KOHEN AVOCATS · PARIS, FRANCE
Charlotte Sweeney ICE Colorado ruling: can ICE rearrest you after a judge orders release?

Hassan Kohen is a Paris-based lawyer admitted to the Paris Bar (Avocat au Barreau de Paris). Initial strategic analysis: €80 incl. VAT, personal reply within 24 hours.

100% confidential · Legal professional privilege · No obligation

Article generated by artificial intelligence, through a process designed and supervised by the firm

Sources: U.S. court opinions and official U.S. government publications, read for this article

Maître Hassan KOHEN, avocat au Barreau de Paris
Hassan KOHEN
Attorney at the Paris Bar

Charlotte Sweeney ICE Colorado ruling: can ICE rearrest you after a judge orders release?

Speak to a lawyer 06 89 11 34 45

On November 18, 2026, a federal judge in Colorado found that immigration officials had been in contempt of court for rearresting a detainee within moments of his court-ordered release, then declined to impose any penalty because the practice had stopped. The decision, issued by Judge Charlotte N. Sweeney of the United States District Court for the District of Colorado, answers a question that matters far beyond one case: when a court orders a detained person released, can immigration agents simply arrest the same person again at the jailhouse door? The short answer from Colorado is no, at least not by walking the person into a public lobby and immediately shackling them again while a release order stands. The longer answer is more careful, because rearrest after release is sometimes lawful, the contempt finding bound only the parties before that court, and detention law turns on exactly which statute the government invokes. The discussion that follows is general in scope: it describes checks useful in comparable situations and makes no findings about Ramon Quintero Alvarez’s situation. It explains what the November 18 ruling decided, why the judge punished no one, which detention statute controls whether bond is available, and how habeas courts measure compliance with their own orders.

I. What the Colorado court actually decided

A. A release order, a lobby rearrest, and eighteen cases

The case is Quintero Alvarez v. Baltazar, No. 1:26-cv-04891-CNS (D. Colo. Nov. 18, 2026), a habeas petition filed under the federal habeas statute by Ramon Quintero Alvarez, a Mexican-born man who told the court he had lived in the United States for more than thirty years and had been held in immigration detention since February 2026. Anyone studying this ruling should read the contempt discussion itself, because the judge ties every finding to testimony about the October 27 lobby rearrest and the later decision to stop the practice. The procedural sequence, as recited in the November 18 order, began on October 20, 2026, when the court directed the government to show cause by October 27 why the habeas petition should not be granted, adding that if no response arrived by that date, the petitioner was to be released from custody no later than twenty-four hours after the deadline. According to the order, the government’s November 3 status report stated that the petitioner had been released from immigration custody on October 27, 2026, was immediately re-detained, and remained in detention, while government counsel affirmatively represented that it would not argue that the ongoing detention was lawful.

What happened at the detention facility on October 27 matters because the court treated the details as proof of disobedience rather than of compliance. Testifying at a November 4 hearing, Nichole Black, an assistant field office director for Immigration and Customs Enforcement’s Enforcement and Removal Operations in Denver, said she had received a direct instruction, passed down from Deputy Field Office Director George Valdez and Acting Field Office Director Nathan Rasper, to re-detain habeas petitioners who had some type of criminal history. She confirmed that she knew of the court’s order requiring release, that the petitioner was handed his personal property, walked out to the lobby, stepped into a public area, and was rearrested by officers waiting outside. She also acknowledged, according to the order, that her information about pending criminal charges against the petitioner was wrong: the charges underlying his most recent arrest had been dropped on April 24, 2026. At the time of the first hearing she knew of about ten similar immediate re-detentions after court-ordered releases; with later filings bringing the district-wide total to eighteen such cases, according to the order.

The second hearing, on November 9, shifted the focus from the officers who carried out the rearrest to the official who ordered the policy. Rasper testified under oath that he understood immigration authorities had an obligation to make lawful arrests, that colleagues had told him rearrest after a release was permissible, and that the new approach of releasing and then immediately re-detaining petitioners was consistent with the agency’s statutory authority and with court orders. He also testified that this approach had begun in his district on October 27, 2026, that he had given verbal instructions to review each release case with legal counsel before rearrest, and, critically, that after the November 4 hearing he had directed the practice to stop and those rearrested under it to be released. Counsel for the government represented at the same hearing that the practice of immediate re-detention had ceased. On that record, the court granted the habeas petition in a separate final judgment, denied the remaining motions as moot because the petitioner had been released, and directed the clerk to close the case on November 18, 2026. In short, the court found contempt for the October 27 rearrest, held it purged by the November 9 testimony stopping the practice, granted habeas, denied remaining motions as moot, imposed no further sanction, and closed the case.

The order adds one institutional detail that explains why the judge treated the rearrests as a deliberate practice rather than isolated mistakes. The court noted that the judicial district had entered a memorandum of agreement with the United States Attorney’s Office under which habeas releases would be carried out consistent with court orders, and that officials and the new immediate re-detention approach proceeded despite that framework. That context supported the finding that the October 27 rearrest was not a misunderstanding at the facility door but part of an approach announced to agents across the district on that same day.

B. Why the court found contempt, and why it imposed no penalty

Civil contempt in federal court asks three practical questions, and the November 18 order works through each of them with testimony rather than assumptions. The governing framework, drawn from 18 U.S.C. section 401, gives federal courts power to punish disobedience or resistance to a lawful writ, order, or command, and the Tenth Circuit rule quoted in the order requires clear and convincing evidence of three things: a valid court order existed, the defendant knew of it, and the defendant disobeyed it. The Supreme Court has long described this authority as inherent in every court, and the order cites that principle alongside the reminder that all orders and judgments of courts must be complied with promptly. Applied to the facts, the court found each element satisfied: the October 20 order was valid and its release language plain, Black testified that she knew of it, and the lobby rearrest within moments of release was disobedience, not compliance, because escorting a person to a public area before shackling them again does not even attempt to carry out a release. The order stresses that in civil contempt proceedings disobedience need not be willful, and that a person can be held in contempt for failing to be reasonably diligent and energetic in attempting what was ordered.

The absence of any fine or jail time is not an acquittal; it is the other half of how civil contempt works. Unlike criminal contempt, which punishes past defiance, civil contempt is coercive: it pressures the disobedient party back into compliance, and the contemnor can purge the contempt by affirmatively coming into compliance. The order relies on exactly that doctrine, holding that Rasper’s sworn representation at the November 9 hearing, that the immediate re-detention approach had ceased and that any petitioner ordered released would no longer be released and immediately re-detained, cured the contempt that had run from October 27 to November 9. Because the violation was purged and the petitioner had been released, the court wrote that it declines to take any further action and closed the case. Readers should notice the boundary this draws. A contempt finding on these facts binds the parties and officials before that court; it does not by itself free any other detainee, invalidate any other arrest, or decide whether a future rearrest supported by genuinely new facts and a fresh warrant would be lawful. And a useful comparison comes from a Utah habeas case decided in March 2026: in Vielma Uzcategui v. Brooksby, No. 4:26-cv-00020, decided March 5, 2026, the United States District Court in Utah granted a habeas petition, ordered immediate release under prior release terms, required a compliance report within four business hours, and barred any renewed detention of the petitioner under the same statutory provision. The pair of rulings makes the method clear: courts measure what officials actually did against the precise words of the order, and cosmetic compliance, like a walk to the lobby before rearrest, fails that measurement while real compliance ends the dispute.

Contempt also requires fair procedure for the accused party, and the order documents each step. At the November 4 hearing the court gave formal notice that it viewed the re-detention as contempt of the October 20 order, and the November 9 evidentiary hearing gave officials the chance to question witnesses, present the statutory-authority defense, and argue why no finding should issue. Only after that adversarial testing did the court conclude that the knowledge and disobedience elements were proved by clear and convincing evidence, including testimony from the officials’ own witnesses. The method matters for readers hoping to use this ruling elsewhere: a contempt motion without notice, a hearing, and specific evidence tied to the order’s words will not produce the same result.

II. When detention, release, and rearrest are lawful

A. The statute behind the custody decides whether bond is available

Immigration detention is not one single regime, and the Colorado dispute shows why the label on the government’s authority changes everything. In Quiroz v. Mullin, No. 26-6019 (10th Cir. June 30, 2026), the United States Court of Appeals for the Tenth Circuit, whose decisions control federal cases in Colorado, confronted the government’s 2025 position that noncitizens found inside the country could be held under the mandatory detention provision written for arriving aliens. For readers in Colorado and the wider Tenth Circuit, this is the controlling comparison: interior detention usually means section 1226(a), and section 1226(a) means a chance to ask for bond. The court held that noncitizens who entered the United States and were later detained in the country’s interior are usually subject to section 1226(a), and therefore eligible for bond, rather than to section 1225(b)(2)(A), reversing a district court that had denied both release and a bond hearing. The opinion traces decades of agency practice treating section 1225 as the border provision and section 1226 as the interior provision, notes that five circuits had already rejected the government’s new reading, and holds that the Laken Riley Act’s amendments left that understanding in place.

The Tenth Circuit also rejected the argument that a 2025 immigration statute had silently rewritten these categories. Reviewing the Laken Riley Act’s amendments to sections 1225 and 1226, the court found no language displacing decades of practice and precedent treating section 1225(b)(2)(A) as the border provision and section 1226(a) as the interior one, and it treated that silence as ratification of the prior understanding. The breadth of participation in the case underlines its practical weight: immigration-law scholars, the American Immigration Lawyers Association, and more than twenty states filed friend-of-the-court briefs, and the published opinion runs through text, history, purpose, and constitutional avoidance before reaching its bond-eligibility holding.

The text the court construed is worth reading directly, because it contains both the detainee’s best argument and the government’s. Under 8 U.S.C. section 1226(a), a noncitizen may be arrested and detained pending a decision on removal, and the government may either continue detention or release the person on a bond of at least $1,500 or on conditional parole. But the very next subsection, section 1226(b), provides that the government may at any time revoke a bond or parole granted under subsection (a), rearrest the person under the original warrant, and detain them. That revocation language is the lawful rearrest power the Colorado officials gestured toward when they invoked statutory authority, and it explains why the judge’s analysis focused on the court order rather than on whether rearrest is ever allowed: a valid release order from a habeas court sets the compliance baseline, and rearrest under the same paperwork, moments after a court-ordered release the government had conceded was required, could not satisfy it. Anyone reviewing a detention file should therefore ask first which provision the government actually invokes, because mandatory detention categories, bond eligibility, and rearrest authority all flow from that choice.

The Supreme Court’s framework sharpens the point by refusing to read bond rights into statutes that do not contain them. In Jennings v. Rodriguez, 583 U.S. 281 (2018), decided February 27, 2018, the Supreme Court of the United States held that three detention provisions, sections 1225(b), 1226(a), and 1226(c), do not implicitly entitle detainees to periodic bond hearings every six months, reversing a Ninth Circuit rule that had required them. For a detainee hoping a statute silently promises a bond hearing every six months, this decision closes that shortcut: the text must actually grant it, or a court will not invent it. A March 2026 recommendation shows how lower courts apply that framework to individual cases: in Manjot Kaur v. Lyons, No. CV-26-00217, decided March 24, 2026, the United States District Court for the District of Arizona addressed an asylum seeker held at the San Luis Regional Detention Center who sought release or a bond hearing, and recommended a bond hearing within thirty days with the government bearing the burden of showing danger or flight risk. A March 2026 decision applied the same reading and recommended a bond hearing within thirty days, with the government bearing the burden on danger and flight risk. Together these cases describe a coherent method: identify the detention statute, ask what process that statute and its regulations actually provide, and test the government’s answer in a habeas petition if custody rests on nothing else.

B. Habeas courts enforce the limits, and check compliance case by case

The writ of habeas corpus is the enforcement tool behind each of the rulings above, and its statute is deliberately fast. Under 28 U.S.C. section 2241, federal district courts may grant the writ to a person held in custody in violation of the Constitution, laws, or treaties of the United States, and under 28 U.S.C. section 2243 a judge entertaining an application must promptly either award the writ or issue an order directing the custodian to show cause why it should not be granted, with the return due within three days unless good cause extends the time. The Colorado sequence follows that template: show-cause order, missed deadline, release directive, then contempt when release was undone. The order also invoked the All Writs Act, 28 U.S.C. section 1651(a), to bar moving the petitioner out of the district without court permission, a reminder that a habeas court can freeze the status quo while it decides whether custody is lawful. For families, the practical consequence is that habeas petitions in detention cases move on timelines measured in days, and a missed government deadline can itself trigger release.

The outer limit on how long the government may hold someone comes from the Supreme Court’s post-removal detention cases. In Zadvydas v. Davis, 533 U.S. 678 (2001), decided June 28, 2001, the Supreme Court of the United States construed the post-removal-period detention statute to avoid a serious constitutional problem, holding that once removal is no longer reasonably foreseeable, continued detention is no longer authorized by statute, and assigning federal habeas courts the job of deciding whether particular circumstances fall within or beyond the reasonably necessary period. For anyone held months after a final removal order with no country willing to take them, this is the precedent that turns waiting time into a legal argument a habeas judge can decide. To guide lower courts, the decision recognized six months as a presumptively reasonable period, while stressing that the presumption does not mean every person must be released at six months: after that point, a detainee who gives good reason to believe there is no significant likelihood of removal in the reasonably foreseeable future shifts the burden to the government to show otherwise.

The August 2026 Florida decision applies that limit the way courts actually use it, which is narrower than headlines suggest. The Vielma Uzcategui v. Brooksby, No. 4:26-cv-00020 decision, issued March 5, 2026 by a United States District Court in Utah, applied that habeas framework to order release: the court granted the writ, restored prior release terms, put the asylum hearing back on track, and demanded proof of compliance within hours, while forbidding reuse of the same detention provision. A March 2026 decision shows the remedy end of that framework: habeas granted, immediate release ordered, a compliance report due within four business hours, and no second detention under the same provision. The lesson for detainees and families is concrete and unsentimental. Get the court’s order in writing and read its exact words, because compliance is measured sentence by sentence. Document what officials actually do after release, including times, places, and any new paperwork, because cosmetic steps do not satisfy a release directive. Ask counsel promptly which detention statute the government relies on and whether bond was ever offered, since section 1226(a) cases and mandatory detention categories lead to different hearings. If circumstances change after a bond denial, ask the immigration judge for redetermination rather than filing the same petition twice, because courts treat repeat petitions without new facts as already decided. And do not mistake any single district ruling for nationwide law: the Colorado contempt finding bound its parties, the Tenth Circuit’s interior-detention reading binds its circuit, and the governing provision in another state may be construed differently until the Supreme Court resolves it.

Two background principles from the Colorado order help readers place these enforcement cases in context. First, the order recalls that due process protections apply to all persons within the United States, including noncitizens regardless of immigration status, citing decisions stretching from the nineteenth century to Zadvydas itself. That principle is why habeas courts reach detention disputes at all: custody must have a lawful statutory basis even where the underlying removal case belongs to the immigration courts. Second, the burden structure in extended-detention cases is genuinely two-sided. Zadvydas does not order release at six months automatically; it asks first whether the detainee gives good reason to doubt removal in the reasonably foreseeable future, and only then requires the government to justify continued custody. A detainee who cannot make that first showing, or who remains subject to mandatory detention categories during ongoing proceedings, will generally stay detained even after long months inside.

Conclusion

A court-ordered release is a legal event, not a suggestion, and the Colorado ruling shows what follows when officials treat it as a revolving door: a contempt finding supported by clear and convincing evidence, purged only when the practice verifiably stopped. The same body of law also shows the boundaries. Detention authority depends on the statute actually invoked, bond depends on what that statute and its regulations provide, extended post-removal detention ends where reasonable foreseeability of removal ends, and compliance with any order is judged on what was really done. Anyone facing detention should therefore start with the paper: the exact order, the exact statutory citation, and the exact record of what happened after release. Those three documents decide whether the next step is a bond request, a redetermination motion, or a habeas petition, and they are the materials any lawyer will need first.

📄 Official circular