NextFin News - The Justice Department has asked a federal judge to consider contempt sanctions against an immigrant-rights group and its lawyers after they published an independent analysis showing that 42% of the names on the government's list of suspected noncitizen voters in California could not be found in the state's voter rolls. The Sunday filing marks a sharp escalation in a legal battle the administration has already lost repeatedly in court, and it puts the department in the unusual position of defending the accuracy of a database it has used to threaten election officials with criminal prosecution just days before an election.
The question the sanctions fight raises is whether the government can shield a contested voter list from outside scrutiny by confining access to a narrow procedural purpose, or whether courts will treat an analysis that challenges the list's accuracy as part of the same standing inquiry the judge authorized.
The Filing and the Analysis It Targets
The dispute centers on a list of 20,930 California registrants that the Justice Department compiled and demanded state election officials treat as possible illegal voters, including by separating their ballots during the Election Day count. On Friday, Chief Judge James Boasberg ordered the department to turn the list over to attorneys for the Coalition for Humane Immigrant Rights, known as CHIRLA, and the UCLA Voting Rights Project, which sued to block the policy. Boasberg permitted access for the narrow purpose of establishing legal standing and ordered the plaintiffs not to disclose any names.
On Saturday, the plaintiffs filed court papers presenting an analysis by two data analysts at UCLA. The research found that 8,880 of the names, or 42% of the list, do not correspond to anyone in California's current voter file. It also identified 22 people on the list who were likely naturalized U.S. citizens, while saying that thousands more U.S. citizens appeared to be on it as well. The filing, cosigned by Xavier Becerra, a senior adviser to the UCLA project and the Democratic candidate for California governor, argued:
The expert analysis further makes clear that the mandated voter list endangers a significant number of eligible California voter.
The Justice Department's Sunday filing responded by arguing that the analysis had nothing to do with the plaintiffs' standing and therefore violated Boasberg's order.
The report's political grandstanding about alleged inaccuracies that purportedly impact California voters generally, or comparisons between the government's list and the full "California voter file," has no relevance to whether the particular plaintiffs in this case have standing.
The department suggested that Boasberg refer the matter to the Justice Department for investigation for possible criminal contempt of court.
Alberto Lammers, a spokesman for the UCLA Voting Rights Project, said in a statement: "Plaintiffs' counsel and their experts fully complied with the Court's orders."
A Pattern of Judicial Rejections
The sanctions motion arrives as the administration's broader effort to use the list faces mounting legal resistance. On Friday night, U.S. District Court Judge Trina L. Thompson, a San Francisco-based judge, barred the department from demanding that election offices remove possible noncitizens from voter rolls, separate their ballots, or threaten legal action. The Justice Department appealed that decision on Sunday, and Thompson scheduled a Monday hearing on the appeal. The department has noted that Thompson ruled before the federal government was able to file its initial arguments.
That case is one of several losses in the battle over the alleged noncitizen voter lists. A federal judge in California rejected the demand that election officials be forced to separate ballots cast by those voters, writing that such last-minute orders could "wreak havoc" on the state's election, and two appellate judges agreed. California election officials contend the list is riddled with errors and say they have no way to complete the painstaking investigation required by law to remove voters from the rolls in the days before the election. Under state law, people attest to being U.S. citizens when they register to vote and risk felony prosecution if they cast a ballot while ineligible.
The administration has said its list was carefully compiled from the state's own voter registration data and compared against federal citizenship records. Homeland Security Investigations said it used "multiple data-enrichment and verification methods" to assess whether the individuals identified were likely noncitizens. The Justice Department issued a series of subpoenas for election offices in California and warned election officials who knowingly let noncitizens vote that they could face criminal charges.
Beyond California, the department has urged Washington state to remove what it says were 400 possible noncitizens on its rolls. State election officials there have said that list is also unreliable and includes one voter who cast ballots in the 1990s.
Why the Accuracy Question Is Harder Than It Looks
The credibility of the California list cannot be separated from the track record of the underlying data system. The department's effort grew out of a March 2025 executive order directing states to vet voters using the Department of Homeland Security's Systematic Alien Verification for Entitlements system, known as SAVE, which matches voter registration records against immigration records maintained by U.S. Citizenship and Immigration Services.
That matching process has produced significant errors in practice. In Texas, the secretary of state's office checked the state's 18 million registered voters against SAVE in October 2025 and sent counties a list of 2,724 potential noncitizens. Of the 578 cases that have since been confirmed as U.S. citizens, 506 were flagged only because of a lag in passport data, according to court filings reviewed in the litigation. In Georgia, the state's Department of Audits and Accounts review updated in September 2026 found 2,570 potential noncitizen registrations that had reached active or inactive status because of administrative errors, with 407 of them, about 16%, carrying any recorded voting history; the audit did not establish that all 407 people were noncitizens when they voted.
These episodes matter because they illustrate the mechanism by which a data-merge produces false positives: citizenship indicators in federal databases are not always synchronized with naturalization records, and name-matching routines can conflate lawful permanent residents, naturalized citizens, and noncitizens. When the federal government then demands that states act on those matches days before an election, the burden shifts to local officials to untangle errors under time pressure they say they do not have.
The structural problem is not a one-off bad list; it is a data-matching architecture that generates a predictable false-positive rate, combined with an enforcement timeline that leaves no room for correction. That is a regime-level defect, not a cyclical fluctuation that will self-correct on its own. The errors will recur in each election cycle as long as the matching rules and the enforcement deadline remain unchanged.
The Second-Order Stakes: Chilling Effect Beyond the List
The immediate legal question is whether the plaintiffs violated a court order. The second-order question is what the government's response signals about its willingness to expose the list to adversarial testing. Chad Dunn, legal director of the UCLA Voting Rights Project, said during an earlier hearing:
If you're a voter in California you have to wonder if you're on this list; will you ever be on this list?
He added that CHIRLA had received calls from concerned citizens saying they would not vote or wanted to learn whether they were on the government's list. "It absolutely has a suppressive effect," he said.
A sanctions referral would compound that effect by telling advocacy groups and their experts that analyzing a government-produced voter list, even when a judge has granted them access to it, carries the risk of a contempt investigation. The practical consequence is not limited to this case: it raises the cost of outside verification in future election-integrity litigation, where independent analysis is often the only way plaintiffs can establish the concrete harm that standing requires.
For the Justice Department, the analysis struck at the credibility of a list that has become the centerpiece of its Election Day enforcement posture. A finding that a large share of the names cannot be matched to actual voters undermines the premise that election officials must act on the list immediately. The sanctions request can be read as an effort to contain that damage by narrowing the permissible use of the data.
The Counter-Thesis
The strongest argument for the department is procedural rather than factual. Boasberg authorized access to the list for a specific purpose, and the government contends the plaintiffs exceeded it. Courts routinely enforce protective orders strictly, and a party that uses court-granted access to publish a broader critique can be seen as testing the boundaries of what was permitted. From this view, the accuracy of the list is beside the point; the question is whether the plaintiffs stayed within the four corners of the order.
That argument has force, but it does not fully answer the plaintiffs' position. The analysis was filed with the court as part of the standing showing Boasberg authorized, not published as a freestanding public report. If the judge accepts that the broader comparison is relevant to whether eligible voters face a credible threat, the sanctions request loses much of its footing.
The falsifying signal is specific: if Boasberg refers the matter for a contempt investigation or sanctions the plaintiffs or their counsel, it would indicate the court views the analysis as outside the scope of its order, vindicating the department's procedural theory. If he declines and instead treats the analysis as part of the standing record, the government's attempt to wall off the list from scrutiny will have failed.
What to Watch
In the short term, the focus is on two proceedings: Boasberg's response to the sanctions request, and Judge Thompson's Monday hearing on the government's appeal of her order blocking the department's demands to election offices. A contempt referral would open a separate proceeding running parallel to the underlying challenge, potentially drawing out the litigation past the election. A denial would return the case to the merits of whether the list can be used to alter ballot handling days before voters cast ballots.
Over the medium term, the outcome will shape how much deference courts give to federal data-matching efforts that states say are error-prone, and whether the government can use the threat of criminal liability to compel local election officials to act on lists they dispute. The enforcement block covers nine states that DHS directed agents to prioritize for voter investigations: California, Connecticut, Georgia, Missouri, Nevada, New York, Pennsylvania, Washington and Wisconsin. That block expires when polls close on Nov. 3, so the rulings in these final days will set the terms under which the administration enters the next phase of its election-integrity campaign.
More than 10 million naturalized citizens of voting age live in those nine states, according to Census data, representing more than two in five of all naturalized voting-age citizens nationwide. That demographic weight is why the accuracy of the underlying matching system matters beyond any single list.
Three scenarios frame the path ahead. In the base case, the courts allow the analysis to stand as part of the standing record but stop short of sanctioning either side, leaving the merits to be decided on whether the list's error rate makes it unconstitutional to use. In an upside case for the government, Boasberg sanctions the plaintiffs or refers the matter for contempt, which would deter similar outside analyses and preserve the list's insulation from adversarial review. In the downside case for the government, the courts not only reject the sanctions request but also issue a broader ruling that limits how federal agencies can use database matches to pressure state election officials, setting a precedent that constrains the SAVE-based enforcement model in other states.
The central tension is straightforward: a federal agency says it has identified a large pool of ineligible voters using federal databases, while courts, states, and now an independent analysis say the underlying data is too unreliable to justify last-minute changes to how ballots are handled. The next rulings will determine which version prevails, and whether a government voter list can be both shielded from scrutiny and used as the basis for last-minute election intervention.
Explore more exclusive insights at nextfin.ai.
