From Backfire to Prairieland: Drawing a Line Through 20 Years of Repression
Disclosure: I'm friends with three of the defendants convicted in the federal Prairieland trial and have given money to their defense. I've written about those relationships directly in "The Seeds of Prairieland." The piece below is analysis, not neutral reporting, and I'd rather you read it knowing that. I'm also emphatically not a lawyer.
On June 23, 2026, two federal judges in the Northern District of Texas handed down a combined sentence of roughly 450 years to eight defendants convicted in connection with a July 4, 2025 demonstration outside the Prairieland ICE Detention Center in Alvarado.
Benjamin Song, identified by prosecutors as the only defendant who fired a weapon that night, received 100 years. Maricela Rueda received 70. Five other defendants received 50 years each. Daniel Sanchez-Estrada, who by the government's own account was never present at the scene, received 30.
From the bench in one courtroom, Judge Reed O'Connor reportedly described the sentences as a message to "anyone who shares a similar ideology." In another courtroom, Judge Mark Pittman chastised Song for making a political speech and for being afraid of the government.
Both moments fed a story that has dominated coverage of the case: that the Prairieland prosecution is the leading edge of the Trump administration's campaign against "antifa," authorized by the September 2025 executive order designating it a domestic terrorist organization and the national security memo that followed. That framing is not wrong, but it is incomplete. The executive actions supplied the vocabulary; they did not supply the means.
What makes the Prairieland sentences notable is not that the government reached for some new authority. It is that it reached for old ones. Every instrument that produced these sentences—the statute charged, the theory of liability, the sentencing enhancement that multiplied the exposure, the choice of a federal forum—was built years ago and used against earlier movements. The case is best understood not as a departure but as a recombination: a small inventory of legal tools, assembled over a quarter century, applied here to a wider class of people than ever before.
Sentence enhancements vs. the "radical" left
The clearest line runs through a single sentencing rule—though the precise sentencing factors that produced the Prairieland sentences is not, as of this writing, a matter of public record.
What is documented is the policy. On December 4, 2025—before the Prairieland defendants went to federal trial—the Attorney General issued a memorandum to all federal prosecutors implementing National Security Presidential Memorandum-7, the directive that reoriented the Justice Department toward "Antifa-aligned" domestic terrorism. Its closing instruction is explicit: in calculating the sentencing range, "prosecutors should seek all applicable enhancements under the United States Sentencing Guidelines, including the terrorism enhancement under U.S.S.G. § 3A1.4."
That enhancement does two things at once: it sharply raises the offense level, and it jumps the defendant to the highest criminal-history category regardless of any prior record or the complete lack of one. Its effect is to convert ordinary sentences into terrorism sentences, and to fix the label of "terrorist" on the people it reaches. The sentences handed down on June 23—50 to 100 years on conduct that would otherwise carry far less—are consistent with its application, and the government had instructed its own prosecutors to pursue it.
The enhancement itself is not new. Its signature application against the left came nearly two decades earlier, during what is now known as the "Green Scare." The term refers to the wave of federal prosecutions in the early 2000s that targeted radical environmental and animal-rights activists as domestic terrorists. Property destruction and sabotage that injured no one were reframed as terrorism, prosecuted under enhanced charges, and met with sentences far longer than comparable property crimes typically drew. In the 2007 sentencing of defendants in Operation Backfire—the federal prosecution of a multi-year campaign of Earth Liberation Front and Animal Liberation Front arsons—a federal judge applied § 3A1.4 after finding the arsons were intended to coerce the government. Defendants who had the terrorism enhancement applied received between seven and thirteen years in prison; those who did not received fewer than four.
The same enhancement returned decisively at Prairieland to push exposures that might otherwise have run 10 to 20 years into sentences of 50 to 100. Defense counsel argued that the government could have sought the same enhancement against January 6 defendants and largely did not. The disparity they pointed to has a recent judicial echo: in a 2025 resentencing, a federal judge openly measured the punishment of a foreign-terrorism defendant—who joined no group and harmed no one—against two January 6 leaders, who drew lighter sentences for an armed conspiracy to block the transfer of presidential power, during which at least 174 police officers were injured. Courts asked to apply the enhancement in foreign-terrorism cases have sometimes balked: the Ninth Circuit reversed its use against a defendant whose "material support" amounted to opening social-media accounts.
A slow but steady scope creep
Underneath the § 3A1.4 terrorism enhancement is a longer pattern, visible when several cases are read in sequence. Across roughly two decades, the category of who gets prosecuted has steadily widened: from people who committed acts, to people who offered support for acts, to people who were present for acts, to the infrastructure of community support itself.
Before the category began to widen, prosecutors mostly charged the people who did the thing. When Timothy McVeigh bombed the Murrah Building in 1995, the government tried McVeigh, his co-conspirator, and the friend who knew and stayed silent, but it did not indict the militia movement, the gun-show circuit, or the literature that moved through them as a criminal enterprise. The act was treated as the work of a few men. Leftist defendants who committed acts were charged the same narrow way. In 2001, the environmental activist Jeff "Free" Luers set fire to three SUVs at a Eugene, Oregon dealership to protest climate change; no one was hurt and the fire was out in twenty minutes. He was sentenced at the state level to twenty-two years and eight months, the longest sentence ever handed down for environmental sabotage in the United States, and longer than many Oregon arsonists who had destroyed more and endangered lives. The act was his, and so was the liability.
But as the 21st century wore on, prosecutions of leftist activists began to expand their scope. In 2006, activists known as the SHAC 7 were convicted under the Animal Enterprise Protection Act for running a website that publicized and supported a campaign against an animal-testing company. Prosecutors did not allege the defendants personally carried out the underlying acts; the conviction rested on a conspiracy theory aimed at their communications and support, including doxxing animal researchers, as inciting the acts.
In 2010, the Supreme Court extended the logic to speech directly. In Holder v. Humanitarian Law Project, the Court upheld the material-support statute, 18 U.S.C. § 2339B, as applied even to peaceful speech and training—in this case, instructing a designated group in pursuing its aims nonviolently. The statute's definition of "support" is broad enough to include "service," "personnel," "training," and "expert advice or assistance." Those terms matter later: they are what eventually make a person's mere presence chargeable.
The collective-liability theory got its fullest dress rehearsal in 2017, at the inauguration protests in Washington known as DisruptJ20. After police kettled and arrested more than 200 people, federal prosecutors charged somewhere between 194 and 234 protesters (numbers vary in the reporting) with felony rioting, conspiracy, and property-destruction counts carrying decades of potential prison time, without alleging that most of them had personally damaged anything. The theory was that presence and participation made each defendant liable for the acts of a few. The case eventually collapsed: the first defendants were acquitted, the remaining cases were dropped, and the prosecution ended with no convictions after a disclosure violation surfaced.
The 2020 protests following the murder of George Floyd marked a further turn toward the federal forum. The Justice Department brought hundreds of federal cases, leaning on two older statutes: the civil-disorder provision, 18 U.S.C. § 231(a)(3), a near-dormant 1968 law, and the federal arson statute, 18 U.S.C. § 844, with its mandatory minimums. An analysis by M4BL and CUNY's CLEAR clinic found that the overwhelming majority of these federal cases could have been brought in state court, and that the federal charge typically carried a harsher penalty. The federal government was not filling a gap; it was choosing the heavier punishment.
In the wake of those 2020 protests, the city of Atlanta approved plans to build a sprawling police and fire training facility in a DeKalb County forest, and a broad movement—environmentalists, abolitionists, and neighborhood residents—grew up to oppose what the movement called "Cop City." As the standoff over the forest hardened, the target of prosecution moved to the apparatus of solidarity itself. Dozens of forest defenders were arrested, some almost a mile away from alleged property destruction, and charged under the state's domestic-terrorism statute. A 2023 state RICO indictment then named the entire movement a criminal enterprise, pointing to the fact that some of the arrestees had the number for a bail fund written in Sharpie on their arms. Organizers of that bail fund were charged with money laundering and charity fraud, and activists who distributed flyers were charged with intimidation—the act of bailing out and defending protesters reframed as racketeering. In December 2025, a judge dismissed the RICO charges on a procedural ground—that the attorney general lacked authority to bring the case without the governor's written request—not on the merits.
That collapse did not end the prosecution; it relocated it. In June 2026, just months after the federal convictions for Prairieland, a federal grand jury indicted two people for allegedly throwing "explosives" (fireworks) and setting fires at the offices of Brasfield & Gorrie, the project's contractor, during a 2022 protest. The charges rested on the same two statutes the Justice Department had leaned on after 2020—the federal arson provision, § 844, and the civil-disorder law, § 231—and the Acting Attorney General personally certified the intimidation count as "necessary to secure substantial justice." On its face the indictment is narrow: two named people, charged for acts they are alleged to have personally committed. But it arrives wrapped in what the government hopes is new precedent. The DOJ filed it under the National Security Presidential Memorandum 7 initiative and its Joint Task Force Vanguard, a standing interagency body whose stated mandate is to target not only individuals but "organizations engaged in political violence." The sprawling collective-liability theory had failed in state court; what replaced it was not a retreat but a retrenchment at the federal level, now emboldened by Prairieland to call fireworks "explosives" and a broad-based solidarity movement "antifa."
The blunt instrument of § 2339A
The Prairieland indictment did not charge 18 U.S.C. § 2339B, the Holder statute that applies to foreign terrorist organizations. It charged its close relative, 18 U.S.C. § 2339A: providing material support to terrorists.
This distinction means everything. Section 2339B, the statute upheld in Holder v. Humanitarian Law Project, requires a designated foreign terrorist organization as its anchor. Section 2339A requires no organization at all, only that a defendant knowingly provide support toward an underlying federal "crime of terrorism" as a predicate. As Thomas Brzozowski, the former DOJ Counsel for Domestic Terrorism, puts it, § 2339A is "organization-agnostic": it "targets support for crimes, not support for groups." The government therefore did not need "antifa" to be a real, identifiable organization. It needed a qualifying predicate act—and could then sweep in anyone alleged to have supported the event.
The predicates the government invoked are attempted murder of a federal officer (§ 1114) and depredation of government property (§ 1361). The latter, a Talking Points Memo investigation found, had not been used as a material-support predicate in at least a decade before 2025. Because the breadth of the predicate sets the severity of the § 2339A charge, using a statute that can reach minor property damage (spray-painting, say) to unlock a 15-year terrorism penalty dramatically lowers the threshold for what counts as a terrorism offense.
By the government's own enumeration, the "material support" at issue included property, services, training, communications equipment, weapons, "explosives" (again, consumer-grade fireworks), transportation, and "personnel (including themselves)." At trial, prosecutors described wearing black, using the Signal messaging app, printing zines, and reposting on social media as forms of support.
What makes the § 2339A theory worth watching is not that material support can attach without a designated organization—it always could. It is, according to Brzozowski, that the constraints that once disciplined that power appear to be eroding. The statute's original 1994 text included a First Amendment protection clause barring investigations predicated solely on protected activity; Congress struck it in 1996 at DOJ's urging. What informally replaced it was an internal vetting process requiring material-support charges to be staffed through Main Justice. Brzozowski notes that DOJ's announcement of the Prairieland indictment makes no mention of National Security Division attorneys—a departure from standard practice—following Attorney General Bondi's December memorandum directing prosecutors to charge "the most serious, readily provable offenses" and seek terrorism enhancements as a matter of course. A precision instrument, in his framing, is being turned into "a blunt force instrument." If that approach holds, it normalizes pairing a trivial predicate with a terrorism penalty against conduct adjacent to protest.
What is not yet settled
Two questions remain genuinely open, and nothing above is meant to foreclose either. But the structure traced through these sections changes what is at stake in each.
The first is what happened on July 4, 2025, outside the Prairieland Detention Center. A law-enforcement officer was wounded by gunfire; whether the wound came from an aimed shot, a ricochet from suppressive fire, or a ricochet from his own round striking a rifle turned sideways to him was contested through trial and, by available accounts, never definitively resolved. That uncertainty matters enormously for Song, the one defendant found to have fired a weapon. For the others it matters far less than one might expect—and that is the whole point of the theory used against them. Daniel Sanchez-Estrada was never at the scene. The five who drew 50 years each were reached not through what they did with a gun but through § 2339A, whose exposure is set by the predicate, and a predicate as slight as depredation of government property will do. Section 3A1.4 then multiplies that exposure into a terrorism sentence whether or not a defendant has any record or ever touched a firearm. A case built out of support and presence rather than acts makes the acts themselves nearly incidental to the punishment. The ballistics may decide Song's appeal; for the people sentenced for wearing black and printing zines, they barely register.
The second is whether any of it survives review. The convictions are headed to the Fifth Circuit, where the § 2339A theory—a terrorism statute with no organization at its anchor and a minor-property-damage predicate at its base—will be tested at this scale for the first time. The Cop City RICO dismissal is itself under appeal, with state domestic-terrorism charges available for revival even as the prosecution has already migrated to federal court. And the instruments described here have a history of being swung hard and then mildly reined in. The collective-liability theory that swept up more than 200 people in D.C. in 2017 produced no convictions and collapsed under its own overreach; the Ninth Circuit has reversed the terrorism enhancement when the "material support" was opening social-media accounts. Whether that correction reaches the combination assembled at Prairieland—or whether, this time, the tools hold—remains to be seen.
What is not in doubt is the direction of travel, and it runs along all three axes the case brings together. The reach has widened: from people who committed acts, to people who supported them, to people who were merely present, to the very infrastructure of communities. The severity has climbed: the same enhancement that turned property destruction into a 13-year terrorism sentence in 2007 now makes presence a 50-year sentence in 2026. And the constraints that once disciplined these powers—the First Amendment clause Congress struck in 1996, the National Security Division vetting that kept material-support charges rare—have thinned to the point that a precision instrument can be wielded as a blunt one. None of the tools is new. What is new is the willingness to assemble them at once, and the breadth of the class of people now within their reach. The executive orders supplied a name for that class. The existing machinery supplied everything else.