From Parole to Prison: How 500,000 Legal Immigrants Became Undocumented Overnight

On a date in early 2025, the Department of Homeland Security published a notice in the Federal Register announcing the termination of the CHNV humanitarian parole program. The language was administrative, procedural, almost anodyne: parole previously granted to nationals of Cuba, Haiti, Nicaragua, and Venezuela would be revoked. Work authorizations tied to that parole would expire. Beneficiaries were expected to depart the United States or face removal proceedings.

The notice did not mention Emmanuel Damas by name. It did not need to. Damas, a 56-year-old Haitian father of two who had entered the United States in 2024 under that same program, was one of an estimated 530,000 people whose legal status evaporated with the stroke of a bureaucratic pen. He was working for his brothers' transportation company in Boston at the time. According to his family, he did not know his status had changed until Immigration and Customs Enforcement agents took him into custody after an arrest months later. By early March 2026, he was dead in an Arizona detention facility, an infection from an untreated toothache having spread through his body until he went into septic shock.

The distance between a Federal Register notice and a man shackled to a hospital bed, unable to blink, is the distance this article covers.

The Architecture of Parole

The legal authority behind the CHNV program sits in a single clause of American immigration law: Section 212(d)(5) of the Immigration and Nationality Act, enacted in 1952. The provision grants the Attorney General, and by later delegation the Secretary of Homeland Security, discretionary power to "parole into the United States temporarily under such conditions as he may prescribe only on a case-by-case basis for urgent humanitarian reasons or significant public benefit any alien applying for admission."

The language is significant in what it implies and what it omits. Parole is not a visa. It is not asylum. It is not a green card or any form of permanent immigration status. It is a temporary, discretionary grant of lawful presence that can be revoked at any time. The provision was written for individual cases, for the Hungarian refugees of 1956 who needed immediate entry before their paperwork could be processed, for the Vietnamese evacuees of 1975 who arrived faster than the State Department's visa machinery could handle.

For decades, that was how parole worked: a safety valve for individual humanitarian emergencies. Presidents of both parties used it sparingly, and Congress periodically complained that even sparing use exceeded the statute's intent. The Refugee Act of 1980 was in part a response to the perception that parole had been stretched beyond its original purpose when the Ford and Carter administrations used it to admit hundreds of thousands of Indochinese refugees.

The Biden administration's CHNV program pushed parole authority further than it had ever gone. Launched in October 2022 for Venezuelan nationals and expanded in January 2023 to include Cubans, Haitians, and Nicaraguans, the program allowed nationals of those four countries to apply for two-year parole with work authorization. The requirements were substantial: applicants needed a US-based financial sponsor who filed Form I-134A, underwent background checks and biometric screening, and arranged commercial flights coordinated through the USCIS portal. This was not an open border. It was a structured, vetted process that required applicants to identify themselves to the US government, submit to screening, and enter through official ports of entry.

By the time the program was terminated, approximately 530,000 people had been admitted under it.

Four Countries, One Program

The four countries targeted by the CHNV program shared one characteristic: each was in acute crisis, and each was producing migration flows that overwhelmed the existing immigration system. Beyond that commonality, their situations were distinct.

Cuba was enduring its worst economic crisis since the "Special Period" of the early 1990s, compounded by the government's crackdown on the July 2021 protests that had brought tens of thousands into the streets across dozens of cities. Fuel shortages, rolling blackouts, and food scarcity drove a wave of emigration not seen since the Mariel boatlift of 1980.

Venezuela's economic collapse under the Maduro government had been unfolding for nearly a decade, scattering more than seven million Venezuelans across the Western Hemisphere in one of the largest displacement crises in modern Latin American history. The Colombian border, the Darien Gap, and the Mexican transit corridor were all straining under the volume.

Nicaragua under Daniel Ortega had intensified repression since the 2018 protests, stripping citizenship from dissidents, shuttering independent media, and jailing political opponents. Emigration was both an escape valve and, for some, a forced exile.

Haiti occupied a category of its own. The assassination of President Jovenel Moïse in July 2021 had collapsed what remained of governmental authority. By 2023, armed gangs controlled an estimated 80 percent of Port-au-Prince. The country had no functioning elected government, no reliable police force, and no judicial system capable of processing basic criminal cases. Haitians who held Temporary Protected Status in the United States had been living in renewable legal limbo for years. The CHNV program offered something different: a structured pathway with work authorization and a defined, if temporary, legal status.

Emmanuel Damas entered under this program in 2024. He joined relatives in Boston, home to one of the largest Haitian communities in the United States, and went to work.

The Revocation

The Trump administration moved to terminate the CHNV program shortly after taking office in January 2025. The termination proceeded through executive action, with DHS issuing notices that existing parole grants would be revoked and that beneficiaries should prepare to depart or face removal.

The legal basis for the revocation was straightforward in statutory terms. Section 212(d)(5) of the INA gives the executive branch broad discretion to grant parole, and that same discretion has been consistently interpreted to include the power to revoke it. There is no statutory requirement for a grace period. There is no administrative appeal process specific to parole revocation. The grant is discretionary, and what discretion giveth, discretion may take away.

But the practical implications of exercising that discretion against 530,000 people who had entered through a government-created program were without recent precedent. These were not people who had crossed the border unlawfully. They had not overstayed tourist visas or worked without authorization. They had done exactly what the government asked: identified themselves, submitted biometrics, passed background checks, secured sponsors, booked flights, and entered through designated ports of entry. Every one of them was in a US government database with a photograph, fingerprints, and a verified address.

The revocation converted that database from an immigration processing tool into an enforcement targeting list.

The Notification Gap

USCIS possessed contact information for every CHNV beneficiary. The application process required current addresses, email addresses, phone numbers, and the contact details of US-based sponsors. The agency had the means to reach every person whose status it was revoking.

Whether it did so effectively is a different question. DHS published the termination in the Federal Register, which satisfies legal notice requirements under administrative law. But a Federal Register publication is a legal fiction of notification: it presumes that every affected person monitors the daily output of the Government Publishing Office. For a Haitian parolee working a transportation job in Boston, a Venezuelan parolee caring for elderly relatives in Miami, or a Nicaraguan parolee whose English was still developing, the Federal Register was not a meaningful channel of communication.

The Damas family's account suggests that individual notification either did not occur or failed to reach him. His family said he was unaware his status had changed until his arrest in Boston, months after the revocation took effect. If USCIS sent letters to the address on file, they may have gone to an old address. If notices were posted to the USCIS online portal, they required a beneficiary to log in and check. If emails were sent, they may have been filtered as spam or gone unread by someone who communicated primarily in Haitian Creole.

The notification gap matters because it determines whether the government afforded even minimal procedural fairness. A legal system that retroactively revokes status without ensuring the affected person knows about the change is not administering law. It is setting a trap. The person continues to live, work, and raise children in the reasonable belief that they remain lawfully present, while the government accumulates the grounds for their removal.

Historical Precedents

The United States has traveled this road before, though rarely at this speed or scale.

The Chinese Exclusion Act of 1882 began as a bar on new immigration but progressively expanded to strip rights from Chinese nationals already lawfully present. The Geary Act of 1892 required all Chinese residents to carry identification certificates or face deportation, effectively converting lawful residents into conditional ones. Those who failed to register were presumed to be unlawfully present until they could prove otherwise. The Supreme Court upheld the scheme in Fong Yue Ting v. United States (1893), reinforcing the plenary power doctrine that still governs immigration law: Congress's authority over immigration is essentially unreviewable by the courts.

Operation Wetback in 1954 resulted in mass removals of Mexican nationals. The US Immigration and Naturalization Service claimed approximately one million deportations, though historians have since estimated the actual figure was closer to 300,000. Whatever the precise number, the operation swept up a significant number of US citizens and lawful permanent residents who were caught in the dragnet because they looked Mexican. The Bracero Program, which had actively recruited Mexican agricultural workers to enter the United States legally, was winding down during the same period, creating a population of workers whose legal status was evaporating even as the farms that employed them still needed their labor.

More recently, the first Trump administration's attempt to rescind the Deferred Action for Childhood Arrivals program tested similar ground. In DHS v. Regents of the University of California (2020), the Supreme Court ruled 5-4 that the rescission was arbitrary and capricious under the Administrative Procedure Act. The Court did not say DACA recipients had a constitutional right to the program. It said the government had failed to adequately consider the reliance interests of the approximately 700,000 people who had structured their lives around it. Chief Justice Roberts wrote that DHS had "failed to consider the conspicuous issues of whether to retain forbearance and what if anything to do about the hardship to DACA recipients."

The DACA parallel is instructive but imperfect. DACA was a deferred-action policy, not parole. It provided protection from deportation and work authorization but did not grant lawful status in the same sense as parole. The CHNV program went further: it created a legal status, required people to enter the country in reliance on it, and gave them work authorization and a defined period of lawful presence. The reliance interest is arguably stronger. The disruption of revocation is certainly greater.

Temporary Protected Status terminations under the first Trump administration also produced litigation. When the administration moved to end TPS for Haitians, Salvadorans, and Hondurans, federal courts in multiple jurisdictions issued injunctions. The core argument was similar: the government had repeatedly renewed TPS for years, creating a population that had built lives in the United States, and a sudden termination required adequate consideration of those interests under the APA.

The Legal Battlefield

Multiple legal challenges to the CHNV revocations were filed in federal courts beginning in 2025. The litigation landscape is complex and unresolved, with cases proceeding in several jurisdictions simultaneously.

The principal legal arguments fall into three categories. First, the Administrative Procedure Act argument: that the blanket revocation of parole for hundreds of thousands of people constitutes a substantive rule change that required notice-and-comment rulemaking, not merely an executive announcement. The APA requires federal agencies to publish proposed rules, allow public comment, and respond to significant comments before finalizing regulatory changes. If a court determines that the CHNV revocation was a "rule" rather than an exercise of case-by-case discretion, the entire revocation could be vacated as procedurally deficient.

Second, the due process argument. The Fifth Amendment's due process clause applies to all "persons" on US soil, not only citizens. The Supreme Court confirmed this principle in Zadvydas v. Davis (2001), holding that non-citizens physically present in the United States have due process rights that constrain the government's ability to detain them indefinitely. Applied to the CHNV context, the argument is that parolees who entered lawfully and built lives in the United States have a protected liberty interest that cannot be extinguished without some form of individualized process: notice, an opportunity to be heard, and a reasoned decision.

Third, the reliance interest doctrine drawn from the DACA precedent. If the Supreme Court found in 2020 that DHS needed to consider the reliance interests of DACA recipients before rescinding that program, the same logic should apply with even greater force to CHNV parolees, who possessed a more robust form of legal status and who had uprooted their lives at the government's explicit invitation.

Some federal courts have issued preliminary injunctions blocking aspects of the revocation, while others have declined to do so. The legal landscape remains fragmented, with different circuits reaching different conclusions about the scope of executive discretion over parole. The cases are likely to reach the appellate courts, and possibly the Supreme Court, within the next one to two years.

The Constitutional Question

Beneath the statutory and procedural arguments lies a constitutional tension that American law has never fully resolved.

On one side sits the plenary power doctrine, rooted in Chae Chan Ping v. United States (1889) and reinforced by more than a century of case law. Under this doctrine, the federal government's authority over immigration is a sovereign power that courts review with extreme deference. Congress can exclude, admit, and remove non-citizens on virtually any grounds, and the executive branch exercises delegated portions of that authority with broad discretion. Under the logic of plenary power, the CHNV revocation is simply the executive branch exercising the same discretion it used to create the program. If the president can parole 530,000 people in, the president can un-parole them.

On the other side sits the due process clause of the Fifth Amendment, which the Supreme Court has repeatedly held applies to non-citizens who are physically present on US soil. In Zadvydas, the Court emphasized that "once an alien enters the country, the legal circumstance changes, for the Due Process Clause applies to all 'persons' within the United States." Under the Mathews v. Eldridge (1976) framework, courts balance the private interest at stake, the risk of erroneous deprivation under current procedures, and the government's interest in efficiency. For a CHNV parolee who entered lawfully, holds work authorization, has US-citizen children, and faces deportation to a country controlled by armed gangs, the private interest is substantial and the risk of erroneous deprivation through blanket revocation without individual notice is high.

The tension between plenary power and due process has produced an unstable equilibrium in immigration law for decades. The CHNV revocations stress that equilibrium in a new way because they involve a population that did not enter informally or overstay a visa. These are people the government sought out, vetted, approved, and transported. The government created the legal pathway, staffed the processing centers, coordinated the flights, and issued the work permits. The question is whether that active solicitation creates any obligation beyond what the plenary power doctrine traditionally recognizes.

No court has squarely answered this question. The CHNV litigation may force the issue.

530,000 People in Limbo

While the courts deliberate, the people at the center of this legal conflict continue to live in the United States. Some have applied for asylum, though the backlog of immigration court cases stretches to years. Some may qualify for Temporary Protected Status if their country of origin retains or regains that designation. Many have US-citizen children born after their arrival, adding a layer of family unity considerations that courts are obligated to weigh.

Most, however, exist in a state of legal suspension that has no clean resolution. Their parole has been revoked, but their removal has not been individually ordered. They remain in the country, working, paying rent, dropping children at school, building the ordinary architecture of American life on a legal foundation that the government has declared void. Each interaction with any system that checks immigration status carries the risk of triggering the enforcement pipeline: a traffic stop that runs their name through databases, a call to police by a neighbor, a background check for a new job.

Emmanuel Damas encountered that pipeline after a family gathering in Dorchester. Police arrived for a well-being check after neighbors called with concerns. A misunderstanding escalated. Officers arrested him on a charge of assault and battery. One of his brothers posted bail.

Then ICE took custody. Damas was transferred from Boston to a detention center in Florence, Arizona. His family says they were not informed where he was being held until he was already critically ill. He died on March 2, 2026. According to his relatives, an infection that started in his tooth was treated with repeated doses of ibuprofen and nothing more until it spread to his lungs and sent him into septic shock.

The legal machinery that produced this outcome was not arbitrary. It was precise. A program was created. A man applied, was vetted, was approved, and was admitted. He complied with every requirement. Then the program was revoked, his status was voided, and the same database that had processed his lawful entry was used to locate and remove him. From parole to prison, the system worked exactly as designed.

The question before the courts is not whether it was legal. Under the plenary power doctrine, it almost certainly was. The question is whether a legal system that operates this way can call itself just. That is not a question courts are designed to answer. But 530,000 people are waiting for someone to try.