Guardian Documents DHS Using a Customs-Import Statute, 19 U.S.C. § 1509, to Take Journalists' Phone Records and Union Financial Records Without a Judge

confirmed Importance 9/10 ~12 min read 2 sources 7 actors

Opening

The Guardian (Sam Levine, 2026-08-29) documents DHS using 19 U.S.C. § 1509 — a customs-import records provision — to obtain phone records, financial records, and platform account data on journalists, unions, and nonprofits, with no judge and no grand jury. A § 1509 administrative summons requires only the sign-off of a DHS official, and instructs recipients to keep it secret.

The sequence that makes this an evasion finding, not just a surveillance finding

This is the load-bearing structure. Report it as a sequence.

  1. February 2026 — federal prosecutors twice sought search warrants for the YouTube account information of Georgia Fort and Don Lemon, journalists who had pleaded not guilty to charges from a January Minneapolis church protest they were covering.
  2. A judge rejected the request twice, writing the government had failed to establish probable cause and that he wanted Lemon and Fort notified so they could challenge it.
  3. Late February / March — the government said it was withdrawing the request.
  4. Less than a month later — DHS served Google with a § 1509 summons for the same YouTube information. No judge required.

The same data, sought through a channel with no judicial review, weeks after a judge said no. The Guardian’s framing — “an end run around a judge” — is the reporter’s characterization; the sequence itself is the evidence, and it is documented in court papers.

What was actually obtained

  • Six months of Georgia Fort’s phone records from T-Mobile — more than 10,000 calls and text messages. She was not notified until mid-July, when government lawyers produced them to her own lawyers. Her lawyers wrote this week that they were “stunned.”
  • Google refused, responding that DHS had not shown how the request related to a customs investigation.
  • In the separate 15-activist conspiracy case: DHS successfully obtained financial records of the Sunrise Movement, SEIU, and the Communications Workers of America, plus Venmo records for Voices for Racial Justice. None of these organizations is charged with any crime, and DHS offered no explanation for why it needed the records.

The other YouTube accounts sought — and the range is the point: Democracy Now, conservative podcaster Megyn Kelly, the Milwaukee Journal-Sentinel, and independent journalist Brendan Gutenschwager. Some cited videos were not livestreams of the protest at all — the Democracy Now video was a news report plus an interview with Nekima Levy Armstrong, who led the protest; the Kelly video was an interview with Jonathan Parnell, the church’s pastor.

A demand reaching Megyn Kelly and Democracy Now in the same summons set is not ideological targeting; it is a demand keyed to COVERAGE OF AN EVENT. State it that way — it is both more accurate and harder to dismiss.

DOJ’s position, from a December 2025 court filing:

“Although § 1509 references ‘duties, fees, and taxes,’ the plain language of the statute does not limit DHS’s investigative authority to those subjects; instead, DHS is authorized to investigate potential crimes to ensure ‘compliance with the laws of the United States administered by the United States Customs Service’”

The predicate DOJ asserted for the church case: protesters entered the church because a local ICE official was a pastor there and could have been assaulted or interfered with — even though, per the Guardian, that man does not appear to have been present, and no DHS officials were at the church or involved in the protest.

Against it — and both speakers are former DHS insiders, which is why they matter:

  • Chris Duncan, former DHS lawyer: “These laws have absolutely nothing to do with a domestic situation at a church, a social media post, even an immigration matter.” And: “It’s not a free-for-all that was thrown in there. Congress does not operate that way… Congress obviously intended to authorize only records, demands and interviews in furtherance of investigations into customs violations, not wild goose chases.”
  • John Roth, DHS Inspector General 2014-2017: “It’s outrageous conduct on so many levels… This is an improper use of the subpoena under any circumstances. This is not a customs case; it is not a customs violation. They are not investigating a customs violation.”

The withdrawal pattern — the sharpest structural observation in the piece

DHS has repeatedly withdrawn a § 1509 summons after it was challenged in court and before a judge could rule.

  • 2017: Twitter sued over a summons seeking to unmask @alt_uscis, an account critical of DHS. DHS withdrew.
  • 2025: DHS served Meta to unmask the user behind an account monitoring ICE agent activity in the Philadelphia suburbs. Arguments were heard January 15; DHS withdrew the next day.

Lauren Regan, Civil Liberties Defense Center: “They don’t want a judge to take away this scary tool because they are getting stuff out of it. Once a court ruling says ’thou shalt not use this statute’, it does not apply.”

Withdrawing before a ruling preserves the tool. That is a function claim, provable from the pattern of outcomes — do not write it as a claim about anyone’s intent. The Guardian says it “may be a deliberate strategy”; the corpus should say what the sequence does, and note the intent question is not established. See function-not-intent-the-analytical-rule-2026.

Why the scale is unknowable, and why that is itself the finding

Summonses stay hidden unless the company or user challenges them. Companies are NOT required to notify users. The Guardian: “unable to identify any cases where the government attempted to get a court order to enforce a 1509 summons outside of the traditional customs context” — i.e. companies that refuse are apparently not pursued, which suggests the tool’s power is compliance, not enforceability.

NYT reported in February 2026 that DHS had served HUNDREDS of administrative subpoenas on social media companies. Attribute that to NYT; it is not independently verified here, and “administrative subpoenas” may be a broader category than § 1509 specifically.

Nathan Freed Wessler, ACLU: “Without knowing how many of these subpoenas there are and what they’re being used for, there’s no way for courts or lawmakers or the public to put checks on executive branch abuses.”

The cost asymmetryF. Mario Trujillo, EFF: “If a user actually hired a lawyer, it would cost tens of thousands of dollars to fight one of these… they’re pushing that cost onto users and onto non-profit groups when they could easily get their high-powered lawyers who are being paid $500 to $1,000 an hour to fight these.”

Why This Event Matters — and what it does to two existing corpus threads

1. It generalizes the Whipple Shield finding from Minnesota to a nationwide practice. 2026-02-15–project-whipple-shield-hsi-financial-surveillance-minnesota-unions documented HSI using warrantless administrative summonses under customs authority to take years of union and nonprofit financial records on a “domestic terrorist financing” theory. The Guardian shows the same statute deployed against journalists, platforms, and payment processors across multiple states and years. Whipple Shield is not a local anomaly — it is one instance of a standing instrument.

2. It adds a fourth surveillance channel to the Minnesota 15 record. The corpus holds Operation Keyhole (a Certified Undercover Operation), Operation Puppet Master (infiltration), and Project Whipple Shield (financial). § 1509 platform and telecom demands are a distinct fourth channel — and the financial records of SEIU, CWA, Sunrise, and Voices for Racial Justice were obtained in that same 15-activist case.

3. The 2017 DHS OIG report is the strongest available evidence that this is a KNOWN, DOCUMENTED failure. The OIG found “inconsistent — and, in some cases, improper” use, located the problem in CBP’s Office of Professional Responsibility, recommended reforms, and the office agreed to them. A documented 2017 finding, agreed reforms, and the same misuse in 2025-26 is an accountability-mechanism story — and it is checkable against the report itself.

Research Gaps — ordered by value

  • PULL THE 2017 DHS OIG REPORT. Get its number, its actual findings, the reforms recommended, and whether any were implemented. The Guardian quotes it secondhand; do not repeat the quoted phrase without reading the document. Highest-value item here.
  • The Fort filing “this week” — get the exact docket number and date. The “stunned” quote is from a court filing and should be cited to it, not to the Guardian.
  • The December 2025 DOJ filing containing the § 1509 statutory-interpretation argument — docket and document number. That is the government’s own theory in its own words.
  • Whether any court has EVER ruled on § 1509’s scope outside customs. The Guardian implies no. A verified absence here would be a genuine finding — but it requires a real search, and RECAP is a mirror of what people have fetched, not a register of what exists.
  • The NYT February 2026 “hundreds of administrative subpoenas” report — read it directly and establish whether it counts § 1509 specifically or a broader category.
  • The MontCo Community Watch docket — the REAL § 1509 Philadelphia-suburbs case (EFF tracker, ACLU-PA case page). Subpoenaed 2025-09-11, withdrawn after a hearing 2026-01-16 — note the Guardian’s “January 2025” is a YEAR off; it is January 2026. Docket not yet located.
  • The Reddit case — docket, and whether it produced any ruling before withdrawal.

WHAT THIS CORPUS ALREADY HELD — read before drafting, we are ahead of the Guardian on three points

Checked 2026-08-30. The Guardian is not the origin of this thread here; it is national corroboration of a mechanism this corpus documented from primary sources first. Three things we hold that the Guardian does not:

1. THE SUMMONS FORMS THEMSELVES, quoted verbatim from the document. 2026-08-28-whipple-shield-verified-from-summons-formsHSI Form 3115 states its own authority in plain text:

“Issued under authority of section 509, Tariff Act of 1930, as amended by Pub. L. No. 95-410 (19 U.S.C. § 1509); 44 F.R. 2217; Homeland Security Act of 2002.”

We read the instrument; the Guardian describes it. That entry also establishes what it is not — not 8 U.S.C. § 1225(d)(4) immigration authority, not an IRS-style summons, not a grand jury subpoena. Cite the form, not the newspaper, for what the authority is.

2. A SHARPER MECHANISM THAN “NO JUDICIAL OVERSIGHT”: THE ASYMMETRY IS NOTICE, NOT STANDING. standing-to-resist-an-administrative-subpoena-why-unions-did-not-move-to-quash-when-doe-plaintiffs-did carries a conductor correction dated 2026-08-20 establishing this, and it is the more precise claim:

  • The Doe/LBRRN account-holder could and did move to quash: DHS’s subpoena to Meta was dated 2025-09-03; the motion followed 2025-09-19, sixteen days later.

    CORRECTION 2026-08-30 — the conductor’s error, caught by the § 1509 fact-check. I originally wrote that Doe/LBRRN was “the Philadelphia-suburbs ICE-monitoring account the Guardian mentions.” It is not, and it is not even a § 1509 case. A checker pulled the actual motion-to-quash PDF from RECAP, docket 3:25-mc-80288 (N.D. Cal.): it invokes 8 U.S.C. § 1225(d) throughout, never mentions § 1509 or the Tariff Act in 27 pages, and its fact pattern is Long Beach, CaliforniaLBRRN = Long Beach Rapid Response Network, centred on an “Officer Simeon” incident.

    The source note I drew from says “8 U.S.C. § 1225(d) / 19 U.S.C. § 1509” — naming BOTH authorities — and never says Philadelphia. I resolved that ambiguity in the wrong direction and added a geography that was not in the source. Two failures at once: collapsing two statutes into one, and inferring a location.

    Why it matters beyond the geography: § 1225(d) is immigration authority — precisely what the § 1509 forms do NOT cite. So using this docket as the § 1509 comparison case argues against the very distinction the thread rests on.

    A real § 1509 Philadelphia-suburbs case does exist — “MontCo Community Watch,” per EFF’s tracker and ACLU-PA’s case page — but its docket number was not located (search rate limit; recorded as an access blocker, not an absence). Do not cite a docket for it until one is obtained.

    The notice-not-standing MECHANISM still holds — a party who learns of a summons in time can move to quash; one who does not, cannot. But Doe/LBRRN illustrates it under § 1225(d), not § 1509, and any piece using it must say so.

  • The Whipple Shield unions did notnot because they lacked standing, which they doctrinally likely had, but because nothing in the record notified them in time relative to compliance.

“No judge in the loop” is true but incomplete. The operative variable is WHO FOUND OUT, AND WHEN, RELATIVE TO COMPLIANCE. A challengeable summons that nobody learns about is unchallenged, not unchallengeable. The note explicitly requires distinguishing CANNOT from DID NOT — do not write “the unions couldn’t fight it.”

3. THE AGGREGATE THAT SIZES THE PATTERN. litigation-surface-search-strategy-and-the-wider-pattern, citing Just Security’s tracker (2026-07-20): 550+ § 111 charges nationally against people connected to protests or immigration enforcement · at least 213 dismissed, dropped, or acquitted · at least 16 grand jury subpoenas blocked or withdrawn, 12 of them to perceived political opponents · and grand-jury “no bills,” which the tracker calls “almost unheard of for most career federal prosecutors.”

That aggregate is the denominator the Guardian piece lacks — it says the scale is unknowable, and for § 1509 specifically that is right, but the adjacent § 111 record is measured.

What the Guardian genuinely ADDS

  • The evasion sequence in one case: warrant refused twice by a judge → withdrawn → § 1509 summons for the same data weeks later. That is new and it is the strongest single fact on the thread.
  • T-Mobile complied; Google refused. A named telecom handing over 10,000+ call and text records on a journalist is new.
  • Two former DHS insiders on the record — a former DHS lawyer and the 2014-17 DHS Inspector General — calling it improper. Insider attribution this thread did not previously have.
  • The 2017 OIG report and its agreed reforms — see pull-2017-dhs-oig-report-1509-summons-misuse-and-the-agreed-reforms.
  • The recipient breadth: Democracy Now, Megyn Kelly, Milwaukee Journal-Sentinel, Brendan Gutenschwager.

Drafting note. The corpus rule is that our edge is holding the whole sequence when other outlets hold single incidents (feedback_prior_coverage_is_not_a_fence). This is that case exactly: Levine has the national frame; we have the form, the notice mechanism, the union case, and the § 111 denominator. Acknowledge and credit the Guardian, then assemble what only this corpus can.

Sources & Citations

Tiers Tier 1 court records & gov docs · Tier 2 established outlets · Tier 3 regional & specialty press · Tier 4 opinion or single-source. Methodology →
Cite this entry
The Cascade Ledger. “Guardian Documents DHS Using a Customs-Import Statute, 19 U.S.C. § 1509, to Take Journalists' Phone Records and Union Financial Records Without a Judge.” The Capture Cascade Timeline, August 29, 2026. https://capturecascade.org/event/2026-08-29--dhs-19-usc-1509-customs-summonses-journalists-unions-nonprofits-no-judicial-review/