No. 24-1191

HABEAS AND IMMIGRATION LAW SCHOLARS; AMICA CENTER FOR IMMIGRANT RIGHTS; THE NATIONAL IMMIGRATION PROJECT OF THE NATIONAL LAWYERS GUILD; ROBERT F.

Page 2Appeal from the United States District Court for the District of Colorado (D.C. No. 1:22-CV-03043-RM)

Kyle W. Brenton, Assistant United States Attorney, Denver, Colorado (Matthew T. Kirsch, Acting United States Attorney, Denver, Colorado; Brian M. Boynton, Principal Deputy Assistant Attorney General, William C. Peachey, Director, and Alexa S. White, Trial Attorney, Civil Division, Department of Justice, Washington, D.C., with him on the briefs), for Respondents-Appellants.

John V. Hoover of Arnold & Porter Kaye Scholer LLP, Washington, D.C. (Sean A. Mirski, R. Stanton Jones, Andrew T. Tutt, and Casey Corcoran of Arnold & Porter Kaye Scholer LLP, Washington, D.C.; William T. Sharon and Nicole L. Masiello of Arnold & Porter Kaye Scholer LLP, New York, New York; Laura P. Lunn and Laura L. Shoaps of Rocky Mountain Immigrant Advocacy Network, Westminster, Colorado, with him on the brief), for Petitioner-Appellee.

Jeremy M. Christiansen, Matt Gregory, and Hayley N. Lawrence of Gibson, Dunn & Crutcher LLP, Washington, D.C., filed an amicus curiae brief for Habeas and Immigration Law Scholars.

Peter Cameron Alfredson and Amelia Christine Dagen of Amica Center for Immigrant Rights, Washington, D.C., filed an amicus curiae brief for Amica Center for Immigrant Rights.

Charity E. Lee and Sarah E. Libowsky of Cleary Gottlieb Steen & Hamilton LLP, New York, New York, filed an amicus curiae brief for The National Page 3Immigration Project of the National Lawyers Guild, Robert F. Kennedy Human Rights, The Center for Constitutional Rights, The National Immigrant Justice Center, The American Immigration Council, and The American Immigration Lawyers Association.

Before HOLMES , Chief Judge, KELLY , and FEDERICO , Circuit Judges.

FEDERICO , Circuit Judge.

After Eva Daley successfully petitioned for habeas relief from her immigration detention, she moved for attorneys' fees under the Equal Access to Justice Act (EAJA), 28 U.S.C. § 2412. The district court granted Daley’s fee motion. Respondents (the Government) now appeal that order. The Government's appeal presents a single question 1 : Does the EAJA authorize – and therefore waive sovereign immunity for – the award of fees in habeas actions challenging immigration detention?

Answering that question is a matter of statutory interpretation. The EAJA authorizes fees in “any civil action.” 28 U.S.C. § 2412(d)(1)(A). We must therefore decide if habeas actions challenging immigration detention Page 4are 'civil actions.' We conclude they are, so we exercise our jurisdiction under 28 U.S.C. § 1291 to affirm.

Daley is a Guatemalan national. At the age of twelve, she entered the United States without inspection or admission. Later, as an adult, Daley was convicted of second-degree murder in California state court. Although her murder conviction was ultimately vacated and substituted with a reduced conviction for assault with a deadly weapon, Daley served nearly fifteen years in prison.

While in state custody, Daley interviewed with U.S. Immigration and Customs Enforcement (ICE), during which she declined a stipulated order of removal. In the absence of a stipulated order, ICE decided to immediately detain Daley upon her release from state prison on November 1, 2021. ICE then transferred her to a facility in Aurora, Colorado. Roughly three months later, Daley applied for asylum, withholding of removal, and protection under the Convention Against Torture.

On November 22, 2022, having spent over a year in immigration detention without receiving a timeline for release or a bond hearing, Daley filed a petition for a writ of habeas corpus under 28 U.S.C. § 2241. Through her petition, Daley sought to challenge her detention. The district court granted habeas relief and ordered the Government to provide an Page 5individualized bond hearing to Daley. On January 25, 2023, an immigration judge held a bond hearing as ordered and released Daley on a $1,500 bond. Daley had spent 450 days in ICE custody.

Subsequently, Daley moved for attorneys’ fees under the EAJA. The district court ruled that Daley was entitled to fees and awarded $18,553.92. This appeal followed.

''[W]e review the decision to award attorney fees, and the amount awarded, for abuse of discretion,’ though any legal analysis underlying the award is reviewed de novo.' First Am. Title Ins. Co. v. Nw. Title Ins. Agency , 906 F.3d 884, 900 (10th Cir. 2018) (quoting Xlear, Inc. v. Focus Nutrition, LLC , 893 F.3d 1227, 1233 (10th Cir. 2018)). Here, our review is de novo because only interpretation of the EAJA, a legal question, is at issue.

Unless waived, sovereign immunity generally shields the Government, including agencies and officers in their official capacity, from suit. Normandy Apartments, Ltd. v. U.S. Dep’t of Hous. & Urb. Dev. , 554 F.3d 1290, 1295 (10th Cir. 2009). This includes shielding the Government from claims for attorneys’ fees. Adamson v. Bowen , 855 F.2d 668, 670 (10th Cir. 1988) (citing Ruckelshaus v. Sierra Club , 463 U.S. 680, 685 (1983)). Congress waived that immunity for certain types of fees when it enacted Page 6the EAJA. Ardestani v. INS , 502 U.S. 129, 137 (1991). The harder question, and the one we must confront in this appeal, is how to determine which fees fall within the EAJA’s scope.

When interpreting the scope of a sovereign immunity waiver, we strictly construe the relevant statutory language in favor of maintaining immunity. Under this canon of interpretation – sometimes referred to as a clear statement rule – waivers of immunity encompass only those claims for which there is 'an unmistakable statutory expression of congressional intent to waive the Government’s immunity.” FAA v. Cooper , 566 U.S. 284, 291 (2012); see also Lane v. Peña , 518 U.S. 187, 192 (1996) (similar). An unmistakable intent need not be expressed using “magic words” or “in any particular way.” Dep’t of Agric. Rural Dev. Rural Hous. Serv. v. Kirtz , 601 U.S. 42, 48 (2024) (quoting Cooper , 566 U.S. at 291). No matter how Congress makes its intent clear, so long as its intent is clear, sovereign immunity is waived. But if there is any ambiguity in the text of a waiver, meaning that there is a “plausible” interpretation “preserv[ing] sovereign immunity,' we must adopt that 'plausible' interpretation. Lac du Flambeau Band of Lake Superior Chippewa Indians v. Coughlin , 599 U.S. 382, 388 (2023) (quoting Cooper , 566 U.S. at 290).

That said, we may not resort to the sovereign immunity canon at the first sign of any potential ambiguity in the text as this would abdicate our Page 7responsibility to interpret the statutes in front of us. Cf. Loper Bright Enters. v. Raimondo , 603 U.S. 369, 385 (2024). Before turning to the canon, we must exhaust all “traditional tools of statutory construction.” Cooper , 566 U.S. at 291 (quoting Richlin Sec. Serv. Co. v. Chertoff , 553 U.S. 571, 589 (2008)). After scrutinizing the language 'in light of traditional interpretive tools,” we may discover that what originally appeared ambiguous has a clear meaning. See id. Put differently, we cannot reflexively apply the sovereign immunity canon to adopt an immunitypreserving interpretation simply because that interpretation appears plausible on preliminary review. We apply the canon only after analyzing the statutory language and finding it ambiguous.

With these principles in mind, we now proceed to interpret the EAJA.

The statutory text is the core of our inquiry, so we start there. See Kirtz , 601 U.S. at 49. The EAJA authorizes fees against the Government “in any civil action (other than cases sounding in tort), including proceedings for judicial review of agency action,' subject to exceptions not applicable here. 28 U.S.C. § 2412(d)(1)(A). The key question we must answer is whether habeas petitions challenging immigration detention are 'civil actions.” Since the EAJA does not define “civil action,” we look to the term’s ordinary meaning at the time of the EAJA’s enactment. M.S. v. Premera Page 8Blue Cross , 118 F.4th 1248, 1266 (10th Cir. 2024) (citing Food Mktg. Inst. v. Argus Leader Media , 588 U.S. 427, 433–34 (2019)).

When Congress passed the EAJA in 1980, the notion of a “civil action” had already existed for at least two hundred years. See Pub. L. 96-481 tit. II, 94 Stat. 2321, 2325–30 (1980). “Civil action” was a common law “term[] of art in which [was] accumulated the legal tradition and meaning of centuries of practice.' Sekhar v. United States , 570 U.S. 729, 733 (2013) (quoting Morissette v. United States , 342 U.S. 246, 263 (1952)). Congress’s decision to use “civil action” in the EAJA therefore brought the “old soil” of that term's history into the statute. SEC v. Jarkesy , 603 U.S. 109, 125 (2024) (quoting United States v. Hansen , 599 U.S. 762, 778 (2023)). So as “a settled principle of interpretation,' we must presume that Congress incorporated the common law understanding of “civil action” into the EAJA. Sekhar , 570 U.S. at 732.

The concept of a civil action has its roots in the English common law tradition. Common law recognized civil actions as a distinct category of legal proceeding based on the division of wrongs into “private wrongs” and “public wrongs.” 3 William Blackstone, Commentaries *2. Private wrongs stemmed from the “infringement or privation of the private or civil rights belonging to individuals . . . and [were] thereupon frequently termed civil injuries .' Id. To remedy a civil injury, individuals made “application to the[]Page 9 courts of justice; that is, by civil suit or action .” Id. at *3 (emphasis added). By contrast, public wrongs were those stemming from 'a breach and violation of public rights and duties, which affect the whole community . . . and are distinguished by the harsher appellation of crimes and misdemeanors .” Id. at *2; see also 4 Blackstone, supra , at *5 (similar).

Of course, “every public offence is also a private wrong” because “it affects the individual, and it likewise affects the community.” 4 Blackstone, supra , at *5. Yet, commentators like Blackstone still considered the “distinction of crimes from civil injuries [to be] very apparent.” Id. at *6. Blackstone illustrated this distinction through the example of battery. Battery has a criminal (public) component since it 'disturb[s] the public peace.” Id. But it also has a separate civil (private) component because the injured individual could seek a “private remedy . . . for the injury, which he in particular sustains, and recover a civil satisfaction in damages.” Id.

In sum, English common law defined “civil actions” as actions that seek redress for an individual's private injury, or vindication of an individual's private right. And to determine if an action satisfies that definition, the common law focused on the remedy being sought, not the conduct being challenged.

American courts picked up on this thread and adopted it. For example, the Supreme Court recognized that civil proceedings are those “which affect Page 10private rights” in multiple contexts. Watson v. Mercer , 33 U.S. (8 Pet.) 88, 110 (1834); see also Doyle v. London Guarantee & Accident Co. , 204 U.S. 599, 605 (1907) (civil contempt is distinguished from criminal contempt because civil contempt is 'chiefly concerned' with protecting individuals' “private rights and remedies”); Bessette v. W.B. Conkey Co. , 194 U.S. 324, 328 (1904) (similar); Bradlie v. Maryland Ins. Co. , 37 U.S. (12 Pet.) 378, 402 (1838) ('proceedings of a [] civil nature' are those 'to enforce private rights”).

So too did other courts around the country. The Circuit Court of the District of Columbia observed that a “civil action” is one “to try a private right.” United States v. Columbian Ins. Co. of Alexandria , 25 F. Cas. 585, 588 (C.C.D.C. 1821) (No. 14,840). The Kentucky Court of Appeals wrote, “In general, the enforcement of a private right, or the redress of a private wrong, is attainable, in a court of justice, only through the instrumentality of a civil action .' Jones v. Hoffman , 57 Ky. (18 B. Mon.) 656, 656 (1857). And the Minnesota Supreme Court noted that “the ordinary civil action” is one that seeks 'to redress private wrongs or enforce private rights.' True v.

Page 11True , 6 Minn. 458, 465 (1861).² Some states even codified this understanding, expressly defining civil actions as those based on private rights or wrongs. E.g. , Idaho Const. art. V, § 1 (adopted 1889); Cal. Civ. Proc. Code § 307 (enacted 1872); Civil Practice Act, § 1, 1871 Mont. Laws 28; 3 N.C. Const. of 1868, art. IV, § 1; Ga. Code § 3177 (1861); 1848 N.Y. Laws 497, 510.

Legal dictionaries from the time of the EAJA’s passage also accord with this common law understanding. See Food Mktg. Inst. , 588 U.S. at 434 (consulting contemporary dictionaries). One defined 'civil action' as an “[a]ction brought to enforce, redress, or protect private rights,” encompassing “all types of actions other than criminal proceedings.” Civil Action , Black's Law Dictionary (5th ed. 1979). Another likewise defined “civil action” as one “brought to enforce a civil right” or “for the enforcement or protection of a private right.” Civil Action , Ballentine’s Law DictionaryPage 12 (3d ed. 1969). It, too, observed that a civil action 'comprehend[ed] every conceivable cause of action . . . except such as are criminal.” Id. And a third did the same, defining “civil action” as one “brought to enforce a civil right or obtain redress for its violation.” Civil Action , Radin Law Dictionary (2d ed. 1970).

Given the fact that multiple sources – commentators, courts, statutes, and dictionaries - across such a long period of time have consistently defined “civil action,” we construe the EAJA in line with that definition. As used in the EAJA, “civil action” unambiguously refers to any legal action brought to enforce a private or civil right, or to redress a private wrong.

Now that we have defined “civil action,” our next task is to apply that definition to habeas actions challenging immigration detention.

Again, common law tradition provides a starting point. See Sekhar , 570 U.S. at 732. The common law recognizes a right of “personal liberty,” defined as the right to be free from “imprisonment or restraint, unless by due course of law.” 1 Blackstone, supra , at *130. That right is a private one because it attaches to an individual’s person rather than to the public at large. Id. at *118. And since a wrong is 'nothing else but a privation of right,” 3 Blackstone, supra , at *2, the violation of personal liberty through Page 13unlawful detention is a private wrong. Id. at *127. Any action that seeks to redress this private wrong is a civil action under our construction of the EAJA. Following this logic, the writ of habeas corpus is a civil action: it is the common law mechanism used to challenge the legality of detention and therefore to redress unlawful detentions. Id. at *133; see also 1 Blackstone, supra , at *131.

The Supreme Court has said as much. A full century before Congress passed the EAJA, the Court explained that “[t]he writ of habeas corpus is the remedy which the law gives for the enforcement of the civil right of personal liberty.” Ex parte Tom Tong , 108 U.S. 556, 559 (1883). For that reason, '[s]uch a proceeding . . . is, in [the Court's] opinion, a civil proceeding.' Id. at 560. In the years that followed, the Court repeatedly and consistently affirmed this holding. E.g. , Santa Clara Pueblo v. Martinez , 436 U.S. 49, 60 (1978) ('Congress clearly has power to authorize civil actions against tribal officers, and has done so with respect to habeas corpus relief[.]”); Fay v. Noia , 372 U.S. 391, 423 (1963) (acknowledging 'the traditional characterization of the writ of habeas corpus as an original . . . civil remedy'), overruled on other grounds by Coleman v. Thompson , 501 U.S. 722 (1991); Riddle v. Dyche , 262 U.S. 333, 335–36 (1923) (“The writ of habeas corpus is . . . an independent civil suit[.]”); Borrego v. Cunningham , 164 U.S. 612, 618 (1896) (“[A] proceeding in habeas corpus [is] a civil . . .

Page 14proceeding[.]”); Cross v. Burke , 146 U.S. 82, 88 (1892) ('[A] proceeding in habeas corpus is a civil . . . proceeding.').⁴ These decisions were so extensive that, by the time the EAJA was enacted in 1980, it was “well settled” under Supreme Court precedent 'that habeas corpus is a civil proceeding.' Browder v. Dir., Dep’t of Corr. of Ill. , 434 U.S. 257, 269 (1978).

Moreover, every federal circuit court existing at the time of the EAJA's passage had held that habeas was civil.⁵ See, e.g. , Mercado v. United States , 183 F.2d 486, 487 (1st Cir. 1950) ('[H]abeas corpus [] has always Page 15been regarded as a civil proceeding[.]”); United States ex rel. Meadows v. New York , 426 F.2d 1176, 1183 n.9 (2d Cir. 1970) ('Since habeas corpus proceedings are civil in nature, they are subsumed under the phrase ‘any civil action.’”); United States ex rel. Thomas v. New Jersey , 472 F.2d 735, 741 (3d Cir. 1973) ('Habeas corpus is a civil proceeding.'); Long v. Robinson , 436 F.2d 1116, 1119 (4th Cir. 1971) (describing habeas corpus as a “civil process”); Estep v. United States , 251 F.2d 579, 581 (5th Cir. 1958) (“Habeas corpus petitions . . . are, of course, civil proceedings[.]'); Bowdidge v. Lehman , 252 F.2d 366, 368 (6th Cir. 1958) ('Habeas corpus is a civil proceeding[.]”); United States ex rel. Rebenstorf v. Pate , 417 F.2d 1222, 1225 (7th Cir. 1969) (“[T]he federal habeas corpus remedy[] is civil in nature.”); Burgess v. King , 130 F.2d 761, 762 (8th Cir. 1942) (“Habeas corpus is in its nature a civil rather than a criminal proceeding[.]”); Collins v. Heinze , 217 F.2d 62, 62 (9th Cir. 1954) (“A habeas corpus proceeding is in the nature of a civil action[.]'); Hunter v. Thomas , 173 F.2d 810, 812 (10th Cir. 1949) (“Habeas corpus is a civil proceeding.”); Jackson v. United States , 353 F.2d 862, 865 n.4 (D.C. Cir. 1965) (“Of course, habeas corpus is considered a civil proceeding[.]”).

To be sure, there are some cases suggesting that habeas is not purely civil. For instance, in Harris v. Nelson , the Supreme Court accepted that “habeas corpus proceedings are characterized as ‘civil’” but cast doubt on Page 16the appropriateness of that characterization. 394 U.S. 286, 293–94 (1969). The Harris court criticized the “civil” label as “gross and inexact,” and it opined that habeas 'is unique.' Id. at 293-94. Later in Schlanger v. Seamans , the Court construed the term 'civil action' in an unrelated statute to not include habeas. 401 U.S. at 490 n.4. Both Harris and Schlanger , however, are readily distinguishable.

Harris dealt narrowly with whether the Federal Rules of Civil Procedure governing and authorizing discovery applied to habeas actions. 394 U.S. at 289-90. In answering that question, the Harris court focused on how habeas practice differed from general civil practice . Id. at 294. But practice and procedure offer little insight into whether an action is civil in nature in the sense that it is geared to redressing private wrongs. After all, there are several specialized civil actions that have unique procedures not applicable to the more typical or classic civil action. Admiralty claims are one example because they are subject to supplemental rules that no other civil claims are subject to. See Fed. R. Civ. P., Supp R. for Admiralty.

In fact, the EAJA itself makes clear that its waiver of sovereign immunity extends to actions that do not follow the normal rules of civil procedure. The EAJA expressly includes “proceedings for judicial review of agency action' within the broader umbrella of 'civil action.' 28 U.S.C. § 2412(d)(1)(A). These proceedings include social security appeals. And Page 17social security appeals are governed by unique procedural rules. See Fed. R. Civ. P., Supp. R. for Soc. Sec. The EAJA also extends its waiver to cases brought in the United States Court of Appeals for Veterans Claims. 28 U.S.C. § 2412(d)(2)(F). But that court does not have procedures resembling the Federal Rules of Civil Procedure. See U.S. Vet. App. R., https://perma.cc/FVD2-7BZF (posted Feb. 24, 2025). Statutory context, then, demonstrates that differences in procedure do not inform our assessment of whether habeas is a “civil action” under the EAJA.

Meanwhile, Schlanger construed the meaning of 'civil action' by applying legislative history. 401 U.S. at 490 n.4; see also Stafford v. Briggs , 444 U.S. 527, 542-43 (1980) (discussing Schlanger ). When construing waivers of sovereign immunity, though, we focus on the meaning of the text and cannot attempt to divine Congress's intentions through legislative history. Kirtz , 601 U.S. at 49. In any case, Schlanger construed another statute, not the EAJA, so its analysis of legislative history has no bearing on the question before us now.

More fundamentally, there is still an overwhelming judicial consensus that habeas actions are civil actions even when considering Harris and Schlanger . There need not have been unanimity among the courts for us to find that habeas is unambiguously a 'civil action.' It is enough that “‘ nearly all [] of the' relevant judicial decisions have given a Page 18term or concept a consistent judicial gloss.” Bruesewitz v. Wyeth LLC , 562 U.S. 223, 243 (2011) (emphasis added) (quoting Merck & Co. v. Reynolds, 559 U.S. 633, 659 (2010) (Scalia, J., concurring)). That is true here, where courts have given a 'consistent judicial gloss' to the idea that habeas is civil. Id. We can safely presume that Congress intended to incorporate that judicial gloss into the EAJA. Id. ; see also Davis v. Mich. Dep’t of Treasury , 489 U.S. 803, 813 (1989). Combining that judicial perspective with common law history shows that habeas is unambiguously civil, at least in the abstract.

The preceding conclusion does not fully resolve the issue before us. After all, petitioners do not file habeas actions in the abstract. They file habeas actions to challenge particular restraints on their freedom. Our precedent teaches that the nature of the challenged restraint matters.

In Ewing v. Rodgers , 6 we held that 'a habeas petition challenging confinement arising from a criminal judgment is not a ‘civil action’” within the meaning of the EAJA. 826 F.2d 967, 971 (10th Cir. 1987) (footnote omitted). We reached that conclusion because petitioners in such habeas Page 19cases have access to appointed counsel, and their criminal imprisonment creates a 'custodial incentive' to sue. Id. at 971 & n.5. Neither point is inherent to habeas; they are both characteristics that stem from the underlying criminal judgment. As such, Ewing requires us to consider the underlying proceedings when deciding if a habeas action is civil under the EAJA.⁷

With that foundation established, we can resolve the issue raised by this appeal. For the reasons explained above, habeas is civil. So too are the underlying proceedings being challenged. Immigration detention is an 'aspect of the deportation process.' Demore v. Kim , 538 U.S. 510, 523 (2003). In turn, a deportation proceeding itself is 'a purely civil action.' INS v. Lopez-Mendoza , 468 U.S. 1032, 1038 (1984). As a result, immigration detention is civil. Zadvydas v. Davis , 533 U.S. 678, 690 (2001). Because both habeas and the underlying immigration proceedings are civil, habeas actions challenging immigration detention are purely civil. Ewing , which addressed criminal habeas rather than immigration habeas, does not require that we hold otherwise. See Vacchio v. Ashcroft , 404 F.3d 663, 672Page 20 (2d Cir. 2005) (finding that precedent excluding criminal habeas from the EAJA did not control the question of whether immigration habeas fell within the EAJA’s scope).

We therefore hold that, under the EAJA, habeas actions challenging immigration detention are unambiguously 'civil actions.' The EAJA authorizes the award of attorneys’ fees to petitioners who prevail against the Government in such actions.

Because text, common law, and judicial history so strongly characterize immigration habeas actions as civil, we need not turn to any other tools of statutory interpretation. Nonetheless, we also observe that the EAJA’s stated statutory purpose supports our conclusion.

As a threshold matter, statutory purpose is relevant to our interpretation of sovereign immunity waivers. The Supreme Court has previously endorsed the consideration of statutory purpose. West v. Gibson , 527 U.S. 212, 222 (1999). While the Court has subsequently disavowed legislative history when interpreting waivers, Kirtz , 601 U.S. at 49, it has not rejected the use of purpose. Certainly, the Court's instructions to disregard legislative history prevents us from using such history to discern purpose. But here, Congress included an express statement of purpose in the bill enacting the EAJA. Pub. L. 96–481, § 202(c), 94 Stat. 2321, 2325.

Page 21There is no need to guess at purpose through legislative history. The purpose is right there in the text, passed by both chambers of Congress and signed by the President.

The EAJA's stated purposes are twofold: firstly 'to diminish the deterrent effect of seeking review of, or defending against, governmental action by providing in specified situations an award of attorney fees, expert witness fees, and other costs against the United States”; and secondly “to insure [sic] the applicability in actions by or against the United States of the common law and statutory exceptions to the ‘American rule’ respecting the award of attorney fees.' Id. The parties do not address the second purpose, so we focus on the first.

Habeas petitions play an important role in challenging - and deterring – unlawful detention. Yet, the high cost of retaining an attorney to file a petition can dissuade many people from doing so. That cost barrier is especially high in the immigration detention context. Immigration is a sprawling and complicated area of law that can be difficult to navigate without the aid of counsel. It is also difficult for those in immigration detention to afford counsel because they are unable to work while detained. See Ingrid V. Eagly & Steven Shafer, A National Study of Access to Counsel in Immigration Court , 164 U. Pa. L. Rev. 1, 35 (2015). The availability of Page 22EAJA fees in habeas actions challenging immigration detention therefore plays an important role in lowering financial barriers to filing such actions.

Permitting fees would incentivize private counsel to accept representation of indigent, immigration detainees. It would also boost the availability of pro bono services in this area. Although pro bono counsel are willing to represent clients free of charge, their services are not costless. Instead, pro bono counsel must secure their funding from sources other than their clients. For legal services organizations that already offer pro bono immigration services, EAJA fees would supplement their funding streams and provide an opportunity for them to expand their services. See Amica Ctr. Am. Br. at 20-23. For organizations that did not previously offer immigration services due to lack of funding, the availability of fees could enable them to offer immigration services for the first time. Id.

The Government suggests that the EAJA's fee incentives are unnecessary in the context of habeas challenges to immigration detention, so statutory purpose does not favor permitting such fees. For one, the Government claims that the Criminal Justice Act authorizes district courts to appoint counsel for indigent individuals challenging their immigration detention through habeas. Reply Br. at 25 (citing 18 U.S.C. § 3006A(a)(2)(B)). However, that proposition is both contested and underdeveloped since the Government raised it for the first time in reply, Page 23so we do not consider the argument. See United States v. Wells , 739 F.3d 511, 530 n.27 (10th Cir. 2014).

The Government separately contends that detention status gives individuals a “custodial incentive” to challenge their immigration detention. It is true that we considered “custodial incentives” in the past when we held that EAJA fees were not available in habeas actions challenging criminal imprisonment. Ewing , 826 F.2d at 971 & n.5. But we also explicitly distinguished the custodial incentive in cases of criminal imprisonment from the incentive in cases involving immigration. Id. Once a criminal judgment is affirmed on appeal, the only way to seek release from the courts is to file a habeas petition. There is no other choice.

When it comes to immigration detention, though, an individual can avoid continued confinement by agreeing to removal. See 8 U.S.C § 1226(a) (authorizing immigration detention 'pending a decision on whether the alien is to be removed from the United States'). Thus, if that individual does not have the means to hire an attorney to challenge her detention, she may be forced to accept deportation in exchange for being released from custody, even if she has a right not to be detained and to remain in the United States. The availability of EAJA fees in that circumstance is vital to preserving the individual’s ability to challenge her detention.

Page 24Experience, too, illustrates the difference between criminal and immigration habeas. As the Ewing court observed, '[t]he volume of habeas petitions processed in the federal courts . . . suggests that there is little, if any, economic deterrent in seeking review of the legality of criminal confinement.” Ewing , 826 F.2d at 971. The volume of habeas petitions challenging immigration detention is not so high. According to ICE’s own statistics, there were approximately 38,000 immigrants in detention at the end of fiscal year 2024. U.S. Immigration and Customs Enforcement , FY 2024 Detention Statistics , https://perma.cc/FA4X-FZ7S. But court statistics show that just 448 habeas petitions were filed on behalf of alien detainees in calendar year 2024. U.S. Courts, Statistical Tables for the Federal Judiciary C-2 (Dec. 31, 2024 reporting period), https://perma.cc/E4E3JHJF. Interpreting the EAJA to allow fees in immigration habeas actions therefore furthers its statutory purpose.

We next address the Government’s remaining arguments, which we find to be unpersuasive.

The thrust of the Government’s position is that habeas is not a civil action because, by its inherent nature, habeas is a special form of hybrid action with both civil and criminal characteristics. According to the Page 25Government, that hybrid nature exists regardless of whether the underlying proceedings being challenged are civil or criminal. This position cannot square with the history of the term “civil action” that we catalogued above, nor with the text of the EAJA.

First, the Government's support for its hybrid thesis comes almost exclusively from cases recognizing that habeas is procedurally distinct from the average civil case. Op. Br. at 19–23. As we already explained earlier in this opinion, though, the fact that the EAJA includes 'proceedings for judicial review of agency action” renders procedural differences irrelevant. Supra Section III.B.1.

Second , the Government argues that all habeas actions are of the same nature because “the essence or function of the habeas application [is] to seek release.” Reply Br. at 17 (quoting Obando-Segura v. Garland , 999 F.3d 190, 194 (4th Cir. 2021)). That argument is correct as far as it goes. But it cuts against the Government’s position. The act of seeking release from unlawful custody is civil because unlawful custody is a private wrong. See 1 Blackstone, supra , at *130; 3 Blackstone, supra , at *127. So, if the Government’s argument about the essential nature of habeas controlled, that argument would support the conclusion that all habeas actions are civil, not that all habeas actions are hybrid.

Third , the Government takes issue with the fact that many of the Page 26cases in which courts have held habeas to be civil took place in contexts other than sovereign immunity. To the contrary, the fact that habeas has been held to be civil in so many diverse contexts strengthens our conclusion that habeas is civil in this context, too. Given that habeas has uniformly been understood as civil in virtually every area of law, it would be unusual for Congress to implicitly single out the attorneys’ fees context as the sole area in which habeas is not civil.

Finally , the Government encourages us to follow the Fourth and Fifth Circuits in finding that habeas is a hybrid proceeding no matter the nature of the underlying detention. See Obando-Segura , 999 F.3d 190 (4th Cir. 2021); Barco v. Witte , 65 F.4th 782 (5th Cir. 2023). However, neither the Fourth nor Fifth Circuits conducted the type of close historical analysis that undergirds our conclusion here, and we respectfully disagree with the path that each took to conclude that immigration habeas actions are civilcriminal hybrids.

In Obando-Segura , the Fourth Circuit primarily relied on three cases to reach its holding: the Supreme Court's decisions in Harris and Schlanger , and the Fourth Circuit’s own decision in O’Brien v. Moore , 395 F.3d 499 (4th Cir. 2005). We have already explained why Harris and Schlanger do not establish that habeas is hybrid under the EAJA. Supra Section III.B.1. And O’Brien dealt with habeas actions challenging criminal imprisonment, not Page 27immigration detention. O’Brien held that habeas actions challenging criminal imprisonment were hybrid specifically because those actions “assume[] part of the underlying case’s criminal nature.” 395 F.3d at 505. Its focus on the nature of the underlying detention is therefore consistent with our reasoning here.

Obando-Segura declined to apply this aspect of O’Brien because O’Brien did not specifically differentiate habeas actions challenging criminal detention from those challenging civil detention. Obando-Segura , 999 F.3d at 194. Apart from O’Brien ’s silence on the issue, Obando-Segura identifies no other source of non-civil character in habeas actions challenging immigration detention. Perhaps the Obando-Segura court thought itself bound by O’Brien ’s silence. We certainly are not, and we do not find it persuasive to read so much into O’Brien ’s silence, as the panel there may not have had occasion to draw any distinction between the two types of habeas. See In re Cox Enters., Inc. Set-top Cable Television Box Antitrust Litig. , 835 F.3d 1195, 1212 (10th Cir. 2016).

We find the Fifth Circuit's decision in Barco to be of limited assistance as well. The extent of Barco 's analysis was to cite four cases (including O’Brien and Ewing , which we have discussed above) in which courts held that habeas petitions challenging criminal imprisonment were civilcriminal hybrid actions. Barco , 65 F.4th at 785 & n.1. Barco did not consider Page 28how habeas petitions challenging immigration detention might be different. For this reason, we do not find Barco to be compelling.

B

In a variation of its main argument, the Government suggests that our decision interpreting the Prison Litigation Reform Act in McIntosh v. U.S. Parole Commission , counsels in favor of interpreting the EAJA to avoid “case-by-case” adjudication of different types of habeas petitions. Reply Br. at 15 (citing McIntosh, 115 F.3d 809 (10th Cir. 1997)). Essentially, the Government advocates for a categorical approach. That is, if a category contains both purely civil cases and civil-criminal hybrid cases, EAJA fees are unavailable in all the cases in that category. Were we to adopt that approach, all habeas would be beyond the EAJA's reach because habeas actions challenging criminal imprisonment are not purely civil. See Ewing , 826 F.2d at 971.

The problem is that this approach depends on McIntosh ’s analysis of the Prison Litigation Reform Act, which uses the phrase “ a civil action.' 115 F.3d at 811 (emphasis added). The EAJA uses the phrase “ any civil action.' 28 U.S.C. § 2412(d)(1)(A) (emphasis added). This difference is meaningful. As we have held, the use of the modifier “any” before a term in the EAJA “broadens the scope of” that term. Nelson v. United States , 40 F.4th 1105, 1113 (10th Cir. 2022). The term “any civil action” refers to civil actions ofPage 29 “whatever kind.” Id. at 1115 (collecting cases). That includes immigration habeas actions, which are purely civil, even if they are grouped in the same category as criminal habeas actions, which are not.

What is more, “courts ‘are not at liberty to create an exception where Congress has declined to do so.'' Freytag v. Comm'r , 501 U.S. 868, 874, (1991) (quoting Hallstrom v. Tillamook Cnty. , 493 U.S. 20, 27 (1989)). Congress included just one categorical exception in the EAJA: 'cases sounding in tort.” 28 U.S.C. § 2412(d)(1)(A). The EAJA contains no other categorical exception, much less one for habeas. In fact, the word “habeas” does not even appear in its text. See id. We cannot read exceptions into the EAJA that are unsupported by the text. Accordingly, we reject the categorical approach proposed by the Government.

Lastly, the sovereign immunity canon or clear statement rule, which would instruct us to adopt the Government’s preferred interpretation due to ambiguity in the text, does not compel a different result. Because we conclude that habeas actions challenging immigration detention unambiguously fall within the EAJA’s ambit, we are not required to accept the Government's interpretation. Regardless, under the Government's preferred construction of the EAJA – that fees are available “only [in] those cases that are wholly or purely civil in nature,” Op. Br. at 24 – fees would Page 30still be available in immigration habeas since such habeas actions are purely civil.

The EAJA may be a broad statute, but a statute is not ambiguous just because it is broad. Here, we read the EAJA's broad language to unambiguously authorize fees in habeas actions challenging immigration detention.

AFFIRMED.