Summary

  • Arwa Mahdawi’s Guardian column constructs its critique of ICE’s October 3, 2025 miscarriage-tracking halt through a juxtaposition whose three load-bearing premises — rhetoric-conduct equivalence, CDC-ICE surveillance equivalence, and a count supplied without comparative baseline — the column does not separately defend.
  • ICE recorded 18 miscarriages during the first nine months of the Trump administration before discontinuing public tracking on October 3, 2025, according to the Guardian investigation.
  • The Guardian investigation also reported a record number of pregnant detainee arrests during the same period.
  • The discontinuation sits within a federal pregnancy-surveillance architecture that continues to track outcomes elsewhere, including CDC abortion surveillance and ICE’s smartwatch-based delivery monitoring.
  • A 2025 Pregnancy Justice report documented 400+ pregnancy-related prosecutions across 16 states in the two years following Dobbs, placing the federal-level data gap inside an already-operating state-level enforcement architecture.

ICE stopped publicly tracking miscarriages among women in its custody on October 3, 2025, a Guardian investigation published this week found, after the agency had recorded 18 pregnancy losses during the first nine months of the Trump administration. The discontinuation coincides with a record number of arrests of pregnant women and teenagers by the agency during the same period, the investigation reported, and sits within a federal pregnancy-outcome surveillance architecture that continues to track outcomes at multiple other points — including abortion surveillance at the CDC under Director Erica Schwartz, mandated state reporting in 44 states plus Washington, D.C. as counted by the Guttmacher Institute, and ICE’s own location-tracking of pregnant detainees during hospital deliveries through smartwatches. Arwa Mahdawi’s accompanying Guardian opinion column characterized the data gap as “very strange,” paired it with administration officials’ stated pronatalism, and built its critique through a juxtaposition whose three structural premises the text does not separately defend. A separate 2025 Pregnancy Justice report documented prosecutors in 16 states charging more than 400 people with pregnancy-related crimes in the two years following the Dobbs decision, situating the federal-level absence within an enforcement architecture already prosecuting pregnancy outcomes at state scale.

The documented cessation

ICE recorded 18 miscarriages during the first nine months of the Trump administration before discontinuing public tracking on October 3, 2025, according to the Guardian’s investigation. The October 3 date establishes the halt itself as a measurable event rather than an inferred pattern. The 18-figure floor represents only what was recorded, not necessarily what occurred. The Guardian investigation also reported that ICE has arrested a record number of pregnant women and teenagers during the same period, and that detained pregnant women are reportedly not receiving the prenatal healthcare they need.

A values-versus-conduct juxtaposition

In a Guardian opinion column published this week, Arwa Mahdawi paired the discontinuation with administration officials’ stated pronatalism. Mahdawi wrote that “Maga men have babies on the brain” and cited Donald Trump’s statement that “we want more babies, to put it very nicely,” Trump’s self-description as the “father of IVF,” and Trump’s joking self-reference as the “fertilization president.” Shortly after being sworn in as vice president, JD Vance declared he wanted “more babies in the United States of America,” per the column. The column cited Elon Musk’s reported comments to Ashley St Clair — identified by Mahdawi as the mother of one of Musk’s “many offspring” — that he wanted to have “a legion of children before the civil war,” and noted that at least one of Musk’s children has publicly said she is “not sure he’s considered that they might not all be fighting on the same side.”

Mahdawi characterized the data gap as “very strange,” noting that ICE and the Trump administration are “normally very big on surveillance.” Earlier Guardian reporting described a parallel surveillance program that reached pregnant immigrant women giving birth: pregnant detainees arriving at hospitals to deliver their babies while wearing location-tracking smartwatches as part of an ICE monitoring program. In one case, a woman diagnosed with pre-eclampsia — a condition that can cause severe swelling — was reportedly concerned the smartwatch would cut off her circulation. “She was in tears about [the idea of taking the surveillance device off],” a hospital staffer told the Guardian, adding that the woman “had this deep fear that ICE was going to come to the hospital and take her baby.”

The state-level enforcement context

A 2025 report by the advocacy group Pregnancy Justice documented prosecutors in 16 states charging more than 400 people with pregnancy-related crimes in the two years following the Dobbs decision. The report noted that “the documented cases likely understate the total, with many prosecutions of miscarriage and pregnancy loss never making headlines.” The Pregnancy Justice methodology is not adjudicated within the substrate; the report is identified in the article as the work of an advocacy group.

A structural audit of the column’s argument

The column’s critique is constructed through juxtaposition. Its argumentative force depends on three premises the text moves across but does not separately defend. Removing any one of them materially weakens the critique, and the text does not supply the supporting evidence for any of them.

First, the column treats administration pronatalist rhetoric and ICE detention-data practices as commensurable expressions of a single policy posture. Cultural-political utterances and an operational data decision inside a custodial agency are treated as logically entailing specific data-collection obligations for all pregnant people under federal jurisdiction. The equivalence the column draws is not established in the column itself.

Second, the column uses Dr. Erica Schwartz’s confirmation-hearing statement that “Abortion surveillance is absolutely a critical component of what the CDC is currently doing” — Schwartz was confirmed earlier this year by the U.S. Senate as CDC director — to establish that the administration values pregnancy surveillance, then extends the inference to ICE custody data practices. The move requires treating CDC surveillance and ICE custody tracking as expressions of a single surveillance philosophy rather than separate programs with distinct statutory authorities, populations, and operational contexts, a framing the column does not defend.

Third, the 18-miscarriage count is presented as evidence of a problem without supplying a comparative baseline. The substrate does not establish whether 18 is high, low, or typical relative to the detained pregnant population, to prior administrations, or to expected background rates. Mahdawi characterizes the data gap as “very strange,” not the count. The column’s evidence-weight leans on the existence of a documented count being discontinued, not on the count indicating elevated risk. A reader who treats any discontinuation of custody health data as presumptively problematic will find the column conclusive; a reader who requires the count itself to indicate elevated harm will find the column inconclusive on the present evidence.

The cross-register movement the column performs has been characterized by Nicholas Shackel as a structural pattern in which a defensible narrow claim is paired with a broader claim that retreats to the narrow one when challenged. The defensible administrative claim — that a custodial agency holding pregnant detainees should track pregnancy outcomes for the same reason any custodial agency tracks health events — is one position. The broader political claim, illustrated through Musk’s reported “a legion of children before the civil war” remark and the denunciation by one of his children, operates in a different register. The column moves between them, gaining the institutional weight of the first while operating in the rhetorical register of the second. The second register rests on juxtaposition and cultural critique that the substrate does not quantify.

What the substrate does not establish

The combination of Schwartz’s confirmation-hearing statement and the discontinuation of custody miscarriage tracking may genuinely indicate inconsistent institutional priorities. The column does not, however, demonstrate that the inconsistency operates at the level of evidence or institutional design the juxtaposition implies. The substrate does not establish whether the federal-level cessation reflects a resource decision, a policy direction, or an oversight; the administration retains the option to characterize the cessation as any of the three, and the substrate does not adjudicate among them. The smartwatch reporting grounds the narrower claim about custody treatment and supplies a specific patient case; it does not, by itself, ground a broader pronatalism-versus-data critique. The structural-audit observations on the column do not by themselves rebut the column’s underlying concern; the column’s three load-bearing premises may be defensible, but the substrate does not supply the defense.

Additional considerations

Pregnant women in ICE custody are the population whose pregnancy outcomes are no longer publicly tracked after October 3, 2025. The cessation is anomalous within an institutional architecture the substrate describes as tracking pregnancy outcomes at multiple points — federal surveillance through the CDC, mandated state reporting in 44 states plus Washington, D.C., and ICE’s own delivery-monitoring smartwatch program — making the federal-level data discontinuation an exception within an otherwise-documented enforcement architecture rather than an isolated anomaly.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Argument Audit
A full structural audit of an argument’s premises, inferences, and load-bearing assumptions.