry.
  • Banister v. Davis is used to underscore that the Supreme Court continues to describe habeas proceedings as civil in nature.
  • The opinion also uses Ex parte Yerger and Kurtz v. Moffitt to show that habeas has been treated as a civil proceeding even when deployed against executive detention—supporting the conclusion that a civil immigration-detention habeas petition is, if anything, the paradigmatic “civil” habeas.

B. When “civil action” does not automatically include habeas: confined, statute-specific carve-outs

The Government relied on cases recognizing that habeas is “unique” and sometimes treated differently from ordinary civil litigation. The Third Circuit distinguishes them sharply:

  • Schlanger v. Seamans (venue/service provision not extending to habeas) is treated as a statute-specific reading that nevertheless concedes habeas is “technically ‘civil’.” The Third Circuit emphasizes Schlanger’s reliance on legislative history—an interpretive move the present opinion declines to replicate, especially in sovereign-immunity contexts.
  • Harris v. Nelson (discussing habeas’ procedural uniqueness and the “gross and inexact” “civil action” label) is read not as repudiation of habeas’ civil status, but as recognition that habeas is civil while not “automatically subject to all the rules governing ordinary civil actions.”
  • Santana v. United States (PLRA “civil action” not encompassing habeas) is treated as an example of a context where courts departed from plain meaning using legislative history and structural inferences. The Third Circuit notes that modern textualism—particularly post-Food Mktg. Inst. v. Argus Leader Media—does not follow that approach, and sovereign-immunity doctrine reinforces the focus on statutory text.
  • Callwood v. Enow is used to demonstrate the Third Circuit’s own recognition of habeas as civil, even when described as “somewhat of a hybrid.”

C. Sovereign immunity: clear statement, but only after ordinary tools are used

  • FAA v. Cooper and Lane v. Peña supply the general rule: waivers must be “unequivocally expressed,” ambiguities resolved in favor of immunity.
  • Dep't of Agric. Rural Dev. Rural Hous. Serv. v. Kirtz and Richlin Sec. Serv. Co. v. Chertoff provide the crucial methodological refinement: courts first use ordinary tools of interpretation; if “no ambiguity left,” they do not “resort to the sovereign immunity canon.”
  • Ardestani v. INS confirms EAJA is a waiver, while also supporting the opinion’s caution that broad EAJA “purpose” cannot expand text beyond what Congress clearly enacted.

D. “Any civil action” and the importance of textual context

  • Conn. Nat'l Bank v. Germain supplies the “cardinal canon” that legislatures mean what they say.
  • Ex parte Collett is used for its direct statement that the “reach of ‘any civil action’ is unmistakable.”
  • United States v. Johnson provides the inference from express exceptions: by excluding only torts, Congress signals that other exclusions should not be implied.
  • To show harmony with the broader U.S. Code, the Court cites examples where “any civil action” reaches habeas unless expressly excluded: Boumediene v. Bush (transfer under § 1404(a)), German Santos v. Warden Pike Cnty. Corr. Facility (jurisdiction under § 1331), and statutory examples like 28 U.S.C. § 1914(a) and 18 U.S.C. § 3626(g) (which expressly excludes habeas from PLRA’s prison-conditions definition—confirming that Congress knows how to carve habeas out when it wants to).

E. Circuit split on EAJA-habeas fees in immigration detention

The Court canvasses conflicting appellate approaches:

  • O'Brien v. Moore (Fourth Circuit) and Obando-Segura v. Garland (Fourth Circuit) are rejected as unpersuasive in this context, chiefly because they: (i) underweight “any” and the tort exception, (ii) build on a “hybrid” premise tied to criminal custody, and (iii) do not translate well to civil immigration detention.
  • Barco v. Witte (Fifth Circuit) is similarly rejected for relying on the “hybrid” framing without explaining why civil immigration detention imports criminal features sufficient to overcome the EAJA’s breadth.
  • In re Hill (Ninth Circuit) and Vacchio v. Ashcroft (Second Circuit) reach the same outcome but rely on purpose/legislative history in ways the Third Circuit declines to emulate in sovereign-immunity cases.
  • Daley v. Ceja (Tenth Circuit) is treated as the most illuminating aligned approach (text, history), though the Third Circuit notes it does not itself rely on EAJA purpose.

F. Abioye’s “substantial justification” analysis: due process limits on prolonged mandatory detention

  • Demore v. Kim establishes that 8 U.S.C. § 1226(c) is not facially unconstitutional but does so on an assumption of relatively short detention. This assumption becomes a measuring stick.
  • German Santos v. Warden Pike Cnty. Corr. Facility provides the Third Circuit’s as-applied due process framework for prolonged § 1226(c) detention, including the four nonexhaustive factors and the principle that “when detention becomes unreasonable, the Due Process Clause demands a hearing.”
  • Chavez-Alvarez v. Warden York Cnty. Prison (noting it was abrogated in part on other grounds by Jennings v. Rodriguez) supports the time-based skepticism that grows as detention exceeds typical durations.
  • For the EAJA standard itself, the Court relies on Pierce v. Underwood (reasonable person test), Hanover Potato Prods., Inc. v. Shalala (Government bears burden), Cruz v. Comm'r of Soc. Sec. and Morgan v. Perry (three-part substantial-justification showing), and Johnson v. Gonzales (immigration cases require justification for both underlying conduct and litigation position).

3.2. Legal Reasoning

A. The interpretive sequence: text first, then (only if needed) the sovereign-immunity canon

The Court’s method is itself a doctrinal statement: it treats sovereign-immunity strict construction as a tie-breaker, not a thumb on the scale that creates ambiguity. It therefore asks:

  • What did “civil action” mean when Congress enacted EAJA in 1980?
  • How does “civil action” operate in its immediate context: “any civil action (other than cases sounding in tort)”?
  • How does this phrase fit with the broader “corpus juris,” where Congress sometimes expressly excludes habeas when it intends to?

On each level—plain meaning, term-of-art meaning, rules context, common-law pedigree, and statutory context—the Court finds the same answer: immigration habeas under § 2241 is a civil action, and EAJA’s “any” plus express tort exception leave no plausible room for an implied habeas exception.

B. The “hybrid” argument is defused by the civil character of immigration custody

The Government’s key rhetorical move is to label habeas “unique” and “hybrid.” The Court responds with a classification point:

  • Where habeas is said to be “hybrid,” courts generally mean hybrid of civil and criminal because it may collaterally review criminal judgments.
  • These cases are not collateral attacks on criminal custody; they challenge civil immigration detention (citing Zadvydas v. Davis and INS v. Lopez-Mendoza).
  • Thus, whatever “hybrid” might mean in other contexts, it cannot justify treating an immigration-detention habeas petition as anything other than civil.

C. “Any” and the express tort exception do the heavy lifting

Even accepting that some statutes have treated “civil action” as not automatically encompassing habeas, the Court concludes the EAJA text forecloses that move:

  • “Any” signals categorical breadth—civil actions “of whatever kind.” Reading “any” as surplusage is disfavored.
  • Expressio unius: by excluding torts expressly, Congress implies it did not silently exclude other categories (like habeas).
  • Code-wide coherence: in other statutes, Congress either includes habeas within “any civil action” or expressly excludes it; an implied exclusion here would make EAJA anomalous.

D. Abioye: why opposing a bond hearing after 16+ months was not substantially justified

Applying German Santos v. Warden Pike Cnty. Corr. Facility, the Court treats two factors as decisive:

  • Duration (the “most important factor”): detention “becomes more and more suspect after five months,” and became unreasonable “sometime between six months and one year.” At 16+ months (and 18+ by release), the Government’s position lacked a reasonable basis in law.
  • Likelihood of continuation: Abioye’s pending Fourth Circuit review made continued detention likely, strengthening the unreasonableness finding.

The Court acknowledges other factors were less one-sided (reasons for delay; conditions of confinement), but holds they could not rationally outweigh the first two on this record. It also rejects procedural objections (e.g., alleged bright-line rule; failure to “recap” the earlier merits analysis; law-of-the-case arguments based on an offhand “close call” remark).

3.3. Impact

A. Immediate doctrinal effect in the Third Circuit

  • Fee eligibility is clarified: Prevailing immigration detainees who win § 2241 habeas relief (e.g., bond hearings for prolonged detention) may recover EAJA fees if the Government’s position is not substantially justified.
  • Government litigation incentives shift: The Government must now internalize the cost risk of opposing prolonged-detention due process claims in this Circuit—particularly where detention durations exceed the benchmark ranges discussed in German Santos v. Warden Pike Cnty. Corr. Facility.
  • Methodology matters: The decision strengthens a text-and-context approach to sovereign immunity waivers: the sovereign-immunity canon does not license implied narrowing when “any civil action (other than cases sounding in tort)” reads broadly on ordinary tools.

B. Interaction with the circuit split

The opinion deepens an existing split: the Fourth and Fifth Circuits have resisted EAJA coverage for immigration habeas, while the Second, Ninth, Tenth, and now Third Circuits endorse it (though via different reasoning). This division raises the prospect of eventual Supreme Court review, especially because fee-shifting regimes materially affect the volume and quality of detention litigation.

C. Practical consequences for detention litigation and administration

  • More meritorious petitions can be brought: EAJA reduces the financial barrier for detainees to challenge prolonged custody, aligning with habeas’ historic function as a tool to require the executive to justify continued detention with adequate process.
  • Stronger screening of positions: Agencies and government counsel may more carefully evaluate whether to contest bond-hearing demands when detention has become prolonged under Third Circuit precedent.
  • Potential for faster administrative resolution: The cost of defending unjustified positions may incentivize earlier bond hearings or stipulated relief in prolonged-detention cases.

4. Complex Concepts Simplified

4.1. What is EAJA, and why does “civil action” matter?

The EAJA is a fee-shifting statute that can require the United States to pay the prevailing party’s attorneys’ fees and costs in covered cases—but only where Congress has clearly allowed it (because the Government is ordinarily immune from monetary awards). EAJA’s coverage hinges on whether the case is “any civil action (other than cases sounding in tort).”

4.2. Sovereign immunity and the “clear statement” rule

“Sovereign immunity” means you generally cannot obtain money from the United States unless Congress clearly authorizes it. Courts therefore require waivers to be “unequivocally expressed.” This case clarifies the order of operations: first interpret the statute normally (text, context, legal meaning); only if ambiguity remains does the court construe the statute in favor of immunity.

4.3. Habeas under § 2241 in immigration detention

A § 2241 habeas petition is a federal-court action challenging unlawful custody. In the immigration context, detainees often use § 2241 to claim that prolonged detention without an individualized bond hearing violates due process.

4.4. “Substantially justified” in EAJA

Even if the private party wins, the Government does not pay fees if its position was “substantially justified”—meaning reasonably grounded in fact and law. In immigration cases, the Government must justify both (1) the underlying detention decision and (2) its litigation stance defending that detention.

5. Conclusion

The Third Circuit’s precedential rule is straightforward and consequential: an immigration detainee’s § 2241 habeas petition is a covered EAJA “civil action,” because habeas has long been civil in nature and the EAJA’s text—“any civil action (other than cases sounding in tort)”—contains no plausible implicit habeas exclusion. The Court reinforces an interpretive discipline especially important in sovereign-immunity cases: ordinary textual analysis comes first; the sovereign-immunity canon applies only if ambiguity persists.

On the facts, the Court also signals how fee exposure will attach in prolonged-detention litigation: once detention moves well beyond the timeframes contemplated in Demore v. Kim and becomes unreasonable under German Santos v. Warden Pike Cnty. Corr. Facility, the Government risks not only losing on the merits but paying fees for insisting on continued detention without a bond hearing. In that respect, the decision aligns fee-shifting with habeas’ historic mission—compelling reasoned justification and adequate process before the state may continue to confine.