Habeas Jurisdiction Lies in the District of Removal When ICE Conceals Custody Through Rapid, Secret Transfers; INA §§ 1252(g), 1252(b)(9), and 1252(a)(5) Do Not Bar Detention-Only Constitutional Challenges

3.A. Habeas jurisdiction: default rules and two exceptions

Default framework: “district of confinement” + “immediate custodian”

The opinion begins from the canonical rule of Rumsfeld v. Padilla: for challenges to present physical custody, the petition should be filed in the district of confinement, naming as respondent the petitioner’s immediate custodian (typically the warden), rather than “remote supervisory official[s].” It also references the statutory requirement that the custodian be named “if known,” 28 U.S.C. § 2242, and the historic “produce the body” conception of custody (citing Wales v. Whitney).

Precedents cited: the “unknown custodian” exception is not speculative

The court squarely rejects the government’s attempt to cast the exception as dubious. It points to:

• Demjanjuk v. Meese (Bork, J., in chambers): where the detainee’s secret location made it “impracticable” to require counsel to file “in every jurisdiction,” supporting habeas filing where practicable.
• Rumsfeld v. Padilla itself: recognizing that when a prisoner is held in an undisclosed location by an unknown custodian, it is “impossible to apply” the immediate custodian and district-of-confinement rules.
• United States v. Moussaoui (4th Cir.): applying the logic by permitting an “ultimate custodian” respondent where the immediate custodian was unknown.

Application to Suri: (1) unknown district of confinement, (2) unknown immediate custodian

The majority accepts the district court’s findings that Suri’s counsel could not reasonably know where Suri was at filing, because:

• Suri was moved rapidly among multiple Virginia sites and then flown out of state, while being denied calls to update his family for later transfers.
• The ICE detainee locator did not show him until the next day.
• The Notice to Appear listed a Texas “residence” address he never lived at and did not reference Louisiana.

On the “custodian” element, the court emphasizes a timing detail with jurisdictional bite: at the moment of filing, Suri had landed in Louisiana but had not yet been booked into the Louisiana facility, making the identity of any “day-to-day” custodian uncertain even for the government.

Justice Kennedy’s concurrence-based exception: forum manipulation / “filing could not catch up”

As an independent basis, the court adopts the exception “borne out of Justice Kennedy’s concurrence in Rumsfeld v. Padilla,” under which jurisdiction lies in the district from which the detainee was removed if government movement is used (or appears designed) to defeat practicable filing. The court highlights:

• Suri’s abnormal “rapid transfers” across multiple facilities within hours.
• District court findings (uncontested on appeal) that the government’s explanations were “non-responsive or riddled with inconsistencies,” supporting inferences of forum shopping.
• The NTA’s Texas address as an apparent mechanism to steer the case to a preferred immigration court forum, despite Suri’s presence in Virginia at service.

The majority situates this exception within habeas’s “adaptable remedy” tradition (citing Boumediene v. Bush) and emphasizes that flexible jurisdictional rules are necessary to prevent manipulation of the writ’s availability.

Doctrinal significance

The opinion’s most durable doctrinal move is not merely applying existing exception language; it treats Justice Kennedy’s Padilla concurrence as an adopted rule of decision in the Fourth Circuit for cases involving concealment/rapid movement that defeats counsel’s ability to file in the “correct” district.

3.B. Transfer/venue

Reviewing for abuse of discretion (citing Nichols v. G.D. Searle & Co.), the court affirms denial of transfer. Two points are central:

• Forfeiture: the government did not request transfer to Louisiana in the district court, invoking Berg v. Kingdom of the Neth. and Kanai v. McHugh.
• Could-have-been-brought constraint: Texas was not available because Suri had not yet been moved there at filing, and transfer statutes (28 U.S.C. §§ 1406(a), 1631) require a district where the action “could have been brought.”

The court also approves the district court’s reliance on “traditional venue considerations” drawn from Braden v. 30th Jud. Circuit Court of Kentucky, emphasizing where events occurred, convenience, access to counsel/family, and the unfairness of ratifying government forum shopping.

3.C. INA “jurisdiction-stripping” provisions

Framing: presumption of review + “clear statement” requirement

Before parsing statutory text, the majority anchors itself in Supreme Court reviewability principles: the need for a clear statement to repeal habeas jurisdiction (citing DHS v. Thuraissigiam and INS v. St. Cyr), the strong presumption favoring judicial review (citing Kucana v. Holder and Guerrero-Lasprilla v. Barr), and legislative history indicating Congress did not intend to preclude habeas review of detention challenges independent of removal orders (H.R. Rep. No. 109-72).

1) 8 U.S.C. § 1252(g): limited to three “discrete actions”

The court treats Reno v. Am.-Arab Anti-Discrimination Comm. (AADC) as controlling on § 1252(g)’s narrow scope: it bars jurisdiction only over claims arising from decisions to “commence proceedings, adjudicate cases, or execute removal orders,” and targets judicial interference with prosecutorial discretion.

Applying that framework, the majority distinguishes detention challenges from challenges to the commencement of proceedings:

• Suri’s petition—especially as amended—targets detention and conditions allegedly imposed for retaliatory/chilling purposes, not a request to stop removal proceedings.
• Detention and removability are treated as legally distinct in the INA’s structure (contrasting 8 U.S.C. § 1226 with 8 U.S.C. § 1227).
• The court aligns with Öztürk v. Hyde and Mahdawi v. Trump (Second Circuit stay decisions), which held retaliatory-detention claims “independent of, and collateral to, the removal process” do not “arise from” the commencement of proceedings under § 1252(g).

A notable institutional point: the opinion emphasizes that immigration courts generally lack authority to decide constitutional questions (citing Matter of C-), which makes channeling detention-constitution claims into removal proceedings practically incoherent.

2) 8 U.S.C. § 1252(b)(9): confined (here) to review of final orders of removal

The majority adopts a structural reading: § 1252(b)(9) sits within § 1252(b), titled “Requirements for review of orders of removal,” and applies “[w]ith respect to review of an order of removal.” Because Suri does not seek review of any order of removal, the court holds § 1252(b)(9) inapplicable. It relies on its own precedents:

• Casa De Maryland v. United States Department of Homeland Security
• Miranda v. Garland

The court then confronts (and rejects) contrary readings based on Jennings v. Rodriguez and the Third Circuit’s Khalil v. President, United States. Its key moves:

• Marks analysis: the court treats the Jennings v. Rodriguez plurality as expressly declining a comprehensive interpretation of § 1252(b)(9), therefore not abrogating Fourth Circuit precedent.
• Now-or-never injury: detention pending removal is a present injury; delaying judicial review until a final order would come “too late to be meaningful,” with an analogy to “here-and-now” review concepts (citing Axon Enter., Inc. v. Fed. Trade Comm'n).
• Record-development problem: because constitutional issues cannot be developed in immigration court, requiring review only on petition for review would deprive appellate courts of a factual record (given 8 U.S.C. § 1252(b)(4)(A)).

The opinion candidly acknowledges a developing split and expressly disagrees with Khalil v. President, United States; it also notes the Second Circuit’s later contrary decision, Mahdawi v. Trump (Mahdawi II), but finds it distinguishable in part due to differing government representations about administrative custody review.

3) 8 U.S.C. § 1252(a)(5): exclusive review only for “an order of removal”

Consistent with its § 1252(b)(9) reading, the court treats § 1252(a)(5) as irrelevant absent an order of removal, and therefore not a bar to detention habeas jurisdiction.

How Supreme Court detention jurisprudence informs the analysis

The opinion reinforces its construction by observing that the Supreme Court repeatedly entertained detention challenges notwithstanding removal-channeling provisions, citing: Zadvydas v. Davis, Clark v. Martinez, Jennings v. Rodriguez, Nielsen v. Preap, Johnson v. Guzman Chavez, and Johnson v. Arteaga-Martinez. The thrust is functional: if detention legality must await the completion of removal proceedings, detention claims become “effectively unreviewable,” a result the Supreme Court has resisted.

3.D. All Writs Act injunction against removal

The district court barred removal “unless and until” it issued a contrary order, invoking the All Writs Act, 28 U.S.C. § 1651. The Fourth Circuit reviews for abuse of discretion and affirms, relying on:

• Pa. Bureau of Corr. v. U.S. Marshals Serv. and United States v. N.Y. Tel. Co. for the principle that courts may issue orders necessary to effectuate and prevent frustration of their jurisdiction.
• Degen v. United States and Hazel-Atlas Glass Co. v. Hartford-Empire Co. for inherent authority to protect proceedings.
• F.T.C. v. Dean Foods Co. and Arrow Transp. Co. v. S. Ry. Co. for preserving status quo to protect review power.
• A.A.R.P. v. Trump for the proposition that reasonable notice and opportunity for judicial review is due before removal in this posture.

The court treats the injunction as jurisdiction-preserving rather than an impermissible assumption of power to review removal orders. Practically, it prevents the executive from mooting the habeas action by removing the petitioner before the court can decide jurisdiction and interim relief.

3.E. The dissent’s competing framework

Judge Wilkinson’s dissent presents a fundamentally different architecture: a broad reading of “arising from” in § 1252(b)(9) and § 1252(g) that channels detention challenges into the petition-for-review process after a final order of removal, emphasizing congressional efforts to avoid “bifurcated and piecemeal litigation.” The dissent’s historical narrative runs through Foti v. INS, INS v. St. Cyr, and the REAL ID Act, and it underscores statutory emphatics (“any,” “all,” “sole and exclusive,” “only,” “no”).

The majority’s response is not to deny the INA’s channeling goals, but to insist on a limiting principle: channeling provisions directed to removal adjudication cannot be read to extinguish meaningful review of present detention constitutionality—especially where immigration courts cannot adjudicate the constitutional questions, and where executive conduct (rapid, secret transfers) threatens to make even habeas access manipulable.

The decision therefore crystallizes a live, high-stakes doctrinal divide: administrative consolidation vs. habeas as an immediate check on executive detention, particularly in politically sensitive enforcement contexts.