State of Oregon v. Trump

FOR PUBLICATION FILED UNITED STATES COURT OF APPEALS OCT 20 2025 MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS FOR THE NINTH CIRCUIT

STATE OF OREGON; CITY OF No. 25-6268 PORTLAND, D.C. No. 3:25-cv-01756-IM District of Oregon, Plaintiffs - Appellees, Portland v. ORDER

DONALD J. TRUMP, In his official capacity as President of the United States; PETER HEGSETH, In his official capacity as Secretary of Defense; UNITED STATES DEPARTMENT OF DEFENSE; KRISTI NOEM, In her official capacity as Secretary of Homeland Security; UNITED STATES DEPARTMENT OF HOMELAND SECURITY,

Defendants - Appellants.

Before: Susan P. Graber, Ryan D. Nelson, and Bridget S. Bade, Circuit Judges. Concurrence by Judge R. Nelson; Dissent by Judge Graber.

PER CURIAM:

Defendants-Appellants, President Donald J. Trump, Secretary of Defense Peter Hegseth, and Secretary of Homeland Security Kristi Noem, in their official capacities, the Department of Defense (DoD), and the Department of Homeland Security (DHS) (collectively, Defendants), seek an emergency stay of the district court’s October 4, 2025 temporary restraining order (TRO) enjoining the President’s order to federalize 200 members of the Oregon National Guard for 60 days to protect federal personnel and property at the Lindquist Building, an Immigration and Customs Enforcement (ICE) facility in Portland, Oregon. After considering the record at this preliminary stage, we conclude that it is likely that the President lawfully exercised his statutory authority under 10 U.S.C. § 12406(3), which authorizes the federalization of the National Guard when “the President is unable with the regular forces to execute the laws of the United States.” The evidence the President relied on reflects a “colorable assessment of the facts and law within a ‘range of honest judgment.’” Newsom v. Trump, 141 F.4th 1032, 1051 (9th Cir. 2025) (per curiam) (quoting Sterling v. Constantin, 287 U.S. 378, 399 (1932)). We thus conclude that Defendants are likely to succeed on the merits of their appeal, and that the other stay factors weigh in their favor. We grant Defendants’ motion for a stay pending appeal.

I. APPELLATE JURISDICTION

As an initial matter, Plaintiffs dispute whether we have jurisdiction over Defendants’ motion for a stay pending appeal. But “[w]e have jurisdiction to grant such a stay under the All Writs Act.” Id. at 1043 (citing 28 U.S.C. § 1651). Nonetheless, Plaintiffs argue that we should decline to consider Defendants’ stay motion because we lack jurisdiction over the appeal of a TRO. While TROs are normally unappealable, the Supreme Court has “made clear that where an order has the ‘practical effect’ of granting or denying an injunction, it should be treated as such for purposes of appellate jurisdiction.” Abbott v. Perez, 585 U.S. 579, 594 (2018) (citation omitted); see also Sampson v. Murray, 415 U.S. 61, 87 (1974) (holding that a TRO should be treated as a preliminary injunction because it had the same practical effect as a preliminary injunction).

In similar circumstances, we recently held in Newsom that a TRO enjoining the President from deploying members of the National Guard to Los Angeles had the practical effect of a preliminary injunction under Abbott, and therefore “issues of appellate jurisdiction [did] not affect the likelihood of Defendants’ success on their appeal from the TRO.” Newsom, 141 F.4th at 1044. We further explained that when determining whether a TRO should be construed as an appealable preliminary injunction we consider “whether ‘an adversary hearing has been held, and [whether] the court’s basis for issuing the order [was] strongly challenged.’” Id. at 1043 (alterations in original) (citations omitted). We also explained that when an order exceeds the presumptive duration of 14 days, as stated in Rule 65(b)(2) of the Federal Rules of Civil Procedure, it is unlikely that it will be considered a TRO. Id. (citation omitted).

The parties agree that the district court held an adversary hearing and that the request for a TRO was fully briefed and argued. Plaintiffs argue, however, that Newsom is distinguishable based on the duration of the TRO. There, the TRO did not automatically expire and so could be in force for more than 14 days, Newsom, 141 F.4th at 1044, while here the district court ordered that the TRO would expire on October 18, fourteen days after it was entered. But on October 15, the district court extended the TRO for 14 days. The TRO here will be in force for more than the presumptive 14-day period of Rule 65, and Plaintiffs’ argument to distinguish Newsom on this basis fails. Therefore, the TRO has the practical effect of a preliminary injunction, and as we concluded in Newsom, “issues of appellate jurisdiction do not affect the likelihood of Defendants’ success on their appeal from the TRO.” 141 F.4th at 1044. We have jurisdiction over Defendants’ motion for a stay pending appeal.

II. FACTUAL BACKGROUND AND PROCEDURAL HISTORY

A. Disturbances at the ICE Facility in Portland

In the days and months leading up to the federalization order at issue here, immigration-related protests and events have occurred at several ICE facilities around the country. While many of these events have been peaceful, others have been violent and even deadly. For example, on September 24, there was a shooting at an ICE facility in Dallas in which one detainee was killed, and two others were seriously injured. Wamsley Decl. ¶ 21. 1 On September 26, over 200 protesters blocked access to a gate at an ICE facility in Chicago, and one of the protesters was apprehended with a gun in his possession. Id. ¶ 22.

Since June 2025, there have been regular protests at the Lindquist Federal Building, an ICE facility in Portland. Cantu Decl. ¶ 7 2 ; Dobson Decl. ¶ 20 3 (describing protesters gathering since early June outside the ICE facility “on a nightly basis”). Some of these protests have been peaceful, but many have turned violent, and protesters have threatened federal law enforcement officers and the building. Cantu Decl. ¶ 7; Wamsley Decl. ¶ 9 (“protesters have targeted this site and the employees who work there since early June”). Based on its internal risk assessments, the Portland Police Bureau (PPB) activated Crowd Management Incident Commanders (CMICs) and Rapid Response Teams (RRT) on June 11 through 25, and “as a precautionary measure” on July 4 and 17, and again on August 20. Dobson Decl. ¶¶ 11, 22, 23; Dobson Supp. Decl. ¶ 3. PPB made 25 arrests at the ICE facility protests between June 11 and 19. Dobson Decl. ¶ 22.

Indeed, “[a]s a result of the destruction caused by protesters” and threats to the building, DHS was forced to close the facility for more than three weeks, from June 13 to July 7. Wamsley ¶ 13. “This significantly impeded operations.” Id. “While the facility is again currently operational, the windows on the building must remain boarded to prevent further damage to property or attempts at incursion and to provide security to those federal employees working inside.” Id. But on September 18, the City of Portland issued a Notice of Zoning Violation requiring the removal of wood coverings on the windows within 30 days. Id. ¶ 26.

On June 8, protesters placed wood, rocks, and a traffic barrier apparently to impede operations. Id. ¶ 17. On June 11, a man set fire to various materials protesters compiled to barricade a vehicle gate, and other protesters added to the pile of materials, increasing the flames. Id. ¶¶ 11, 17. Protesters then placed a pole against the main lobby entrance. Id. ¶ 17.

On June 14, a serious incident occurred when a group of protesters, including one who was carrying a firearm, advanced up the driveway of the ICE facility towards the main gate. Cantu Decl. ¶ 8. They threw rocks and sticks at the guard shack and fired M80 fireworks at FPS officers. Id. Protesters threw a mortar at the front entrance of the building, injuring an officer. Id. FPS officers had to extinguish two fires in front of the building. Id. Officers were forced to barricade themselves inside the building and protesters placed chains on the exterior doors. Id. Eventually, the protesters attempted to breach the front door and broke the front door’s glass. Id. FPS officers were forced to deploy long guns but did not use them. Id. The protesters continued to throw mortars at the building. Id. The Border Patrol Tactical Unit (BORTAC) deployed an armored tactical vehicle to the building and, working with an ICE Special Response Team (SRT), they eventually pushed protesters back to the street. Id. Three protesters were arrested and charged with assaulting a federal officer. Id.

On June 24, protesters attempted to set a U.S. flag on fire in the driveway of the ICE facility, and later another protester attempted to light an incendiary device next to the guard shack. Wamsley Decl. ¶ 11. When officers approached the protester, she threatened them with knives. Id. Protesters also poured a substance believed to be motor oil in the vehicle entrance. Id. ¶ 17. Also on June 24, one protester shot officers with a paintball gun after an FPS officer warned protesters to clear the building driveway, and another shined a laser in an officer’s eyes. Cantu Decl. ¶ 10. That same night, another protester was arrested for assaulting an officer with a machete and a knife. Id. All three protesters were charged with assault on a federal officer. Id.

On June 29, an individual tampered with a building access card reader and was arrested after a foot pursuit. Id. ¶ 9. He resisted arrest by biting and kicking the FPS officers taking him into custody. Id. Three card readers were vandalized in June, with replacements expected to arrive in mid-October. Wamsley Decl. ¶ 12. While the card readers are inoperable, employees are delayed entering and exiting the building and protesters take their photographs to dox the officers and their family members. Id.

For months, FPS and ICE officers have been the targets of doxing. Cantu Decl. ¶ 11. On several occasions since June, protesters have followed officers who have been deployed to Portland back to their hotels, and groups have protested at the hotels. Id. Several ICE officers at the Portland ERO office have had their names, photographs, and home addresses posted publicly in several municipal locations and residential neighborhoods, along with threatening messages. Wamsley Decl. ¶ 18. Multiple Portland ICE officers have had unknown individuals appear at their residences in vehicles and on foot, peering into their homes and recording officers coming and going. Id.

On July 4, ICE officers observed several individuals defacing ICE property with graffiti. Id. ¶ 10. When ICE attempted to apprehend these individuals, one ran towards an officer and kicked him causing him to fall, and another threw an incendiary device toward the officers, which then detonated near the officers. Id.

The U.S. Attorney’s Office charged 22 defendants between June 13 and July 8 with offenses committed at the ICE facility, including assaulting federal officers, arson, possession of a destructive device, and depredation of government property. Id. Also in June, a man was arrested for pointing a laser at federal officers outside the ICE facility. Id.

Protesters regularly attempt to impede government vehicles as they enter or exit the facility. Cantu Decl. ¶ 12. When a vehicle stops, protesters surround it, shouting threats to the occupants. Id. For example, on August 9, a group of protesters became aggressive when an ICE vehicle was leaving the facility and began striking the vehicle with their fists. Id. FPS officers had to use pepper balls to push the protesters away from the vehicle. Id. In September, ICE officers reported vehicles following them after they left the ICE facility. Wamsley Decl. ¶ 18.

On September 1, protesters assembled a guillotine outside the ICE facility, while two protesters wielded riot shields. Id. ¶ 16. This incident and others were also recorded in multiple emails from PPB officers reporting their monitoring activities at the ICE facility from September 1 to 30.4 On September 1, the groups of protesters at the facility ranged from 40-200 people at different times of the day. Hughes Decl. Exs. 1, 2. ICE agents “did several push outs and arrested a few.” Id. Ex. 2. That night ICE agents “made a push and deployed gas” to clear the front of the building to move vehicles. Id. Ex. 1. September 2 was “un[]eventful” with “around 20 of the usuals” protesting at the building. Id. Ex. 2. On September 3, a person illegally flying a drone near the upper windows of the building was arrested and reacted violently. Id. There were “about 20 people” protesting that day. Id. Ex. 3.

On September 4, a PPB officer reported seeing several older homeless people being coerced into approaching the ICE facility gate; the instigators told these people that they should create a distraction or simply rattle the gate. Id. Ex. 2. Officers confirmed this was happening by watching “the instigators escorting different [older homeless people] up to the gate.” Id. FPS took custody of an elderly man after he asked “if he could just come up to the gate and rattle it so the antifa instigators would leave him and others alone.” Id. He was escorted behind the gates and interviewed. Id. He told agents that he and several other homeless people were being coerced to act “to instigate agent response.” Id.

Also, on September 4, FPS arrested a woman described as “a daily chronic problem,” who is “very aggressive” and believed to be armed with a Glock. Id. She “has made threats to ICE agents and has expended her three strikes at the facility.” Id. According to one PPB officer, she has been arrested many times and “can be very violent.” Id.

On September 6, there were 50 to 60 “very active protest[e]rs/black blockers/agitators” at the gate. Id. Ex. 5. There was lots of noise, flashing lights, and swinging sticks. Id. FPS was pushing out to allow vehicles to enter, and the PPB observed approximately 50 people in the driveway. Id. FPS had 26 officers on duty. Id. On September 5, 7, 8, 9, 10, and 11, PPB officers reported groups of protesters ranging from 7 to 30 people, with “low energy,” some dressed in all black, one with a baton, but no issues. Id. Exs. 4, 6-11.

On September 9, multiple individuals were observed on four occasions shining high-powered flashlights at vehicles departing the building. Wamsley Decl. ¶ 18. In the second week of September, ICE received multiple reports that drivers leaving the ICE facility were temporarily blinded or disoriented after individuals shined high-powered lights at their vehicles. Id.

On September 9, screenshots were captured of threats on social media to murder ICE officers. Id. ¶ 19 (screenshot of posts stating “fed ain’t nothin but a target practice,” “Execute gestapo on sight,” and “Off The Pigs”). On September 12, protesters photographed an unmarked government vehicle as it was leaving the ICE facility and then posted the photograph to a social media account and provided the location of the vehicle after it left the ICE facility. Id. ¶ 18.

On September 12, there were 8 to 10 protesters at the ICE facility, but a vehicle arrived and unloaded “a large quantity of sticks and bats,” and it was “unknown if they [were] gearing up for Sunday night or just to resupply.” Hughes Decl. Ex. 11. On September 13, there were 40 to 50 people at the ICE facility, and someone had a green laser. Id. Ex. 12. On September 14, FPS increased the number of its officers in anticipation of large numbers of protesters; they had not determined why the bats and sticks had been delivered. Id. Ex. 13. On September 15, 16, and 17, there were groups ranging from 10 to 50 people at the building and some were very energetic. Id. Exs. 14-16. On September 18, “the afternoon started with around 150 plus” protesters outside the ICE facility; later there were 50 to 60 of the “normal agitators.” Id. Ex. 17. They screamed and threw objects at a PPB car. Id. A PPB officer reported that “[t]here does appear to be a large camp set up again even though it was cleaned out earlier in the day.” Id.

On September 19, HSI received a tip that someone planned to detonate a bomb at the ICE facility that day. Wamsley Decl. ¶ 14. Although the threatened attack did not occur, preparing for the possibility required an increase in security inspections and diverted important agency resources. Id. A PPB officer summarized the situation by stating that “FPS agents were busy.” Hughes Decl. Ex. 19. That night, there were 50 protesters, who were throwing objects and running at cars. Id. FPS also reported “black blockers” assaulting people, and FPS provided a video to PPB of a “black blocker” assaulting a journalist with a large sword or stick. Id. There was a disturbance outside the building that day and FPS deployed munitions to break up the crowd. Id. Ex. 18.

On September 20, there were 50 protesters gathered at the ICE facility, and they were “more agitated than most nights.” Id. Ex. 20. There was a disturbance involving protesters and counter-protesters. Id. On September 20, 22, 23, and 24, there were groups of protesters of 10 to 20 people, there was one disturbance involving a protester pepper spraying someone else, one “black blocker” in front of the ICE driveway, but no reported assaults. Id. Exs. 21-24.

On September 25 and 26, there were between 10 to 20 protesters, sometimes yelling at police, sometimes minimally active. Hughes Decl. Exs. 25, 26. On September 27, there were groups of 50 to 60 protesters outside the facility, “most of them were blocking the facility’s driveway in anticipation of the shift change” and refusing to leave, and “[m]any were dressed in bloc, with helmets, masks, and other PPE.” Id. Exs. 27, 28. FPS requested assistance from PPB to get vehicles out of the garage for a shift change. Id. Ex. 28. FPS had the resources to clear the driveway, but they were concerned about protesters shining bright lights at them as they drove out. Id. PPB did not assist because officers thought the request might conflict with their previous directions to not assist with facilitating the movement of vehicles in and out of the facility. Id. PPB also did not have the resources to safely accomplish the task with a “crowd of this size and demeanor.” Id.

On September 28, after the announcement of the federalization order, there were 35 to 200 protesters outside the ICE facility throughout the day. Schoening Decl. Exs. B, C.5 PPB attempted to de-escalate conflicts several times, and made two arrests. Id. Ex. B. PPB did not provide traffic control resources, in part, “because of the confrontational nature of individuals at the ICE rally,” and “the static nature of the event.” Id. There were “a couple of confrontations” when federal agents were facilitating vehicles entering or exiting the facility. Id.

On September 29, an FPS officer had to deploy a pepper spray fogger against a protester who refused to comply with several orders to leave federal property. Cantu Decl. ¶ 13. Later that night, FPS warned protesters to clear the facility driveway, and a few minutes later they had to deploy pepper spray foggers at protesters who were shining high-powered strobe lights in the officers’ faces as they were attempting to push the crowd back from the driveway. Id. One of the officers who had the strobe light directed at him experienced pain and disturbances in his vision. Id.

“Normally, FPS would be able to rely on the [PPB] to assist with large scale law enforcement operations related to federal facilities in Portland. FPS has been informed by PPB, however, that they will only respond to ‘life/safety’ situations, but not anything immigration related.” Id. ¶ 15. But when FPS called PPB multiple times on September 9 to request assistance with a woman who was being physically assaulted by a group of 12 to 13 protesters outside the ICE facility, PPB would not intervene or assist. Id. Eventually, FPS permitted the woman onto federal property for her safety and later pushed back protesters so that the woman could escape from the protesters attempting to follow her. Id.

“In June 2025, PPB Police Chief Bob Day spoke publicly about avoiding any actions that might show ‘perceived or actual support’ for immigration agents” after PPB officers cleared a blockade of the ICE driveway to let an empty transport van pass. Wamsley Decl. ¶ 25. In emails summarizing these incidents, PPB officers report that no officers are available to respond to disturbances at the ICE facility. Hughes Decl. Ex. 15 (September 16 email explaining PPB has “no officers to go or call” to respond to disturbance at ICE facility); Id. Ex. 18 (September 19 email stating the PPB had “few to zero officers available” to address a “verbal disturbance” in front of the ICE building); Id. Ex. 19 (September 20 email stating that PPB “would not be able to address the call or make an arrest with the resources we have”).

The lack of support from PPB has forced FPS to rely almost exclusively on ICE and SRT for assistance in preventing violent protesters from attacking officers and the facility in which they work. Cantu Decl. ¶ 16. FPS deployed 115 officers from other regions, working on a 24/7 basis, to maintain order at the Lindquist Federal Building. Id. ¶ 18. And FPS has thus far spent over two million dollars on overtime pay and expenses since June 2025 to maintain order in the wake of these violent protests. Id. ¶ 19. In September, ICE detailed three SRT certified Special Agents to ERO’s SRT efforts in Portland. Wamsley Decl. ¶ 15. Detailing HSI SRT Special Agents to Portland diminishes HIS’s operational capabilities nationwide. Id.

B. The President’s Order Federalizing the National Guard

On September 26, after more than three months of disturbances at an ICE facility in Portland, DHS sent a memorandum to DoD requesting “immediate and sustained assistance” “to safeguard federal personnel, facilities, and operations in the State of Oregon.” In that memorandum, DHS explained that the Lindquist Building had “come under [a] coordinated assault by violent groups” that “are actively aligned with designated domestic terrorist organizations” and that DoD’s assistance was required to prevent the obstruction of “lawful federal enforcement actions” and “to safeguard federal personnel, facilities, and operations in the State of Oregon.”

On September 27, President Trump cited the DHS request and directed Secretary Hegseth to coordinate the deployment of troops to Portland to protect the facility. Donald J. Trump, (@realDonaldTrump), Truth Social (Sept. 27, 2025 at 7:19 a.m. ET), https://truthsocial.com/@realDonaldTrump/posts/ 115276694936263266 (Truth Social Post). DoD contacted the Oregon Adjutant General, who serves as the Director of the Oregon Military Department and Commander of the Oregon National Guard, to request that he deploy National Guard members in a non-federalized status to assist. Or. Rev. Stat. § 396.160. The Governor rejected this request.

On September 28, Secretary Hegseth issued a mobilization memorandum directive stating that “200 members of the Oregon National Guard will be called into Federal Service effective immediately for a period of 60 days.” The mobilization memorandum relied, in part, on a June 7 memorandum President Trump had issued temporarily authorizing the mobilization of the National Guard to protect DHS and other U.S. government personnel on federal property. In the June 7 memorandum, President Trump cited several incidents of violence and disorder taking place across the country and invoked § 12406 to federalize National Guard members to temporarily protect ICE and other government personnel, “including the enforcement of federal law.”6

C. Plaintiffs’ Complaint and TRO Application

In response, Plaintiffs-Appellees, the State of Oregon, the City of Portland, and the PPB (collectively, Plaintiffs), immediately sued, alleging that Defendants’ actions—federalizing the Oregon National Guard—were ultra vires, and violated the Tenth Amendment to the United States Constitution, the Administrative Procedure Act, the Posse Comitatus Act, and separation of powers.7

Plaintiffs also applied for a TRO, which the district court issued on October 4, after a hearing. See Oregon v. Trump, No. 3:25-CV-1756-IM, 2025 WL 2817646, at *15 (D. Or. Oct. 4, 2025). The TRO temporarily enjoins Defendants from implementing Defendants’ September 28, 2025 memorandum ordering the federalization and deployment of members of the Oregon National Guard. Id. The district court determined that Plaintiffs are likely to succeed on their claim that the President’s order is ultra vires because it was not authorized by §§ 12406(2) or (3), and thus also violates the Tenth Amendment. Id. at *9–14. That same day, Defendants appealed and filed an emergency motion to stay the TRO pending appeal.

Later that day, Oregon sought a second TRO in the district court. The initial TRO only enjoined Defendants from federalizing and deploying the Oregon National Guard. Oregon, 2025 WL 2817646, at *15. After the district court entered its initial TRO, Defendants sent approximately 100 federalized members of the California National Guard to the Portland Air National Guard base with plans to send additional troops that same day. The district court issued a second TRO enjoining Defendants from deploying any federalized members of the National Guard in Oregon. Oregon v. Trump, No. 3:25-CV-1756-IM, 2025 WL 2823653 (D. Or. Oct. 5, 2025). Defendants have not appealed the second TRO.8

III. STANDARD OF REVIEW

We review Defendants’ request for a stay pending appeal under the following four factors: “(1) whether the stay applicant has made a strong showing that [it] is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Nken v. Holder, 556 U.S. 418, 426 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). “The first two factors . . . are the most critical.” Id. at 434.

A. Binding Effect of Newsom

Before addressing the stay factors, we note that our analysis here is circumscribed by our legal conclusions on the same issues in Newsom. See E. Bay Sanctuary Covenant v. Biden, 993 F.3d 640, 660 (9th Cir. 2021) (explaining that published motions panel orders “may be binding as precedent for other panels deciding the same issue”); Lair v. Bullock, 798 F.3d 736, 747 (9th Cir. 2015) (explaining that a published motions panel’s decision “binds future panels the same as does a merits panel’s published opinion”).

First, in Newsom, we held that neither the political question doctrine nor the statutory text precludes judicial review of the President’s decision to invoke § 12406 to federalize the National Guard. 141 F.4th at 1045–47. Thus, Plaintiffs’ claims are justiciable and the President’s decision to federalize the National Guard is subject to judicial review. At oral argument, Defendants’ counsel noted that they intend to challenge these conclusions on the merits of the preliminary injunction in Newsom. But they agreed that we could view these holdings as binding and argued that Defendants would nonetheless prevail.

Second, we addressed “whether we owe [the President’s] determination deference, and if so, how much?” Id. at 1047. In answering that question, we considered the history of Congress’s delegation of its constitutional power to call forth the militia to the President and the cases interpreting those delegations, particularly Martin v. Mott, 25 U.S. (12 Wheat.) 19 (1827), and Luther v. Borden, 48 U.S. (7 How.) 1 (1849). Newsom, 141 F.4th at 1047–51. We held that “our review of the President’s determinations in this context is especially deferential,” id. at 1047, “we must give a great level of deference to the President’s determination that a predicate condition exists,” id. at 1048, and “the President’s determination that an exigency exists [must] be given significant deference,” id. at 1050. But we also rejected the government’s argument that the President could federalize the National Guard “on no evidence whatsoever,” and that courts could not “review a decision that was obviously absurd or made in bad faith.” Id. at 1050. In sum, we concluded that courts may review the President’s determination “to ensure that it reflects a colorable assessment of the facts and law within a ‘range of honest judgment.’” Id. at 1051 (quoting Sterling, 287 U.S. at 399).

Finally, in Newsom, we rejected the argument that § 12406(3) requires “total or near total interference” with the President’s ability “to execute the laws of the United States.” Id. We held that while “minimal interference with the execution of laws is, by itself, [not] enough to justify invoking § 12406(3),” the statute does not require that “the President be completely precluded from executing the relevant laws of the United States in order to call members of the National Guard into federal service, nor does it suggest that activation is inappropriate so long as any continued execution of the laws is feasible.” Id.

Applying these legal conclusions, we turn to the first and most important Nken factor, whether Defendants have made a strong showing that they are likely to succeed on the merits.

B. Likelihood of Success on the Merits

The President’s authority to federalize the National Guard is statutory, not constitutional. Newsom, 141 F.4th at 1046. The Constitution vests the power to call forth the militia in the legislative branch. U.S. Const. art. I, § 8, cl. 15. Under that enumeration, Congress has enacted several statutes, including § 12406, which delegate the calling forth power to the President. Section 12406 authorizes the President to federalize members of the National Guard in three circumstances, providing:

Whenever—

(1) the United States, or any of the Commonwealths or possessions, is invaded or is in danger of invasion by a foreign nation;

(2) there is a rebellion or danger of a rebellion against the authority of the Government of the United States; or

(3) the President is unable with the regular forces to execute the laws of the United States;

the President may call into Federal service members and units of the National Guard of any State in such numbers as he considers necessary to repel the invasion, suppress the rebellion, or execute those laws. Orders for these purposes shall be issued through the governors of the States or, in the case of the District of Columbia, through the commanding general of the National Guard of the District of Columbia.

10 U.S.C. § 12406.

Defendants argue that the circumstances at the Portland ICE facility from June to September establish that the President’s actions were authorized by two sections of the statute: first, by § 12406(2), which authorizes federalization of the National Guard to address a “rebellion or danger of a rebellion against the authority of the Government of the United States”; and second, by § 12406(3), which authorizes federalization of the National Guard when “the President is unable with the regular forces to execute the laws of the United States.” Because we determine, as we did in Newsom, that § 12406(3) allows us to conclude that Defendants are likely to prevail, we do not address § 12406(2).

1. Plaintiffs’ Ultra Vires Claim

Whether Defendants have shown a likelihood of success on the merits of Plaintiffs’ ultra vires claim turns on whether the President’s determination under § 12406(3), that he was “unable with the regular forces to execute the laws of the United States,” “reflects a colorable assessment of the facts and law within a ‘range of honest judgment.’” Newsom, 141 F.4th at 1051 (quoting Sterling, 287 U.S. at 399). While the district court cited this highly deferential standard, Oregon, 2025 WL 2817646, at *9, it erred by failing to apply it. Instead, the district court substituted its own assessment of the facts for the President’s assessment of the facts. This is the opposite of the significantly deferential standard of review that applies to the President’s decision to invoke § 12406(3) and federalize members of the National Guard. See Newsom, 141 F.4th at 1051.

First, the district court erred by determining that the President’s “colorable assessment of the facts” is limited by undefined temporal restrictions and by the district court’s own evaluation of the level of violence necessary to impact the execution of federal laws. Thus, the district court determined that it would apply Newsom’s deferential “colorable basis” standard to the facts “as they existed at the time [the President] federalized the National Guard.”9 Oregon, 2025 WL 2817646, at *9. The district court then discounted the violent and disruptive events that occurred in June, July, and August, including the resulting closure of the ICE facility for over three weeks in June and July, id. at *1, 4, and focused on only a few events in September, id. at *10. Thus, the district court discounted most of the evidence of events in Portland from June through September.10

Section 12406 contains no such limitations. Instead, by its plain text, § 12406(3) requires only a determination that “the President is unable with the regular forces to execute the laws of the United States.” The statute delegates the authority to make that determination to the President and does not limit the facts and circumstances that the President may consider in doing so. Indeed, the inherently subjective nature of this evaluation demonstrates that the President has the authority to identify and weigh the relevant facts under § 12406(3). The President can, and should, consider the totality of the circumstances when determining whether he “is unable with the regular forces to execute the laws of the United States.”

Rather than reviewing the President’s determination with great deference, the district court substituted its own determination of the relevant facts and circumstances. That approach is error under Newsom. Our dissenting colleague takes a similar approach. The dissent argues that the plain text of § 12406(3) confines the President’s inquiry to present circumstances because the statute lacks any forward-looking language, as seen in §§ 12406(1) (“invaded” or “in danger of invasion”), and (2) (“rebellion” or “danger of rebellion”). Dissent at 8–11. That observation, however, does not resolve whether the President may also consider past events. All three statutory prongs are the same as to whether prior events may be considered. And the President is entitled, and indeed, duty-bound, to evaluate the entire context of events leading up a decision to invoke § 12406 as part of the constitutional command to “take Care that the Laws be faithfully executed.” U.S. Const. art. II, § 3. Considering the totality of the circumstances, there is a colorable basis for the President’s determination that he is unable with regular forces to execute the laws of the United States.

Second, the district court erred by placing too much weight on statements the President made on social media. Oregon, 2025 WL 2817646, at *11. The district court interpreted President Trump characterizing Portland as “War ravaged,” as the equivalent of the President “ignoring the facts on the ground.” Id. As such, the district court relied on these statements to disregard other facts that do “reflect[] a colorable assessment of the facts and law within a range of honest judgment.” Id. (citation modified).

This was error for two reasons. For one, under Newsom, we consider whether the facts provide a colorable basis to support the conclusion that the executive branch is unable with regular forces to execute the law. Even if the President may exaggerate the extent of the problem on social media, this does not change that other facts provide a colorable basis to support the statutory requirements.11 For another, the President’s social media post characterizing Portland as “War ravaged” may reasonably be viewed as a part of his assessment under § 12406(2)’s “rebellion” prong, which we do not reach. Id.

Moreover, § 12406(3) cannot be analyzed solely based on external protests and violence. We must also consider the executive’s internal assessment of its ability to enforce the laws. The record reflects that 115 FPS officers—nearly 25% of FPS officers nationwide—were diverted to Portland. The President may reasonably rely on this evidence in determining whether he is unable to execute the law. 12 The dissent only reaches a different conclusion by characterizing this evidence as “staffing difficulties” and committing the same error as the district court in discounting, minimizing, and discrediting Defendants’ undisputed evidence on this point.13 Dissent at 17–21. While Defendants bear a burden, the dissent demands (as did the district court) far more than what is required to support a colorable assessment. And that is error under Newsom.

Third, the district court erred by concluding that the surge deployment of FPS officers and HSI SRT officers from around the country to respond to the months of violence at the ICE Facility established that “‘the regular forces,’ i.e. FPS and additional federal law enforcement, were able to execute the laws of the United States.” Oregon, 2025 WL 2817646, at *10. This conclusion can be reached only by overlooking the record. While the record establishes that FPS deployed many officers to Portland—nearly 25% of the agency’s capacity, working 12-hour shifts, seven days a week—it also establishes that these deployments were irregular and unsustainable. Cantu Decl. ¶ 4.

Defendants submitted sworn declarations from Robert Cantu, the Deputy Director of the Federal Protective Services, and Cammilla Wamsley, the DHS Field Office Director who oversees and directs ICE’s enforcement of federal immigration laws in Oregon, Washington, and Alaska, stating that these surges and assignments strained resources and are unsustainable:

While FPS can surge its Inspectors to respond to emergency situations, it is not resourced to provide a large-scale response to ongoing civil unrest or sustained attacks on federal facilities or federal employees working in those facilities. Cantu. Decl. ¶ 4.

FPS’s efforts to contain the violence associated with the protests in Portland have placed an unreasonable strain on the FPS law enforcement community as well as on FPS’s limited resources. Id.

28 25-6268

¶ 19.

The sustained violence associated with the protests in Portland has required FPS Region 10 to deploy officers from the other FPS Regions. To date, 115 FPS officers have had to deploy to Portland to maintain a 24/7 operational tempo. Id. ¶ 18.

[T]he continued deployment of FPS officers in response to the immigration protests stretches an already thin force beyond what is safe and effective. While FPS officers are expected to respond to emergencies after hours and on weekends and holidays, they are not expected to work 12-hour shifts, seven days a week, which has become the norm. Id. ¶ 6.

Because of these concerns and the volatility of the situation in Portland, FPS has been forced to drastically alter its operations in a manner which is currently unsustainable. Id. ¶ 14.

This situation has resulted in FPS having to rely almost exclusively on ICE SRT for assistance in preventing the violent protesters from attacking immigration officers and the ICE facility where they work. . . . [ICE SRT’s] continued assistance to FPS will become a drain on those resources and is not a practical ongoing solution. Id. ¶ 16.

HSI detailed three Special Response Team (SRT) certified Special Agents to ERO’s SRT efforts in Portland in September 2025, and intends to continue to support as directed. . . . Detailing HSI SRT Special Agents to Portland for these purposes diminishes HSI’s operational capabilities nationwide due to personnel limitations. Wamsley Decl. ¶ 15.

[ICE has deployed] officers and agents from around the country, including SRT personnel, to assist FPS in Portland beginning on or about June 9, 2025. This diversion of personnel and resources to assist FPS erodes ICE’s ability to enforce federal immigration laws in the State of Oregon and across the country. Id. ¶ 20.

Disregarding this uncontested evidence, the district court rejected Defendants’ argument that the President had a colorable basis to invoke § 12406(3) because—

29 25-6268 under its own assessment of the facts—these deployments “demonstrate . . . that federal law enforcement officers, unaided by any military forces, were capable of not only ‘quelling’ the violence in June but also ‘prevent[ing] it through September.” Oregon, 2025 WL 2817646, at *10 (alteration in original) (citing Newsom, 141 F.4th at 1051). But the district court turned the proper analysis on its head. By interpreting the modest improvements created by the surge as demonstrating that the President did not have a colorable basis for invoking § 12406(3), and discounting that the surge of FPS officers was inherently irregular, the district court accorded no deference to the President’s determination that he could not execute federal laws with regular forces.14

Thus, considering the totality of the circumstances from June through September, and applying the required highly deferential standard of review, we conclude that the President’s assessment of the situation in Portland—specifically,

14 The dissent also argues that the FPS surge cannot be considered as evidence that regular forces cannot enforce the law, citing a lack of factual support. Dissent at 18. But in the very next paragraphs, the dissent explains that it is discounting the evidence provided by Director Cantu when making this assertion. Id. at 18–21. Our colleague in dissent and the district court cannot summarily dismiss Director Cantu’s sworn and undisputed statements. The issue is whether the President, relying on that evidence, could conclude that he “is unable with the regular forces to execute the laws of the United States,” § 12406(3), and whether that determination reflects a “colorable assessment of the facts and law within a range of honest judgment,” Newsom, 141 F.4th at 1051 (citation modified). The President’s determination satisfies that deferential standard.

the threat to federal personnel and property, and the resulting inability to execute federal laws—“reflects a colorable assessment of the facts and law within a ‘range of honest judgment.’” Newsom, 141 F.4th at 1051 (quoting Sterling, 287 U.S. at 399). Therefore, Defendants are likely to succeed on the merits of Plaintiffs’ ultra vires claim.15

2. Plaintiffs’ Tenth Amendment Claim

The district court also concluded that Plaintiffs are likely to show that Defendants’ federalization of the National Guard violated the Tenth Amendment. Oregon, 2025 WL 2817646, at *14. The district court correctly explained that “[n]o matter the context, the President’s authority to act necessarily ‘stem[s] either from an act of Congress or from the Constitution itself.’” Id. at *13 (citations omitted). And because the Tenth Amendment reserves to the States all rights and powers “not delegated to the United States by the Constitution,” when the President federalizes the National Guard absent an authorization by Congress acting under its militia clause power, he necessarily violates the Tenth Amendment. Id. at *13–14. But as we have already concluded, Defendants are likely to establish that the President lawfully federalized the National Guard as Congress authorized in § 12406(3). We therefore conclude that Defendants are also likely to succeed on Plaintiffs’ Tenth Amendment claim.

C. The Remaining Stay Factors

Apart from the likelihood of success on the merits, we consider whether the applicant will be irreparably injured absent a stay, whether a stay will substantially injure the other parties, and where the public interest lies. Nken, 556 U.S. at 426 (citation omitted). The last two factors “merge when the Government is the opposing party.” Id. at 435; E. Bay Sanctuary Covenant, 993 F.3d at 668 (“When the government is a party, the last two factors (equities and public interest) merge.”).16

In Newsom, under similar circumstances, we concluded that “[b]oth irreparable harm and the public interest weigh in favor of Defendants, who have an uncontested interest in the protection of federal agents and property and the faithful execution of law.” 141 F.4th at 1054. Here, like Newsom, the undisputed facts show that protesters damaged a federal building, leading to its closure for over three weeks, attempted to burn the building down, placed chains on the doors, attempted to breach the front door of the building and broke the front glass door, threw objects at the building, including rocks, sticks, and a mortar, and launched M80 fireworks at federal officers, assaulted federal officers, shined lasers at officers’ eyes, and doxed federal officers. “The federal government’s interest in preventing incidents like these is significant.” Id. (citing United States v. Bader, 698 F.2d 553, 555 (1st Cir. 1983) (“It is well established that the need to safeguard the normal functioning of public facilities is a ‘substantial government interest’ . . . .”); United States v. Shiel, 611 F.2d 526, 528 (4th Cir. 1979) (“The legitimacy of the government’s interest, in the abstract, of insuring the public’s compliance while in or on government property with proper directions of law enforcement officers . . . [is] apparent.”)).

The dissent, however, argues that Defendants face no irreparable harm because the second TRO, which is not before us, still prevents Defendants from deploying the National Guard. See Dissent at 29–31. But this argument lacks merit. The district court stated that it granted the second TRO based on the same legal reasoning it provided in its order issuing the first TRO. Defendants are thus correct that the first TRO and the second TRO rise or fall together on the merits of the issues raised in this motion for a stay pending appeal. And the dissent does not challenge that common sense conclusion.

Whether the second TRO could be extended to preclude the deployment of members of the California National Guard is a separate issue. But this provides no basis for the dissent’s conclusion that Defendants are not harmed when they are prevented from enforcing federal immigration law by federalizing and deploying members of the Oregon National Guard.

Plaintiffs argue that the other stay factors weigh in their favor. While they acknowledge Defendants’ interest in protecting federal agents and property, and the execution of federal law, they argue that Defendants will not be irreparably harmed in the absence of a stay because any threats to federal agents and property have been remedied by federal civilian law enforcement, with support from state and local law enforcement. But the record establishes that the continued deployment of FPS officers to Portland is unsustainable, see supra pp. 27–29, and state and local law enforcement have been unable or unwilling to assist the government’s efforts to protect federal personnel and property at the ICE facility, see supra pp. 14–16.

Finally, Plaintiffs also argue that the “irreparable harm” factor weighs in their favor because the federalization of the National Guard, over the Governor’s objections, will cause a “substantial sovereign injury,” result in the diversion of Oregon National Guard members from their state responsibilities, and threaten irreparable economic and financial harms to Plaintiffs and their residents. Plaintiffs argue that the “public interest” factor weighs against a stay because of “the strong resistance of Americans to any military intrusion into civilian affairs.”

As we stated in Newsom, while Plaintiffs have significant interests, their arguments are, “in essence, a merits argument that we have already resolved. The Constitution assigns the power to ‘call[] forth the Militia’ to Congress, and Congress has delegated portions of that power to the President.” 141 F.4th at 1055 (alteration in original) (quoting U.S. Const. art. I, § 8, cl. 15). Therefore, Plaintiffs’ arguments on the remaining stay factors rise and fall with their argument on the merits. Because we have concluded that Defendants are likely to succeed on the merits by establishing that the President lawfully invoked § 12406(3), Plaintiffs’ arguments on the other factors fail.

IV. CONCLUSION

For these reasons, we GRANT Defendant’s motion for a stay pending appeal.

FILED Oregon v. Trump, No. 25-6268 OCT 20 2025 MOLLY C. DWYER, CLERK R. NELSON, J., concurring: U.S. COURT OF APPEALS

Newsom proceeded to likelihood of success on the merits, it implicitly found standing for the state plaintiffs in that case. Id. It likely got both issues wrong.

As a matter of first principles, I doubt we are bound by Newsom. We have previously held that a published opinion, even at the preliminary injunction stage, binds future panels on pure issues of law. See E. Bay Sanctuary Covenant v. Biden, 993 F.3d 640, 660 (9th Cir. 2021); Ranchers Cattlemen Action Legal Fund United Stockgrowers of Am. v. U.S. Dep’t of Agric., 499 F.3d 1108, 1114 (9th Cir. 2007). But we have never held that an order granting a stay pending appeal of a TRO binds future panels, even on pure issues of law. And it would be odd if the Newsom merits panel would be bound by its initial conclusions made during the truncated stay pending appeal stage. Similarly, it does not make sense that another panel hearing a stay pending appeal on a different TRO should be bound by Newsom. That legal nuance can wait, however, to be decided. As the government requests at this stage, we follow the holdings in Newsom. See Per Curiam at 20. Still, current Supreme Court precedent dictates that Newsom was wrong to hold that the President’s determinations under § 12406 are reviewable by the courts.

A

The government’s argument that § 12406 may not be reviewed by the federal courts tracks Supreme Court caselaw. The government rests its argument on Martin v. Mott, 25 U.S. 19 (1827), and Luther v. Borden, 48 U.S. 1 (1849). Under those two *3 cases, whether the President has lawfully invoked § 12406 presents a political question the courts cannot review.

Martin arose in the wake of the War of 1812 when President Madison invoked the Militia Act of 1795 to mobilize the nation for war. See 25 U.S. at 28. Congress enacted the 1795 Act pursuant to its power to call forth the militia. See U.S. Const. art. I, § 8, cl. 15. To provide the men necessary to defend the nation from British invasion, President Madison directed the governor of New York to raise a militia. Jacob E. Mott, a private in the New York Militia, refused his superior’s order to assemble in support of the war effort, was court martialed, and found guilty of refusing to enter service. See Martin, 25 U.S. at 33. Mott appealed and the Supreme Court of Judicature of the State of New York granted him relief. It concluded that President Madison had not properly adjudged that there “was an invasion or imminent danger of invasion” to initiate the militia draft under the 1795 Act. See id. at 32.

The case then reached the United States Supreme Court. Writing for the Court, Justice Story reversed the lower court. Id. at 38–39. He explained that “the authority to decide whether the exigency has arisen, belongs exclusively to the President, and that his decision is conclusive upon all other persons.” Id. at 30. And then in Luther, also interpreting the 1795 Militia Act, Chief Justice Taney writing for the Court held that “the power of deciding whether [an] exigency had arisen” “is *4 given to the President.” 48 U.S. at 43. The Supreme Court thus twice held that the 1795 Militia Act (and by extension § 12406 which uses similar language and is a successor of that Act) is unreviewable by the federal courts. Compare Act of Feb. 28, 1795, § 2, 1 Stat. at 424–25 to 10 U.S.C. § 12406.

The Supreme Court has said that in our constitutional structure, the safeguards to “guard against usurpation or wanton tyranny” through the Militia Acts is “the frequency of elections, and the watchfulness of the representatives of the nation.” Martin, 25 U.S. at 32. Reasonable minds will disagree about the propriety of the President’s National Guard deployment in Portland. But federal courts are not the panacea to cure that disagreement—the political process is (at least under current Supreme Court precedent).

Further cementing the lack of judicial review, Plaintiffs seek to enjoin the federalization and deployment of National Guardsmen through non-statutory ultra vires review. But the Supreme Court has held such review “applies only when an agency has taken action entirely in excess of its delegated powers and contrary to a specific prohibition in a statute.” NRC v. Texas, 605 U.S. 665, 681 (2025) (cleaned up). And no such specific prohibition exists within the text of § 12406. So Martin, Luther, and the limits the Supreme Court has placed on ultra vires review, all counsel towards us lacking the power to engage in judicial review.

Newsom distinguished Martin and Luther, in part by relying on Sterling v. Constantin, 287 U.S. 378, 399–400 (1932). But Sterling only dealt with a Governor of a state, not the President of the United States. Id. In Sterling, Governor Ross S. Sterling of Texas issued a gubernatorial proclamation that certain Texas counties were in a state of insurrection and had the adjutant general of the state enforce martial law in those geographic regions. 287 U.S. at 387. In that context, the Supreme Court said that while Martin and Luther establish that the President’s determination of an exigency requiring the militia was conclusive, a governor is permitted within a “range of honest judgement” to determine such exigencies. Id. at 399–400.

But Sterling only demonstrated that a governor is not entitled to the same level of deference or judicial review as a President. Id. Sterling also arose with a declaration of martial law. See id. at 387. Sending federal troops to protect a federal building, on its face, is far afield from martial law. Sterling did not hold that the standard set for the President in Martin and Luther no longer bound the federal courts. Id. Thus, Newsom erred by relying on Sterling, rather than Martin and Luther, both of which hold that when the President invokes the 1795 Militia Act or its eventual successor § 12406, that decision is unreviewable by the federal courts.

One could qualify these cases. Justice Story observed in Martin that this awesome presidential power was meant to apply during “sudden emergencies, upon great occasions of state, and under circumstances which may be vital to the existence *6 of the Union.” 25 U.S. at 30. The War of 1812 was such a time when the existence of the American Union was at stake. See GORDON WOOD, EMPIRE OF LIBERTY: HISTORY OF THE EARLY REPUBLIC, 1789-1815 659–701 (2009) (Wood). Congress had explicitly declared war, and any argument that the United States was not being invaded was implausible. See Library of Congress, The War of 1812, https://www.loc.gov/item/today-in-history/june-18/; see also Library of Congress, Image 793 of U.S. Statutes at Large, Volume 2 (1799-1813), 6th through 12th Congress, https://www.loc.gov/resource/llsalvol.llsal_002/ ?sp=793&r=0.49,1.204,0.983,0.471,0 (reproducing the declaration of war).

And Luther arose out of an outright civil war taking place in Rhode Island, the Dorr Rebellion. See 48 U.S. at 35–36; see also AKHIL AMAR, AMERICA’S CONSTITUTION: A BIOGRAPHY 369 (2005) (Amar). The Dorr Rebellion arose in 1841 because unlike the other states, Rhode Island did not adopt a new state constitution during the American Revolution and operated instead under the colonial charter granted by King Charles II in the 1660s. Amar at 369. Civil war broke out when dissatisfied citizens tried to hold their own constitutional convention, creating a new state constitution to extend the right of suffrage to a greater portion of the population. Luther, 48 U.S. at 36. Those citizens took up arms against the charter government claiming that they were the new rightful state government. Id.

Both Martin and Luther could be cabined to distinguish from the events here. Though both cases and the events in Portland represent threats to domestic tranquility, Martin and Luther interpreted the 1795 Act in the context of existential threats to the Union (in Martin) and outright civil war within a state (in Luther). The situation in Portland involves violent protestors preventing the federal government from executing the laws of the United States around one localized federal building with regular forces.

Ultimately, Martin and Luther categorically hold that the President’s decision in this area is absolute. See Martin, 25 U.S. at 30; Luther, 48 U.S. at 43. “As a circuit court, even if recent Supreme Court jurisprudence has perhaps called into question the continuing viability of its precedent, we are bound to follow a controlling Supreme Court precedent until it is explicitly overruled by that Court.” Garfias-Rodriguez v. Holder, 702 F.3d 504, 517 (9th Cir. 2012). Whether Martin and Luther can be distinguished by Sterling or on other grounds is therefore an issue for the Supreme Court to decide. Until the Supreme Court says otherwise, we are bound to follow its precedent. See Mallory v. Norfolk S. Ry. Co., 600 U.S. 122, 136 (2023); Crowe v. Oregon State Bar, 989 F.3d 714, 725 (9th Cir. 2021).

B

Plaintiffs also have serious standing issues that have not been addressed given the preliminary posture of this case (as in Newsom). When federal courts have *8 opined on similar issues, the party bringing the lawsuit usually seeks either a remedy at law or habeas relief. See, e.g., Martin, 25 U.S. at 28 (seeking a remedy at law); Ex parte Milligan, 71 U.S. 2, 107 (1866) (seeking habeas relief). A court may incidentally declare that executive branch actions involving the military were illegal in such cases. But a court does so only to grant a plaintiff a remedy at law.

Plaintiffs bring claims solely for injunctive and declaratory relief—asking us to declare the federalization and deployment unlawful. Plaintiffs allege that federalizing the Oregon National Guard was: (1) ultra vires under § 12406; (2) prohibited by the Posse Comitatus Act, 18 U.S.C. § 1385; (3) an infringement on Oregon’s police power; (4) designed to coerce the city and the state “to accede to the Trump administration’s political prerogatives;” (5) a violation of equal state sovereignty; (6) unlawful administrative action; and (7) contrary to the separation of powers. Some of these claims are bare allegations that the executive violated the law, which cannot be a standalone basis for an injunction. See Trump v. CASA, Inc., 606 U.S. 831, 858 (2025). And the remainder are claims where Plaintiffs allege injury to sovereign political rights as a state within the Union.

But injuries to Plaintiffs’ sovereign political rights are not enough to confer standing. Federal courts only issue injunctive relief when “private rights or private property [are] infringed,” not to protect “the rights of sovereignty, of political jurisdiction, of government, of corporate existence as a State, with all its *9 constitutional powers and privileges.” State of Ga. v. Stanton, 73 U.S. 50, 77 (1867). We thus may not be empowered to issue injunctive relief here, meaning Plaintiffs’ claims lack redressability. See TransUnion LLC v. Ramirez, 594 U.S. 413, 423 (2021).

Plaintiffs also lack an injury in fact because all political rights Oregon may have had were ceded to the federal government during Constitutional ratification. The Militia Clause grants Congress the power “To provide for calling forth the Militia to execute the Laws of the Union, suppress Insurrections and repel Invasions.” U.S. Const. art. I, § 8, cl. 15. States admitted to the Union after the initial ratification acceded to the Constitution, including delegated powers. And Congress enacted § 12406 under one of its delegated constitutional powers. So an illegal federalization and deployment may cause an injury in fact against Congress, but not the states. TransUnion, 594 U.S. at 423. This raises serious issues over the court’s ability to proceed to the merits of Plaintiffs’ claims. The parties have not fully addressed these questions. But before a panel proceeds to the merits, we should assure ourselves Plaintiffs have standing sufficient that we may decide a proper case or controversy.

II

Assuming the President’s determination is reviewable, it is reinforced by the historical tradition of similar presidential actions.

A

Our conclusion in the Per Curiam order that the President’s determination under § 12406(3) is likely lawful is also supported by the historical tradition established by the early Militia Acts. Ultimately, § 12406 is part of a statutory lineage stretching back to the Militia Act of 1792 designed to enable one of the core functions of the Constitution—maintaining the integrity of the Union.

We must therefore be careful when we interpret statutes part of the lineage of the Militia Act of 1792, all of which enable the President to carry out this solemn constitutional function. It requires special justification found in the statutory text in these types of statutes to narrow their scope from Founding Era understandings. Otherwise, we can presume that such a provision carries the “old soil with it,” and rely on Founding Era understandings to determine whether the President lawfully invoked § 12406. United States v. Trumbull, 114 F.4th 1114, 1127 (9th Cir. 2024) (Bea, J., concurring); see Whitman v. Am. Trucking Ass., 531 U.S. 457, 468 (2001) (Congress does not “hide elephants in mouseholes”). Therefore, when the President acts within the bounds of this historical tradition, it provides near irrefutable evidence of good faith lawful action. Indeed, it would satisfy an even less deferential standard to the President than the standard adopted in Newsom.

Three instances of domestic unrest during the Founding (collectively, the tax revolts) support the President’s determination under § 12406(3). In each example, *11 the President federalized the militia to ensure that the law could be fully enforced. While early actions involved the enforcement of tax laws, the tradition they establish applies equally to the enforcement of federal immigration laws today.

The first, Shays’s Rebellion—an uprising in Massachusetts that arose because of the state’s efforts to enforce its new tax policies and the ensuing debt crisis that followed—occurred during the Articles of Confederation period. See AKHIL AMAR, THE WORDS THAT MADE US 298–302 (2021) (TWMU). General Washington feared that American republicanism might fail as the “mobbers” tried “to shut down courts” and attacked government officials who “earnestly [sought] to obey [and enforce] the laws that the people themselves had authorized.” Id. at 300. This episode left an indelible mark on the Founders. They designed the Constitution to provide a Union that could safeguard the lives of law-abiding Americans trying to enforce the nation’s laws the next time a Shays’s Rebellion arose. See THE FEDERALIST NO. 6 (Hamilton) (discussing Shays’s Rebellion and offering it as an argument to adopt the proposed Constitution).

And another tax revolt did arise. In the early 1790s, after the adoption of the Constitution, a group of Americans were outraged by the newly enacted federal whiskey tax. TWMU at 382–85. These so called “whiskey insurrectionists” responded, not through the democratic process, but “by brutalizing and terrorizing federal tax collectors and their local supporters with tar and feathers, torches, whips, *12 guns, and more.” Id. at 382.

But this time the Union was prepared, both with the Militia Clause and the Militia Act of 1792. In 1794, President Washington invoked the Militia Act of 1792. Id. Washington “amassed an overwhelming force of some thirteen thousand militia men from neighboring states.” Id. Washington “rode at the head of this awesome military force into central Pennsylvania” and restored law and order by putting down the domestic unrest. Id.

The events of 1794 established that those who oppose enforcement of the laws of the United States must seek democratic change at the ballot box. Not through violence directed at federal government officials trying to enforce the laws of the Union or at fellow Americans who support those laws. Id. at 382. And if malicious actors choose violence over the democratic process, then to ensure “that the Laws be faithfully executed,” and the lives and safety of both federal employees and civilians are maintained, the President can invoke the powers Congress has delegated to him and use militia forces to restore order.

Fries’s rebellion, the third of the tax revolts, was not nearly as massive as the other two tax revolts. In response to an unpopular land tax in 1799, Americans in eastern Pennsylvania attacked tax assessors coming to their property by pouring boiling water on them from their windows and intimidated those federal officials into resigning their jobs. The Fries Rebellion, AMERICAN HERITAGE (Apr. 1999), *13 [https://perma.cc/H8EF-CN4S] (Heritage). All in all, this unrest never exceeded more than 150 men. And many incidents involved isolated individuals in a discrete geographic region engaging in extended violence against federal officials to oppose the tax policy. Id.

The federal marshals in 1800 could not contain the violence in this geographic area and ensure federal law was faithfully executed. Id. So President Adams sent 500 federal militia men to successfully maintain peace during the domestic disturbance. Id.

These events also underscore that historically the solution to presidential overreach when invoking the Militia Acts was the political process. For example, opponents of the Whiskey Rebellion who felt President Washington had acted with excessive force, helped impel the creation of Democratic-Republican societies. See Wood at 198; TWMU at 383. And those societies eventually helped usher in the Election of 1800 when Thomas Jefferson soundly defeated President Adams, toppling the dominant Federalist Party of Washington and Adams. Wood at 198, 276. Fries’s Rebellion similarly hurt President Adams in the Pennsylvania backcountry, with German-American communities rejecting the Federalist Party during the Election of 1800 and in later elections. See PAUL DOUGLAS NEWMAN, FRIES’S REBELLION: THE ENDURING STRUGGLE FOR AMERICAN REVOLUTION 164–78 (2004). The political process was a powerful tool to check presidential action during *14 the early Republic. We should not lose hope in our constitutional structure as a check against presidential overreach today.

We have also seen § 12406 be used for far more anodyne purposes. In 1970, President Nixon issued Proclamation Number 3972, invoking § 12406 in response to a postal strike. He directed his Secretary of Defense to use the National Guard to ensure “that the laws of the United States pertaining to the Post Office Department may be executed in accordance with their terms.” Proclamation No. 3972, 35 Fed. Reg. 5001 (Mar. 23, 1970). For eight days, National Guardsmen delivered about 150,000 letters across 30 cities until the strike abated. See Erin Blakemore, How a Postal Strike Became a National Emergency for Richard Nixon (June 30, 2025), [https://perma.cc/TK8R-83YV]. Once again, the political process rather than the courts served as a check on President Nixon. Congress subsequently conducted a special investigation into the President’s decision to federalize and deploy the National Guard to deliver mail. Id. President Nixon’s actions (which may be less probative of the historical tradition) still illustrate the President’s broad discretion in using § 12406.

President Trump’s September 28 deployment falls within the history and tradition of the early Militia Acts. Individuals within a group of about 200 people have engaged in violent activity in opposition to a single set of laws that carry out federal immigration enforcement. Rather than try to enact political change through *15 the ballot box, they have assaulted the federal officers in Portland who enforce those laws and other Americans who disagree with them on the wisdom of those laws. They tried to burn down a federal building and forced that building to close for three weeks. And sadly, regular federal law enforcement cannot on their own contain the situation. State and local officials have also made clear that they do not support the enforcement of federal immigration law, further impeding the enforcement of federal law. In response, the President invoked § 12406(3). The situation in Portland echoes the history discussed above, and the federalization and deployment falls within the broad tradition of how the early Militia Acts were used.

B

The Per Curiam order does not address whether the President likely has authority for National Guard federalization and deployment under § 12406(2) when there is “rebellion or danger of a rebellion against the authority of the Government of the United States.” The historical tradition of the early Militia Acts also supports the President’s determination as likely lawful under § 12406(2).

The government urged the district court to construe “rebellion” as “Open resistance or opposition to an authority or tradition.” Rebellion, Black’s Law Dictionary (12th ed. 2024). On the other hand, Plaintiffs urged the district court to define “rebellion” as “Open, organized, and armed resistance to an established government or ruler; esp., an organized attempt to change the government or leader *16 of a country, [usually] through violence.” Id.

The district court adopted an incorrect reading of “rebellion,” providing four criteria. First, it held that a rebellion must not only be violent but also armed. Second, a rebellion must be organized. Third, a rebellion must be open and avowed. And fourth, a rebellion must be against the government as a whole—rather than in opposition to a single law or issue. As the agitators causing issues in Portland do not seek to overthrow the government and are likely opposed to only a single issue (i.e., immigration enforcement), the district court held that § 12406(2) had not been properly invoked. To reach this conclusion, the district court analyzed “rebellion” based on four dictionaries from the late 1800s and early 1900s, near in time with the Militia Act of 1903. For many other statutes, this approach may have sufficed.

But we are better served relying on historical tradition to define the terms of § 12406, including “rebellion,” because this statute is part of the lineage of the Militia Act of 1792.1 Under that tradition, Founding Era Presidents used federal militias to quell domestic disturbances in cases the district court’s definition would render impermissible. For example, the Whiskey Rebellion could be construed as individuals voicing opposition to a single law or issue (i.e., the domestic whiskey tax inciting the rebellion in 1791). See JOHN YOO, CRISIS AND COMMAND 68 (2010).

1 It could also prove valuable in future proceedings to examine a corpus linguistics analysis of the historical meaning of “rebellion.” See Al Otro Lado v. Exec. Off. for Immigr. Rev., 120 F.4th 606, 629–32 (9th Cir. 2024) (R. Nelson, J., dissenting).

Under the district court’s reading, had § 12406(2) been effective in 1794, President Washington would have been acting extralegally by putting down the whiskey insurrectionists.²

And Shays’s Rebellion, like the Whiskey Rebellion, could be reasonably characterized as opposition to a single issue, and thus would stand inapposite to the district court’s definition of “rebellion.”³ Since the Constitution was enacted to allow Congress and the executive to respond to another Shays’s Rebellion, the district court’s definition makes little sense. See THE FEDERALIST NO. 6 (Hamilton). And Fries’s Rebellion, which both saw a similar number of individuals involved in the unrest, and similar types of violence directed at federal officials as the events in

2 The Militia Act of 1792 did not use the term “rebellion” and instead stated “[t]hat whenever the laws of the United States shall be opposed or the execution thereof obstructed, in any state, by combinations too powerful to be suppressed by the ordinary course of judicial proceedings, or by the powers vested in the marshals by this act, the same being notified to the President of the United States, by an associate justice or the district judge, it shall be lawful for the President of the United States to call forth the militia to suppress such combinations, and to cause the laws to be duly executed.” Act of May 2, 1792, ch. 28, §§ 1-2, 1 Stat. 264. But observers at the time understood the Whiskey Rebellion to be a “rebellion” because it constituted a violent opposition to a limited set of federal laws (i.e., the federal whiskey tax). See TWMU at 382. Successor statutes to the 1792 Act codified that idea in through the term “rebellion” in their text. 3 The history of the Whiskey Rebellion also undermines the dissent’s claim facts supporting § 12406 must immediately precede its invocation (i.e., a matter of days). Dissent at 10–11. The violence during the Whiskey Rebellion began in 1791, but President Washington did not invoke his delegated militia powers until 1794. Wood at 136; TWMU at 382.

Portland, serves as an analogous instance of the President invoking his delegated Militia Act powers.⁴ See Heritage. The district court defines “rebellion” in § 12406(2) in a manner that significantly constricts the term from its Founding era understandings in the early Militia Acts. It erred by doing so.⁵

The Per Curiam order correctly addresses the district court’s errors under § 12406(3), which provides the necessary basis for a stay pending appeal. But if domestic unrest today would have been reasonably understood as rebellion during the Founding era, then it should also be understood as “rebellion” under § 12406(2).

4 The dissent concludes that because only a fraction of the population in Portland are involved in the unrest, it cannot be considered a “rebellion” under § 12406. Dissent at 26–27. Fries’s Rebellion contradicts the dissent’s narrative—as does common sense. Section 12406 allows the President to respond to a “rebellion” whatever the number of people involved in that rebellion. The number rebels may be relevant to the President’s proportionate response. But it has little bearing on whether a rebellion exists. 5 The dissent does not defend the district court’s definition of “rebellion,” tacitly admitting that such a definition does not comport with § 12406 or its predecessor statutes. Dissent at 22. Instead, it wrongly characterizes the events in Portland as quintessential First Amendment Activity under cases like Tinker v. Des Moines, 393 U.S. 503, 508 (1969). But assaulting federal immigration officials, setting fire to the Lindquist Building, and firing M80 fireworks at FPS officers is a far cry from students peacefully wearing black armbands to school as an act of protest. See Tinker, 393 U.S. at 504. The latter is protected First Amendment activity; the former may be seen as rebellion against the United States. A cursory review of the facts shows that the events in Portland go far beyond individuals simply wearing chicken suits or inflatable frog costumes, see Dissent at 1, but represent a profound threat to the lives and safety of federal officials at the Lindquist Building.

III

Finally, the proportionate response to the events in Portland further support that the President likely invoked § 12406 lawfully and in good faith. A grossly disproportionate response could suggest that the executive might be acting pretextually, rather than in response to § 12406’s enumerated predicates.

Such a proportionality principle can be seen throughout our nation’s history. During the Whiskey Rebellion, about 6,000 rebels massed in Pittsburgh and were attacking federal officers, to which President Washington responded with about 13,000 militia members. TWMU at 382. During Fries’s Rebellion, President Adams responded with 500 militia members. Heritage. During the start of the Civil War, President Lincoln called 75,000 members of the militia into service after the fall of Fort Sumter. See Abraham Lincoln, Proclamation 80 (Apr. 15, 1861). And in 1957, President Eisenhower ordered 1,000 Guardsmen to help facilitate an end to school segregation in Little Rock, Arkansas. See National Archives, Executive Order 10730: Desegregation of Central High School (Sept. 23, 1957), [https://perma.cc/A3E2-EA77].

In each instance where the President invoked statutory authority to call forth the militia, he did so in a proportionate manner—with enough force to put down lawless activity and restore order (or to deliver mail), but not so much force that military rule would doubtless nest itself in such territory permanently. See Robert J. *20 Delahunty, Structuralism and the War Powers: The Army, Navy, and Militia Clauses, 19 GA. ST. U. L. REV. 1021, 1030–31 (2003) (discussing the Founding generation’s aversion to large standing armies).

Of course, the precise decision of how many troops to send under § 12406 and its predecessor militia statutes is statutorily left to the discretion of the executive and thus warrants the highest possible deference. Section 12406 provides that when lawfully invoked, “the President may call into Federal service members and units of the National Guard of any State in such numbers as he considers necessary.” The decision of how many troops to call into service has thus always fallen exclusively to the President.⁶ But when proportionality is present, it supports that the President acted in good faith rather than pretextually. District courts must not micromanage the President’s use of statutory powers delegated to him under the Militia Clause. A proportionate response presumptively means that § 12406 has been lawfully invoked.

The federalization and deployment of just 200 National Guardsmen for 60 days is well within a deferential proportional response to support good faith. As

6 During the Constitutional Convention, a delegate suggested that the Constitution should restrict the United States to a standing army not exceeding 5,000 troops at any one time. George Washington’s Mount Vernon, The Humor of George Washington, [https://perma.cc/HT2S-QPVQ]. General Washington quipped that an amendment should then provide that no foreign enemy be allowed to invade the United States at any one time with more than 3,000 troops. Id.

recently as September, the protests amounted to about 200 people, and sporadic bursts of illegal and violent activity still occurred when DHS requested federal support. The President federalized and deployed only 200 Guardsmen to Portland. This was a measured response, sending only a minimal number of Guardsmen to restore order around the federal building. The limited nature of the federalization— for just 60 days—bolsters that conclusion. Given the proportionate response, the President likely invoked § 12406 lawfully.

FILED State of Oregon v. Trump, No. 25-6268 OCT 20 2025 MOLLY C. DWYER, CLERK GRABER, Circuit Judge, dissenting from the order: U.S. COURT OF APPEALS

In the weeks preceding the President’s September 27 social media post proclaiming that Portland was “War ravaged” and authorizing Secretary Hegseth to deploy federalized Oregon National Guard members, demonstrations in Portland were non-disruptive and small. Notwithstanding the turbulent events that had occurred several months earlier, the record contains no evidence whatsoever that, on September 27, Immigration and Customs Enforcement (“ICE”) was unable either to protect its Portland facility or to execute the immigration laws it is charged with enforcing. But, in the statute invoked here, Congress has authorized the President to call up the National Guard only to repel a foreign invasion, quell a rebellion, or overcome an inability to execute the laws. Consequently, no legal or factual justification supported the order to federalize and deploy the Oregon National Guard. Given Portland protesters’ well-known penchant for wearing chicken suits, inflatable frog costumes, or nothing at all when expressing their disagreement with the methods employed by ICE, observers may be tempted to view the majority’s ruling, which accepts the government’s characterization of Portland as a war zone, as merely absurd. But today’s decision is not merely absurd. It erodes core constitutional principles, including sovereign States’ control *2 over their States’ militias and the people’s First Amendment rights to assemble and to object to the government’s policies and actions. I strenuously dissent.

DISCUSSION

The government, as the party moving for a stay pending appeal, bears the burden of justifying a stay. Nken v. Holder, 556 U.S. 418, 433–34 (2009). We consider four factors: “(1) whether the stay applicant has made a strong showing that he is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Id. at 434 (citation omitted). The first two factors are the “most critical.” Id.

For the reasons described below, none of the factors supports the granting of a stay. The government has failed to carry its burden to justify a stay pending appeal.

I. Likelihood of Success: Plaintiffs are Likely to Succeed on Appeal.

The district court granted a temporary restraining order, a decision that we review for abuse of discretion. Woratzeck v. Ariz. Bd. of Exec. Clemency, 117 F.3d 400, 402 (9th Cir. 1997) (per curiam). We must defer to the district court’s factual findings unless they are clearly erroneous. Mondaca-Vega v. Lynch, 808 F.3d 413, 426 (9th Cir. 2015) (en banc).

The district court concluded that Plaintiffs met their burden of demonstrating that they were likely to succeed on the merits of their statutory claim that the deployment violated 10 U.S.C. § 12406; that they were likely to suffer irreparable harm without preliminary relief; and that the balance of equities and the public interest weighed in favor of a restraining order. Oregon v. Trump, No. 3:25-cv-1756-IM, 2025 WL 2817646, at *2–15 (D. Or. Oct. 4, 2025) (applying the factors described in Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008)).1 The first factor—likelihood of success—is “the most important factor,” Garcia v. County of Alameda, 150 F.4th 1224, 1230 (9th Cir. 2025) (citation and internal quotation marks omitted), and it forms the focus of the parties’ briefing to us. Because the district court did not abuse its discretion in concluding that Plaintiffs are likely to succeed on the merits, the government has no likelihood of success on appeal.

Title 10 U.S.C. § 12406 provides:

Whenever–

(1) the United States, or any of the Commonwealths or possessions, is invaded or is in danger of invasion by a foreign nation;

1 The district court reached the same conclusions with respect to Plaintiffs’ Tenth Amendment claim on the ground that the government lacked statutory, and thus constitutional, authority to federalize the National Guard in the circumstances. Oregon, 2025 WL 2817646, at *13–14. I agree with the parties that, for the purpose of the present motion, no meaningful difference exists between the statutory claim and the constitutional claim.

(2) there is a rebellion or danger of a rebellion against the authority of the Government of the United States; or

(3) the President is unable with the regular forces to execute the laws of the United States;

the President may call into Federal service members and units of the National Guard of any State in such numbers as he considers necessary to repel the invasion, suppress the rebellion, or execute those laws. Orders for these purposes shall be issued through the governors of the States or, in the case of the District of Columbia, through the commanding general of the National Guard of the District of Columbia.

In Newsom v. Trump, 141 F.4th 1032 (9th Cir. 2025) (order) (per curiam), we interpreted the statute in some detail. Violent protests had broken out in Los Angeles on June 6, 2025. Id. at 1041. The next day, June 7, protesters escalated their dangerous behavior—blocking traffic, damaging property, physically harming federal employees, and threatening many more. Id. “In response to these incidents, the President signed a memorandum on June 7, 2025,” calling National Guard members into service pursuant to § 12406. Id. Violent, disruptive, and dangerous protests continued for the next several days. Id. at 1041–42. The Governor of California and others sued the President, the district court granted a temporary restraining order, and the government appealed and sought a stay pending appeal. Id. at 1042–43.

We first rejected the government’s broad argument that the statute “completely precludes judicial review of the President’s determination that a *5 statutory precondition exists.” Id. at 1047. We held that, although “our review of the President’s determinations in this context is especially deferential,” our deference is not unlimited. Id. Instead, “courts may at least review the President’s determination to ensure that it reflects a colorable assessment of the facts and law within a ‘range of honest judgment.’” Id. at 1051 (quoting Sterling v. Constantin, 287 U.S. 378, 399 (1932)).

Here, as in Newsom, the government argues that the President properly invoked the statute under subsection two, pertaining to rebellions, and subsection three, pertaining to an inability to execute the laws. The majority’s order addresses subsection three only, so I start there.

A. Subsection Three: The President Was Not “Unable With the Regular Forces to Execute the Laws.”

In Newsom, we discussed the proper interpretation of the third category of emergency.2 We rejected, as too narrow, an interpretation that the statute requires “total or near total interference” with the execution of the laws. Newsom, 141 F.4th at 1051. But we also rejected, as too broad, the government’s assertion that “any minimal interference with the execution of the laws is, by itself, enough to

2 As the parties agree, we are bound to follow Newsom. I note that a district court in Illinois recently interpreted subsection three of the statute to authorize the President to call up the National Guard only in much narrower circumstances than Newsom allows. Illinois v. Trump, No. 25-cv-12174, 2025 WL 2886645, at *9– 10, *13–21 (N.D. Ill. October 10, 2025). The government’s arguments here would fail even more spectacularly under the interpretation in Illinois.

justify invoking § 12406(3).” Id. “Subsections one and two of the statute discuss unusual and extreme exigencies—invasions and rebellions—that threaten the normal operations of civil government. If we were to adopt the federal government’s reading of subsection three, it would swallow subsections one and two, because any invasion or rebellion renders the President unable to exercise some federal laws.” Id.

Applying that legal analysis to the facts, we concluded that the government was likely to prevail on appeal with respect to the President’s determination that he was unable with the regular forces to execute the laws:

Under a highly deferential standard of review, Defendants have presented facts to allow us to conclude that the President had a colorable basis for invoking § 12406(3). They presented evidence, detailed [earlier in the opinion], of protesters’ interference with the ability of federal officers to execute the laws, leading up to the President’s federalization of the National Guard on June 7. There is evidence that the day before, protesters threw objects at ICE vehicles trying to complete a law enforcement operation, “pinned down” several FPS officers defending federal property by throwing “concrete chunks, bottles of liquid, and other objects,” and used “large rolling commercial dumpsters as a battering ram” in an attempt to breach the parking garage of a federal building. Plaintiffs’ own submissions state that some protesters threw objects, including Molotov cocktails, and vandalized property. According to the declarations submitted by Defendants, those activities significantly impeded the ability of federal officers to execute the laws. Affording appropriate deference to the President’s determination, we conclude that he likely acted within his authority in federalizing the National Guard under 10 U.S.C. § 12406(3).

Id. at 1052 (paragraph break omitted).

The majority’s order here holds that, as in Newsom, the government is likely to prevail on appeal with respect to the statute’s third subsection. The majority’s decision cannot be squared with the record in this case and cannot be reconciled with Newsom.

The district court’s factual findings, which the government does not challenge on appeal, and which are not clearly erroneous in any event, fully resolve this issue. In the two weeks leading up to the President’s September 27 social media post, there had not been a single incident of protesters’ disrupting the execution of the laws. Oregon, 2025 WL 2817646, at *10. I repeat: not a single incident for two weeks. Here are summaries from police reports for the five days preceding the President’s social media post.

September 22: Approx. 7-10 people. No calls. September 23: Few people. No activity. September 24: Approx. 10 people. No calls. September 25: Approx. 20 people. No calls. September 26: Approx. 15 people. Energy low, minimal activity. No incidents.

A police officer’s report from September 26 stated: “Throughout the day we observed approximately 8-15 people at any given time out front of ICE. Mostly sitting in lawn chairs and walking around.”

It is hard to understand how a tiny protest causing no disruptions could possibly satisfy the standard that the President is unable to execute the laws. The facts at issue in Newsom—significant, violent protests by hundreds of people in *8 several locations the day before and the day of the President’s invocation of the statute—could not be further from the facts here—small gatherings in one location with “no activity” or “minimal activity,” low energy, and no calls for assistance for weeks. See 141 F.4th at 1041; see also Illinois v. Trump, No. 25-2798, 2025 WL 2937065 (7th Cir. Oct. 16, 2025) (per curiam) (reaching the opposite conclusion from the majority’s order when assessing the same legal issue with respect to larger, more violent protests in the days before the President’s invocation of the statute in Illinois).

The government advances two arguments to the contrary. Neither is persuasive.

1. Timing

First, the government asks us (and the majority’s order agrees) to expand the time frame that we consider when asking whether the President is unable to execute the laws of the United States. In the government’s view, even if there is complete calm for weeks beforehand, we should look further back in time to assess whether a present exigency exists. The government is wrong as a matter of both law and fact.

a. Legal Error

As a legal matter, the statute is clear that the President must be unable at the present moment to execute the laws. See Oregon, 2025 WL 2817646, at *9 *9 (“Under Newsom, and as a matter of logic, this Court assesses whether the President invoked Section 12406(3) based on ‘a colorable assessment of the facts’ as they existed at the time he federalized the National Guard.” (citation name altered) (quoting Newsom, 141 F.4th at 1051)). The statute uses the present tense: “Whenever . . . (3) the President is unable with the regular forces to execute the laws.” 10 U.S.C. § 12406(3) (emphasis added). That use of the present tense clearly conveys congressional intent that the present state of affairs is the relevant time. See United States v. Wilson, 503 U.S. 329, 333 (1992) (“Congress’ use of a verb tense is significant in construing statutes.”); United States v. Jackson, 480 F.3d 1014, 1019 (9th Cir. 2007) (holding that Congress’ use of “the present tense verb ‘travels,’ most sensibly read, does not refer to travel that occurred in the past”); Bonnichsen v. United States, 367 F.3d 864, 875 (9th Cir. 2004) (“Congress’s use of the present tense is significant. . . . Congress, by using the phrase ‘is indigenous’ in the present tense, referred to presently existing tribes, peoples, or cultures.”). Nor would it make sense for this narrow statute, which focuses on extreme and unusual occurrences, to allow the President to call up the National Guard to address an exigency that no longer exists.

Moreover, the text of subsection three contrasts with the text of the other two subsections in a way that removes any doubt about the relevant time frame. Cf. Newsom, 141 F.4th at 1051 (interpreting the text of subsection three by *10 considering the scope of the first two subsections). Subsections one and two permit deployment whenever there is an invasion or rebellion or a danger of those events. 10 U.S.C. § 12406(1)–(2). That is, the first two subsections call for a predictive judgment about the possibility of an imminent but future invasion or rebellion. Subsection three, by contrast, asks only whether the President is presently unable to execute the laws; the subsection does not ask in the alternative whether the President is in danger of being unable to execute the laws. In sum, subsection three exclusively concerns the present, not the past and not the future. See City & County of San Francisco v. Env’t Prot. Agency, 604 U.S. 334, 344 (2025) (“‘Where Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.’ . . . We have invoked this canon time and again.” (brackets omitted) (quoting Russello v. United States, 464 U.S. 16, 23 (1983))).

That understanding fully comports with both precedent and historical practice. In Newsom, for example, large, violent, and dangerous protests erupted on June 6 and continued for several days. 141 F.4th at 1041–42. The President called up the National Guard on the second day of protests, June 7—in the midst of protests that, at least colorably, were significantly impeding the execution of the immigration laws. Id. at 1041. And when we assessed whether the President was *11 unable to execute the laws, we examined the events immediately preceding the invocation. Id. at 1052. When the Rodney King riots broke out on April 29, 1992, and continued for many days, President Bush called up the National Guard two days after the riots started, on May 1—in the midst of the worst rioting. Exec. Order No. 12,804, 57 Fed. Reg. 19361 (May 1, 1992); Proclamation No. 6427, 57 Fed. Reg. 19359 (May 1, 1992). When a Postal Service strike began in New York City on March 18, 1970, and spread during the next eight days, President Nixon called up the National Guard to process and deliver the mail five days after the initial strike, on March 23—in the midst of a severe mail backlog that was akin to a modern-day shutdown of the internet. Exec. Order No. 11,519, 35 Fed. Reg. 5003 (March 23, 1970); Proclamation No. 3972, 35 Fed. Reg. 5001 (March 23, 1970). When riots occurred after Martin Luther King, Jr., was assassinated on April 4, 1968, President Johnson called up the National Guard three days later, on April 7—in the midst of terrible rioting. Exec. Order Nos. 11,404–05, 33 Fed. Reg. 5503–05 (April 7, 1968); Proclamation Nos. 3841–42, 33 Fed. Reg. 5497–99 (April 7, 1968). Although the tight deadline to deliver this dissent prevents a full historical analysis, it suffices to note that the government does not suggest that, before this case, a President ever has invoked § 12406 when there was not an extreme and ongoing exigency.3

3 The concurrence asserts that, because violence attributed to (continued)

b. Factual Error

Even if we assume that events occurring weeks or months before the invocation of the statute are relevant considerations, the facts of this case come nowhere near justifying the conclusion that the President was unable on September 27 to execute the laws. Let’s expand the inquiry back to the beginning of September. The district court accurately described and assessed the events in September:

[The government’s] declarants describe only four incidents of protesters clashing with federal officers in the month of September preceding the federalization order—on September 1st, 9th, 12th, and without further specification, the second week of September. The first involved protesters setting up a makeshift guillotine to intimidate federal officials; the second involved four people shining overpowered flashlights in the eyes of drivers; the third involved someone posting a photograph of an unmarked ICE vehicle online; and the last involved additional drivers having flashlights shone in their eyes. These incidents are inexcusable, but they are nowhere near the type of incidents that cannot be handled by regular law enforcement forces. . . . More broadly, these incidents are categorically different from the violent incidents as described in

the Whiskey Rebellion began as early as 1791, yet President Washington invoked the statutory predecessor to § 12406 only in 1794, the Whiskey Rebellion undermines my argument about the requirement of immediacy. Concurrence at 17 n.3. The concurrence misunderstands both the relevant inquiry and the historical facts. As explained in text, the statute makes clear that, at the time when the President invokes the statute, the situation on the ground must be an emergency. Nothing requires the President to invoke the statute as soon as some violence occurs. Turning to the facts, the Whiskey Rebellion lends further historical support to my reading of the statute as requiring an emergency at the time of invocation. President Washington invoked the statute, thus calling up the militia, shortly after the “most violent” events of the rebellion, that is, in the midst of an emergency. Bennett Milton Rich, The Presidents and Civil Disorder 7–8 (1941).

Newsom that took place in Los Angeles when the President federalized the California National Guard on June 7, 2025.

Oregon, 2025 WL 2817646, at *10 (citation name altered) (citations omitted).

Police reports during the month of September reflect the tenor of the protests for days other than the four mentioned by the government. For example, here is a report from September 5:

At [2:00 p.m.] today, I drove by ICE and observed 1 person standing across the street by the entryway. I touched base with Will Turner from FPS and was told there was a small group with low energy.

Drove by around [9:00 p.m.], saw 8 people out front and couldn’t even get one of them to flip me the bird. Very low energy. It appeared two trucks made entry to the building with no issues.

(Typos and missing apostrophe corrected). Here is a summary for September 7: “Approx. 7 people. Low to no energy. No issues. No calls.”

If we expand our inquiry further still, to encompass all of August 2025, here is what we learn: “On August 9, 2025, protesters attempted to block an ICE vehicle and federal officers had to use pepper spray so the vehicle could depart.” Id. at *5. That’s it.

Otherwise, from July 18th to September 27th, Central Precinct police officers handled the ICE protests with routine monitoring. The protests generally were limited to fewer than 30 people and were “largely sedate.” If the protests were to increase or threaten public safety, [the Portland Police Bureau] could call on additional available resources. But the protests have been such a minor issue, that the normal nightlife in downtown Portland has required more police resources than the ICE facility.

Id. (citations omitted). At no point during these two-plus months is there even a colorable assertion that the President was unable to execute the laws without more than truly “minimal interference,” which as a matter of law does not justify invoking subsection three of the statute. Newsom, 141 F.4th at 1051.

The government insists that we expand our inquiry further still, to encompass the four months leading up to September 27, including the more serious protests in June and the resulting closure of the ICE building for more than three weeks, from June 13 to July 7.4 As an initial matter, those events, even at that time, likely did not result in an inability to execute the laws. The government has not asserted, for example, that ICE employees were unable to work, that immigration arrests could not be made, or that ICE officers could not otherwise

4 The government also asserts, and the majority’s order agrees, that the Portland Police Bureau (“PPB”) was unwilling to respond to calls for assistance. The record defies that characterization. To the contrary, the PPB activated its Rapid Response Team during June; began routine monitoring of the situation at the ICE building; and filed daily reports on events at the facility. The government cannot point to a single instance in which federal property or personnel was threatened or in need of assistance and the local police failed to respond to a request for assistance. Instead, for the entire four-month period, the government points to: (a) a citizen’s comment during a 911 call that the police were “useless”; (b) an instance when no call for assistance was made but an officer stated that, because high-priority crimes elsewhere in the city occupied the officers’ primary attention, had a hypothetical call come for a non-urgent matter, the local police would not have had the capacity to respond immediately; and (c) a single failure by the PPB to respond quickly to requests to assist a distressed person passing near the ICE building. The record otherwise shows full assistance, including arrests, by the Portland police.

perform their duties. The record shows the opposite; the government submitted a declaration explaining that “most Portland ICE officers” simply “work[ed] out of an alternative temporary space.” To be sure, the government also asserted that some immigration appointments “had to be rescheduled for alternative locations and/or times,” but the rescheduling of routine appointments cannot justify calling up the militia. See Newsom, 141 F.4th at 1051 (holding that “minimal interference” with the execution of the laws does not suffice). Additionally, unlike the incidents at multiple locations in Los Angeles, at issue in Newsom, the situation in Portland concerned only a single federal building that occupies a single city block.

In any event, even assuming that an emergency within the meaning of § 12406 existed in June, those events do not even colorably show an inability to execute the laws at the end of September. To the contrary, civilian authorities were able to restore order and resume regular operations. It is illogical to assert that, because an emergency existed three months ago, the same emergency exists today. A pot of tepid water is not a pot of boiling water, and it cannot hurt you, even if it was boiling three hours earlier.

The majority’s order not only combines scattered incidents across four months when concluding that an exigency existed on September 27, the order also stretches the facts to suggest a level of continuing violence that the record does not *16 reflect. For example, the order begins its factual recitation by describing a shooting at an ICE facility in Dallas that resulted in the death of a detainee, and the apprehension in Chicago of a protester who had a gun. Order at 4–5. By contrast, in Portland, no one has been killed or seriously injured, and nothing else of similar gravity has occurred. Moreover, no colorable assessment of the situation in Portland can rest on what happened in far-away locations with no connection to protesters in Portland. Were it otherwise, the President could call up the National Guard in any location, anywhere in the country, without making an honest assessment of local conditions.

In sum, the statute requires that we ask whether the President is unable to execute the laws at the time when he invokes the statute. Whether we consider only the days and weeks leading up to September 27 or the full four months beforehand, no colorable assessment of the facts and law can justify a conclusion that an emergency existed on September 27.

2. Capacity of Civilian Law Enforcement

The government’s second argument is that it was able to stabilize the situation in Portland only by using Federal Protective Service (“FPS”) officers from elsewhere in the country and that, because of the situation in Portland, federal law enforcement forces were “overstretched.”

The legal basis for this argument is unclear. The trigger for federalizing the National Guard is an inability to execute the laws, not staffing difficulties that fall short of demonstrably resulting in an inability to execute the laws. The government has not explained how its purported staffing troubles were causing an inability to execute the laws on September 27. As explained above, the protests themselves—small, uneventful, low-energy—were not preventing execution of the laws at that time. The most that can be said is that, because FPS officers were stretched thin, if protests were to flare up in the future and if staffing woes were to lead to insufficient staffing, then an inability to execute the laws might arise at some hypothetical future time. But, as also explained above, subsection three of the statute requires a present-day inability to execute the laws; fear of a future inability is not enough. Nor could staffing difficulties alone support an inability to execute the laws; otherwise, the President could direct scores of FPS officers to a location with minimal security issues and then claim authority to call up the National Guard because those officers are needed elsewhere. In assessing whether the President had a colorable basis for concluding that the statutory requirements were met, we must consider the actual situation being addressed by the FPS officers.

Even assuming that staffing difficulties—without any connection to the facts on the ground—could constitute an inability to execute the laws, the government’s *18 argument fails for a more fundamental reason: lack of factual support. Nothing in the record reasonably supports a conclusion that, in late September, deployment of FPS officers to Portland was causing significant staffing problems.

The majority’s order spells out an argument that the government does not make, presumably because the government recognizes the lack of factual support. The argument in the majority’s order proceeds as follows. FPS has 776 officers, but only 497 officers are trained to protect federal buildings. Robert Cantu, the regional FPS director, asserted that, from June through September, “115 FPS officers have had to deploy to Portland.” The majority’s order first assumes that all 115 officers—nearly a quarter of the agency’s officers with relevant training— were stationed in Portland in late September. The majority’s order next reasons that such a diversion supports an inference that Portland is a significant source of staffing woes.

But that argument impermissibly adds facts to Director Cantu’s vague, carefully worded assertion. Crediting his assertion, we know that a total of 115 officers from elsewhere were deployed in Portland during the preceding four months. The record contains no information about how many officers were in Portland at any given time. For all we know, FPS sent a different 8 officers to Portland every week for 14 or 15 weeks, meaning that Portland’s drain on FPS’s staff from elsewhere on any particular day was 8 people, not 115. Indeed, the only *19 description in the record of a “[s]urge” in officers was the deployment of 8 officers. The fact that there were 26 FPS officers on duty on September 6, as the majority’s order emphasizes, Order at 27 n.13, says nothing about whether any or all of those individuals were from somewhere other than Portland. The record does not reveal the number of local FPS officers.

Even if we assume that FPS deployed all 115 officers in June, it strains credulity to assume that all 115 of them remained in Portland for four months. What were they doing during the month of August, for example, when there was only a single incident at the ICE facility during the entire month? The record does not tell us. Indeed, the record does not shine light on the most pertinent information: in the days leading up to September 27, how many FPS officers from elsewhere were in Portland? The only hint in the record is a reference to some officers from elsewhere leaving Portland and returning to their home stations.

This factual deficit in the record weighs doubly against the government. First, in the present procedural context, the government bears the burden of persuading us that a stay is warranted. See Nken, 556 U.S. at 433–34 (“The party requesting a stay bears the burden of showing that the circumstances justify an exercise of that discretion.”). The government cannot rely on factual ambiguities of its own making to meet its burden.

Second, and perhaps more revealing, the government is the only party with access to the pertinent information. The government presumably has a record of who was deployed to Portland and when. We do not know why the government chose not to file a declaration describing the relevant details. But it cannot submit an ambiguous assertion and then ask us to interpret that assertion in its favor. See Doe #1 v. Trump, 957 F.3d 1050, 1059–60 (9th Cir. 2020) (“The government cannot meet [its] burden [to demonstrate irreparable harm] by submitting conclusory factual assertions and speculative arguments that are unsupported in the record.”); see also Clifton v. United States, 45 U.S. 242, 248 (1846) (“[I]f the weaker and less satisfactory evidence is given and relied on in support of a fact, when it is apparent . . . that proof of a more direct and explicit character was within the power of the party, . . . it may well be presumed, if the more perfect exposition had been given[,] it would have laid open deficiencies and objections which the more obscure and uncertain testimony was intended to conceal.”).5

5 All of the reasoning in text applies equally to another carefully worded statement in the Cantu declaration: “While FPS officers are expected to respond to emergencies after hours and on weekends and holidays, they are not expected to work 12-hour shifts, seven days a week, which has become the norm.” That statement skillfully avoids describing how many officers worked those hours, whether they worked those hours while deployed in Portland, how often they worked such strenuous hours, and so forth. The majority’s order combines its maximal interpretation of 115 officers with a maximal interpretation of this statement, too, leading to the questionable conclusion that Portland requires “nearly 25% of the agency’s capacity, working 12-hour shifts, seven days a week.” (continued)

Whatever staffing problems the government faced on September 27, the present record provides no support for the proposition that a staffing challenge amounted to an inability to execute the law.

B. Subsection Two: There Was No “Rebellion or Danger of a Rebellion Against the Authority of the Government of the United States.”

The government argues, in the alternative, that it is likely to prevail with respect to the second subsection, which authorizes the federalization of the National Guard if “there is a rebellion or danger of a rebellion against the authority of the Government of the United States.” 10 U.S.C. § 12406(2). We did not analyze that subsection in detail in Newsom, but we clearly understood that the term “rebellion” cannot be construed so broadly that it encompasses run-of-themill protests. To the contrary, we described a “rebellion” as an “unusual and extreme exigenc[y].” Newsom, 141 F.4th at 1051.

The district court here, like district courts in other cases, adopted a narrow construction of the word “rebellion.” See Oregon, 2025 WL 2817646, at *12–13 (determining that the “key characteristics” of a “rebellion” are that it is (1) violent and armed; (2) organized; (3) open and avowed; and (4) “against the government as a whole . . . rather than in opposition to a single law or issue”); accord Illinois,

Order at 28. Because the government has the burden, because the government is the only party with access to clearer information, because the majority’s interpretation is implausible, and because sheer diversion without any connection to the facts on the ground is insufficient, the government cannot prevail with only carefully crafted, vague statements.

2025 WL 2886645, at *15; Newsom v. Trump, 786 F. Supp. 3d 1235, 1251–53 (N.D. Cal. 2025). At oral argument, the government’s lawyer suggested that we apply a broader definition, from the 1891 edition of Black’s Law Dictionary: “Deliberate, organized resistance, by force and arms, to the laws or operations of the government.” Oral Argument at 3:27–3:53, https://www.ca9.uscourts.gov/ media/video/?20251009/25-6268/; Rebellion, Black’s Law Dictionary (1st ed. 1891). Determining the precise definition of “rebellion” under § 12406(2) is unnecessary in this case because, even under the government’s proposed definition,6 properly construed in light of the statutory context, the activity in Portland came nowhere close to constituting a rebellion.

As an initial matter, the record contains no evidence that the sporadic violent events occurring over a handful of days during four months of otherwise peaceful protests were in any way organized. For example, there is no evidence of leadership, organizational structure, premeditation, or an overarching plan.

Even putting aside that deficiency, nothing in the record suggests that, on September 27, there was a rebellion or a danger of one. The same reasons given in

6 The concurrence errs when it states that, by declining to address this complicated point, I am “tacitly admitting” that the district court’s definition was wrong. Concurrence at 18 n.5. Avoiding a definitive decision on an unnecessary issue while giving the losing party the benefit of the doubt is a frequent feature of judicial opinions, and it cannot reasonably be construed as an admission, tacit or otherwise.

Part I-A-1, above, apply here. In the two weeks leading up to September 27, there was not a single incident of “force and arms” against ICE’s personnel or facility. And going back more than two months, the record contains only “evidence of sporadic violence against federal officers and property damage to a federal building.” Oregon, 2025 WL 2817646, at *13.

Even considering all four months, the events as a whole did not rise to the level of an “unusual and extreme exigenc[y]” constituting a “rebellion.” Newsom, 141 F.4th at 1051. On almost every day during the four months preceding September 27, the record discloses ordinary political protests in Portland. Ordinary protests—quintessential First Amendment activity—obviously do not constitute a rebellion. See Illinois, 2025 WL 2937065, at *6 (“Nor does a protest become a rebellion merely because of sporadic and isolated incidents of unlawful activity or even violence committed by rogue participants in the protest. Such conduct exceeds the scope of the First Amendment, of course, and law enforcement has apprehended the perpetrators accordingly. But because rebellions at least use deliberate, organized violence to resist governmental authority, the problematic incidents in this record clearly fall within the considerable daylight between protected speech and rebellion.”); cf. Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 508 (1969) (“[I]n our system, undifferentiated fear or apprehension of disturbance is not enough to overcome the right to freedom of *24 expression.”); Milk Wagon Drivers Union v. Meadowmoor Dairies, Inc., 312 U.S. 287, 293 (1941) (“[T]he right of free speech cannot be denied by drawing from a trivial rough incident or a moment of animal exuberance the conclusion that otherwise peaceful picketing has the taint of force.”); Index Newspapers LLC v. U.S. Marshals Serv., 977 F.3d 817, 834 (9th Cir. 2020) (“The many peaceful protesters, journalists, and members of the general public cannot be punished for the violent acts of others.”). Once the ordinary, peaceful protest activities are stripped away, the record reflects only sporadic, uncoordinated criminal acts that were handled promptly and appropriately by civilian law enforcement.⁷ Those incidents do not amount to a “rebellion” under any reasonable definition of the term. Nor is there any suggestion in the record of a “danger” of a rebellion.

Contrary to the concurrence’s suggestion, historical examples affirmatively disprove that the events in Portland are similar to rebellions of the Founding era. The Whiskey Rebellion, Shays’s Rebellion, and Fries’s Rebellion,⁸ for example,

7 Any act of violence against federal officers is a grave event, and I do not diminish the serious nature of the criminal acts that occurred. But the acts of violence catalogued in the record—heavily emphasized by the majority’s order and the concurrence—do not reflect the day-to-day reality of the protests, particularly in the weeks leading up to September 27. On most days, the most notable aspect of the demonstrations here was the presence of people dressed in chicken, frog, or birthday suits. 8 It is not clear that Fries’s “Rebellion” actually was understood as a rebellion at the time. President Adams invoked the predecessor statute’s clause (continued)

each involved hundreds or thousands of people—a sizeable percentage of the population at the time—openly attacking federal officials and properties in clearly organized fashion across a wide geographic area. See generally David P. Szatmary, Shays’ Rebellion: The Making Of An Agrarian Insurrection 57–59 (1980); Bennett Milton Rich, The Presidents And Civil Disorder 3–7, 21–24 (1941). Those rebellions shared several salient characteristics, including a large number of participants relative to the population and to available law enforcement, a wide geographic scope, evident organization and leadership, widespread use of arms, intense ferocity, and the creation of extreme difficulty restoring control by means of ordinary law enforcement.

What occurred in Portland differed in every dimension. As already noted, there is no evidence of organization or leadership, widespread use of arms,

pertaining to the execution of the laws, not the clause pertaining to an insurrection. John Adams, Proclamation 9—Law and Order in the Counties of Northampton, Montgomery, and Bucks, in the State of Pennsylvania Online by Gerhard Peters and John T. Woolley, The American Presidency Project, https://www.presidency.ucsb.edu/node/202751. And many historians, including the one cited by the concurrence, have suggested that Fries’s “Rebellion” is a “misnomer.” The Presidents And Civil Disorder 21 n.1; see Frederic D. O’Brien, The Fries Rebellion, American Heritage (Apr. 1999), https://www.americanheritage.com/1799-two-hundred-years-ago [https://perma.cc/H8EF-CN4S] (cited by the concurrence) (describing Fries’s Rebellion as “a mere riot”). For the purpose of this dissent, I assume that it was a rebellion rather than a riot.

ferocity, or difficulty exerting control by ordinary means. In addition, the population of the Portland metropolitan area exceeds two and a half million, spread across nearly 6,700 square miles. U.S. Census Bureau, Annual Estimates of the Resident Population for Metropolitan Statistical Areas in the United States and Puerto Rico: April 1, 2020 to July 1, 2024; U.S. Census Bureau, State and Metropolitan Area Data Book: 2010, at 110. At their height, the protests in Portland have involved 200 people, or about 0.008% of the population. And they have taken place exclusively around a single city block, or approximately 0.00002% of the Portland metro area.⁹ The few people who did commit sporadic acts of violence have been arrested, processed, and charged by regular law enforcement forces.

In sum, occasional illegal activity by such a small number of people in such a limited area is not at all akin to the Founding-era rebellions. Returning to the

9 The concurrence takes the position that “[t]he number [of] rebels . . . has little bearing on whether a rebellion exists” within the meaning of § 12406, claiming specifically that Fries’s Rebellion contradicts consideration of that factor. Concurrence at 18 n.4. But Fries’s revolt involved more than a hundred and forty armed resisters, under the command of John Fries, organized into two companies of riflemen and one company of horsemen, who marched on a prison housing eighteen of their comrades. W.W.H. Davis, The Fries Rebellion 63 (1899). The group’s size intimidated the much smaller force guarding the prison—twenty men in all, armed with pistols—and the federal forces eventually relinquished the prisoners. Id. at 63–66. I am certain that Pennsylvania farm country at the time was not home to more than 2.5 million people, and that the presence of a large number of armed attackers, in comparison to the population, played a role in the perception of the gravity of the event.

statutory text, the historical events more closely resembled an “invasion by a foreign nation,” 10 U.S.C. § 12406(1), albeit from within, and they bore no resemblance to the recent activities in Portland.

Because the government has no likelihood of success on appeal with respect to either subsection, this factor weighs heavily against granting a stay.

II. Injury to the Government: The Government Will Not Be Injured Absent A Stay.

The second Nken factor is whether the government “will be irreparably injured absent a stay.” 556 U.S. at 434 (citation omitted). For three independent reasons, the government fails to meet this standard.

First, the foregoing analysis of likelihood of success all but resolves this issue. Because the President lacked statutory authority to federalize the National Guard, the government cannot show any injury.

Second, the government fails to meet its burden of showing that continuing protests are causing irreparable injury. See Newsom, 141 F.4th at 1054 (considering the effect of ongoing protests when assessing the harm to the government). As described above, the protests were tranquil during the days and weeks before September 27. The President’s decision to call up the National Guard caused more activity for two—and only two—nights. On September 27, approximately 60 people were present; FPS officers arrested one person for blocking the driveway early on; and, by 11:00 p.m., the crowd was described as *28 “mild,” with “low to no activity,” and no other arrests were made. On September 28, a crowd of 200 rallied at the ICE building,¹⁰ but only around 50 remained later on. Police notes record that a “[v]ehicle attempted to drive into crowd but was surrounded.” Two people were arrested for assaults among the protesters, but “[n]o criminal activity [was] directed towards ICE facility o[r] Feds.” There were a “[c]ouple of confrontations as Feds came out to facilitate vehicles [that were coming] to or from [the] facility.” The criminal acts are inexcusable, but they do not present a risk of irreparable injury to the government.

Perhaps more importantly, the irreparable-injury question asks whether the applicant “will be irreparably injured absent a stay,” Nken, 556 U.S. at 434 (emphases added), thus requiring a predictive judgment about the state of affairs in the coming days if the militia is not deployed, see, e.g., Doe #1, 957 F.3d at 1059 (“[A]t this juncture, the government has the burden of showing that irreparable injury is likely to occur during the period before the appeal is decided.”). The most recent events confirm that the government will not be injured if a stay is denied. After those first two nights of increased activity resulting from the President’s actions, the protests have returned essentially to their earlier calm state. Police notes for September 29 record: “40 people. No issues. No energy.” FPS officers

10 There is no evidence in the record about how many of these persons were protesters, as opposed to bystanders, journalists, live-streamers, and so on. In any event, the record suggests that no more than 50 people remained at the ICE facility for any meaningful duration.

facilitated vehicles arriving and leaving, and they had to remove one person who would not leave the driveway. On September 30, police notes state: “Approx. 50 people. Uneventful. No issues. No [use of force] or arrests.” The government filed its motion for a stay on October 5, and the government did not assert that there had been any trouble at all in the intervening days. On the present record, it is hard to see any injury, and certainly no irreparable injury, that will result if we deny a stay. Put simply, FPS officers do not have an urgent need for military support to deal with uneventful small gatherings.

The majority’s order nevertheless concludes that, without a stay, the government would be irreparably harmed because the National Guard is needed to protect the ICE building. Order at 33–34. The evidence that it cites in support is taken almost entirely from four months ago. No evidence in the record suggests that the events in June are apt to recur, and a proper assessment of the state of the protests both before and after the President’s invocation of the statute shows no threat of irreparable injury.

The third reason why the government cannot show irreparable injury from the district court’s first temporary restraining order (“TRO”) is that a second source—the district court’s second TRO—prevents deployment of the National Guard. The government has not appealed the second TRO, so a stay pending *30 appeal will not have any immediate practical effect. The government will remain barred from deploying the National Guard.

The government’s lawyer asserted during oral argument that the government chose not to file an immediate appeal from the district court’s second order on the theory that, if this panel grants a stay, he “would expect the district court to immediately dissolve th[e] second TRO.” Oral Argument at 24:30–24:35, https://www.ca9.uscourts.gov/media/video/?20251009/25-6268/. That expectation, with which the majority’s order agrees, is unwarranted. The majority’s order only grants a stay, which suspends the effect of the first TRO; it does not dissolve the first TRO or the reasoned decision on which it rests, and the stay certainly does not require dissolution of the second TRO. See Nken, 556 U.S. at 428–30 (discussing the effect of a stay pending appeal). Additionally, the majority’s order concludes only that the government is likely to succeed on appeal; final resolution of the merits of the government’s appeal is not before us. See id. at 432 (“The whole idea is to hold the matter under review in abeyance because the appellate court lacks sufficient time to decide the merits.”). Moreover, although the district court granted the second TRO in part for the same reasons that it had granted the first TRO, the motion for a second TRO raised additional issues and involved an additional party, and the district court granted the second TRO also in part because of “the newly submitted declarations.” Oregon v. Trump, No. 3:25-*31 cv-1756-IM, 2025 WL 2823653, at *1 (D. Or. Oct. 5, 2025) (order). If the district court entertains a motion to dissolve the second TRO,¹¹ the court may decide to keep that order in place: either because the court remains convinced of its original reasoning despite the majority’s prediction that the government will prevail in this appeal, or because other reasons support the TRO, or both.¹² All of this is to say that the government’s decision thus far not to appeal the second TRO means that it has fallen short of showing irreparable injury as a result of the first TRO alone.

For each of those three independent reasons, the government has not shown irreparable injury.

III. Substantial Injury to Interested Parties: Plaintiffs And Others Will Suffer Substantial Harm.

The third Nken factor is “whether issuance of the stay will substantially injure the other parties interested in the proceeding.” 556 U.S. at 434. Although this factor is less critical than the first two, it also supports the denial of a stay.

11 Earlier this week, the district court set a tight schedule for the filing of any motion to dissolve the second TRO. The government has 48 hours to file such a motion, and Plaintiffs have 24 hours to respond. 12 For reasons that escape me, the majority’s order states that I “do[] not challenge” the conclusion that “the first TRO and the second TRO rise or fall together on the merits of the issues raised in this motion for a stay pending appeal.” Order at 34. Not so. As explained in text, the issues raised in this motion for a stay pending appeal include the government’s likelihood of success on its challenge to the district court’s grant of the first TRO. Nothing in the majority’s order does—or could—require the district court to dissolve the second TRO, which is not before us.

Permitting the illegal deployment of Oregon National Guard members is a direct affront to Oregon’s sovereignty. See Bond v. United States, 564 U.S. 211, 221 (2011) (“[The] allocation of powers between the National Government and the States . . . preserves the integrity, dignity, and residual sovereignty of the States.”); see also Illinois, 2025 WL 2937065, at *7 (holding that deploying National Guard troops in Illinois over the State’s objection would harm the State). More concretely, deployment would deprive the State of more than half its reserves who are trained to respond quickly to emergencies, including natural disasters. Oregon, 2025 WL 2817646, at *14.

Plaintiff City of Portland has a strong interest in preserving the peace. As the district court found, the deployment of troops in Portland is likely (if not certain) to aggravate the situation at the ICE building. Id. Finally, nearby businesses have economic interests that are likely to be harmed by the deployment of troops.

The third Nken factor thus supports the denial of a stay.

IV. The Public Interest: The Public Interest Supports Denial of a Stay.

The final factor—consideration of the public interest—merges with the second factor in cases, like this one, in which the government is the party seeking a *33 stay.¹³ Index Newspapers, 977 F.3d at 838. For the reasons stated in Part 2, this factor, too, strongly supports the denial of a stay.

CONCLUSION

The Founders recognized the inherent dangers of allowing the federal executive to wrest command of the State militia from the States. Congress authorized the President to deploy the National Guard only in true emergencies— to repel an invasion, to suppress a rebellion, or to overcome an inability to execute

13 The majority’s order legally, and logically, errs by asserting that the third and fourth factors merge whenever the government is a party. Order at 32–33. When the government is the party opposing a stay, the third and fourth factors merge. Nken, 556 U.S. at 435. But when the government is the party seeking a stay, the second and fourth factors merge. Index Newspapers, 977 F.3d at 838. The guiding principle is that the public interest inquiry merges with the inquiry into harm to the government. The majority’s order fundamentally misunderstands the statement in Nken that the last two factors merge when the government is the “opposing” party. What Nken meant is that those factors merge when the government is the party opposing a stay, not whenever the government is the defendant. See, e.g., Garcia, 150 F.4th at 1234 (“Where, as here, the party opposing injunctive relief is a government entity, the third and fourth factors—the balance of equities and the public interest—merge.” (emphasis added) (internal quotation marks omitted) (quoting Fellowship of Christian Athletes v. San Jose Unified Sch. Dist. Bd. of Educ., 82 F.4th 664, 695 (9th Cir. 2023) (en banc))). We could not have been clearer in Thakur v. Trump, 148 F.4th 1096 (9th Cir. 2025): “the third and fourth factors merge . . . when the government is the party opposing a stay, rather than the party seeking one.” Id. at 1109 n.7. The majority’s order provides no explanation for its failure to apply the law correctly; instead, it states only, in a footnote, that it would reach the same conclusion regardless. Order at 33 n.¹⁶. The confusion in the majority’s order on this point allows it to minimize its analysis of the second factor, which is one of the two most critical factors and which plainly supports the denial of a stay here. Once that factor is properly assessed, the government cannot show irreparable harm, and the public interest supports denial of a stay.

the laws. 10 U.S.C. § 12406. Congress did not authorize deployment in merely inconvenient circumstances, and Congress unquestionably did not authorize deployment for political purposes. Article III commands that we enforce those limits. The majority’s order abdicates our judicial responsibility, permitting the President to invoke emergency authority in a situation far divorced from an enumerated emergency.

Today’s President seeks to bring troops into one set of States to enforce one set of laws; a future President may seek to bring troops into a different set of States to enforce a different set of laws. Partisans who cheer this President’s use of troops to protect personnel who are enforcing federal immigration laws would do well to consider whether they would be equally pleased if a future President uses troops to protect personnel who are enforcing laws that they vehemently dislike. Cf. Greer v. Spock, 424 U.S. 828, 839 (1976) (noting “the American constitutional tradition of a politically neutral military establishment under civilian control”).

Except in true emergencies, and by design of the Founders and Congress, our civil society resolves its disputes without domestic military intervention. Cf. Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 650 (1952) (Jackson, J., concurring) (“[The Founders] knew what emergencies were, knew the pressures they engender for authoritative action, knew, too, how they afford a ready pretext for usurpation. We may also suspect that they suspected that emergency powers *35 would tend to kindle emergencies.”). The Founders’ deep concerns about the domestic use of the militia by the federal executive animated the drafting of the Constitution. See, e.g., Perpich v. Dep’t of Def., 496 U.S. 334, 340 (1990) (describing the Founders’ “widespread fear that a national standing Army posed an intolerable threat to individual liberty and to the sovereignty of the separate States”). Pertinent here, the Founders reserved to Congress the power to “provide for calling forth the Militia” and limited the circumstances in which the militia may be pressed into federal service. U.S. Const. art. I, § 8, cl. 15. Beginning in 1792, Congress has delegated some—but not all—of that power to the President. That delegation has evolved over time, but it always has come with specific limits on the President’s authority: Congress clearly contemplated that the President could call up the National Guard only during emergencies. See, e.g., Sterling, 287 U.S. at 399 (“The power [of the President to call up the militia] ‘is to be exercised upon sudden emergencies, upon great occasions of state, and under circumstances which may be vital to the existence of the Union.’” (quoting Martin v. Mott, 25 U.S. 19, 30 (1827))).

We have come to expect a dose of political theater in the political branches, drama designed to rally the base or to rile or intimidate political opponents. We also may expect there a measure of bending—sometimes breaking—the truth. By design of the Founders, the judicial branch stands apart. We rule on facts, not on *36 supposition or conjecture, and certainly not on fabrication or propaganda. I urge my colleagues on this court to act swiftly to vacate the majority’s order before the illegal deployment of troops under false pretenses can occur. Above all, I ask those who are watching this case unfold to retain faith in our judicial system for just a little longer.

Footnotes

[2] 25-6268

[3] 25-6268 1 Declaration of Camilla Wamsley, ICE, Enforcement and Removal Operations (ERO), Field Office Director (FOD), Seattle Field Office (Wamsley Decl.). ICE is the largest investigative branch of DHS and is charged with the

[4] 25-6268 enforcement of more than 400 federal statutes. Wamsley Decl. ¶ 6. The Enforcement and Removal Operations (ERO) Seattle Field Office covers Oregon, Alaska, and Washington, and has three offices in Oregon, including the Portland sub-office, and 30 officers. Id. ¶¶ 2, 3. 2 Declaration of Robert Cantu, Deputy Director, Federal Protective Service (FPS), Region 10 (Cantu Decl.). The Federal Protective Service is a federal law enforcement agency under the umbrella of the Department of Homeland Security, which provides security services for federal buildings. 3 Declaration of Craig Dobson, Assistant Chief of Operations, PPB (Dobson Decl.).

[5] 25-6268

[6] 25-6268

[7] 25-6268

[8] 25-6268 4 These emails are attached to the declaration of Brian Hughes, PPB, Central Precinct Commander, (Hughes Decl.). The record does not include summary daily reports from PPB officers monitoring the ICE facility for any dates before September 1.

[9] 25-6268

[10] 25-6268

[11] 25-6268

[12] 25-6268

[13] 25-6268 5 Declaration of Franz Schoening, City of Portland, Commander, Specialized Resources Division (Schoening Decl.).

[14] 25-6268

[15] 25-6268

[16] 25-6268 6 In response to disturbances in Los Angeles arising from federal enforcement of immigration laws, the President invoked § 12406 “to order 4,000 members of the National Guard into federal service for 60 days to protect federal personnel performing federal functions and to protect federal property.” Newsom, 141 F.4th at 1040. The State of California challenged the deployment, and a district court issued a TRO enjoining the government’s actions. Newsom v. Trump, 786 F. Supp. 3d 123 (N.D. Cal. 2025). We stayed that order pending appeal, concluding that the President would likely prevail on the merits that he had acted within his lawful discretion in invoking § 12406(3), and that the other stay factors weighed in favor of granting the stay. Newsom, 141 F.4th at 1055–56.

[17] 25-6268 7 Plaintiffs have since filed an amended complaint, which added the State of California as a plaintiff.

[18] 25-6268 8 At oral argument, Defendants’ counsel explained that they did not appeal the second TRO, which relies on the same legal reasoning as the first TRO, because they concluded that the district court would be required to dissolve the second TRO if Defendants succeed on their motion for a stay pending appeal.

[19] 25-6268

[20] 25-6268

[21] 25-6268

[22] 25-6268

[23] 25-6268 9 The district court did not define what period would be encompassed in an assessment of the facts “at the time,” of the President’s federalization order. Instead, it cited Newsom as considering facts “leading up to” the President’s order and then referred more generally to “days” and “weeks” leading up to that order. Id. at *9– 10. It is unclear what facts may be considered under the district court’s standard to determine whether there is a colorable basis for the President’s decision. 10 The district court also clearly erred in characterizing the events in September leading up to and preceding the federalization order. See Elosu v. Middlefork Ranch Inc., 26 F.4th 1017, 1023 (9th Cir. 2022). There were numerous instances of violent and disruptive behavior at the ICE facility in September. On September 1, 200 protesters were disrupting operations at the ICE facility, and ICE agents conducted “several push outs,” arrested protesters, and had to deploy gas to clear the front of the building to move vehicles. Hughes Decl. Exs. 1, 2. On September 6, there were 50-60 “very active” protesters that FPS was “pushing out” in order to get vehicles to enter the facility. Id. Ex. 5. On September 12, protesters delivered “a large quantity of sticks and bats” outside the facility. Id. Ex. 11. On September 19, there was a threat that a bomb would be detonated at the ICE facility, and although the attack did not occur, it required increased security and diverted agency resources. Wamsley Decl. ¶ 14. That same night, there were 50 protesters outside the facility, “black blockers” were reported “assaulting people,” including a

[24] 25-6268 journalist, and there was disturbance outside the building and FPS deployed munitions to break up the crowd. Hughes Decl. Exs. 18, 19. On September 27, there were 50 to 60 protesters outside the facility, blocking the driveway in anticipation of the shift change and refusing to leave, and “[m]any were dressed in bloc, with helmets, masks, and other PPE.” Id. Exs. 27, 28. While the factual background above more thoroughly details these events, even a cursory glance at the facts shows that the district court erred by simply discounting these facts when it concluded that the President did not have a colorable basis for his determination. See Newsom, 141 F.4th at 1052.

[25] 25-6268

[26] 25-6268 11 The dissent erroneously asserts that we “accept[] the government’s characterization of Portland as a war zone.” Dissent at 1. Clearly, we do not. 12 FPS has 776 law enforcement officers. Cantu Decl. ¶ 4. Of those, only 497 are inspectors, whose primary mission is to patrol and respond to law enforcement incidents at federal buildings. Id. The surge of 115 officers therefore comprises 23.1% of FPS’s total capacity in terms of inspectors. 13 The dissent speculates that “FPS sent a different 8 officers to Portland every week for 14 or 15 weeks, meaning that Portland’s drain on FPS’s staff from elsewhere on any particular day was 8 people.” Dissent at 18. But we know this is inaccurate because the PPB reported that on September 6, FPS had 26 officers on duty. Hughes Decl. Ex. 5.

[27] 25-6268

[30] 25-6268 15 The Seventh Circuit recently granted in part and denied in part a motion for a stay pending appeal of a TRO enjoining the President’s invocation of § 12406 in Chicago, Illinois. Illinois v. Trump, No. 25-2798, 2025 WL 2937065, at *1 (7th Cir. Oct. 16, 2025) (per curiam). The Seventh Circuit’s order is not binding and not analogous to the facts here. In Chicago, “the federal government has been able to protect federal property and personnel without the National Guard’s help.” Id. at *7. By contrast, protesters in Portland have actively assaulted federal employees for months, including by kicking them, Cantu Decl. ¶ 9, firing an M80 explosive firework, mortar, and other incendiary devices at them, id. ¶¶ 8-9, doxing them online, id. ¶ 11, threatening them with knives, a machete, and other objects, id. ¶ 10; Wamsley Decl. ¶ 11; Hughes Decl. Ex. 19, and attempting to blind them with high-powered flashlights as they leave the building, Wamsley Decl. ¶ 18. In Chicago, federal facilities “have remained open.” Illinois, 2025 WL 2937065, at *7. But in Portland, violent protests forced the federal government to close the Lindquist Building for over three weeks and keep the windows boarded once reopened to protect federal personnel and property. Wamsley Decl. ¶¶ 18, 26. In Chicago, the “sporadic disruptions [] have been quickly contained by local, state, and federal authorities.” Illinois, 2025 WL 2937065, at *7. In Portland, protests have endured for months, and the PPB has been either unwilling or unable to respond to the disturbances at the Lindquist Building, see Wamsley Decl. ¶ 25; Hughes Decl. Exs. 15, 18, 19, creating an irregular and unsustainable strain on ICE and FPS, Cantu Decl. ¶ 4. While the situation in Portland is not static, it is distinguishable from the events in Chicago.

[31] 25-6268

[32] 25-6268 16 The dissent argues the last two factors merge only when the government is the party opposing the stay. Dissent at 33 n.13. The remaining stay factors nonetheless favor Defendants even if analyzed independently. Defendants have an interest in the protection of federal agents and the faithful execution of the law, and preventing the executive from protecting federal agents and property amounts to irreparable harm. In contrast, Plaintiffs suffer no injury from federalization. They contend that federalization harms them by putting into federal service members of the National Guard who would otherwise be serving in state roles. That possibility is not only speculative, it is inherent to a system where National Guard units may temporarily be federalized pursuant to a federal statute. And the public interest independently favors allowing, on a temporary basis, a limited number of National Guard members to ensure federal personnel and property are protected from violence. Thus, an independent analysis of the remaining Nken factors has no practical bearing on the outcome. See 556 U.S. at 426.

[33] 25-6268

[34] 25-6268

[35] 25-6268

[36] 25-6268 I concur in the Per Curiam order which concludes that the government is likely to succeed on the merits of its claim that the President’s assessment reflects a colorable assessment of the facts and law to sustain his designation under 10 U.S.C. § 12406(3). Normally, for a stay pending appeal, that would be enough. But given that these legal issues are moving quickly, I write separately to emphasize four points. First, under current Supreme Court precedent, presidential determinations under § 12406 are not reviewable by federal courts. Second, Plaintiffs have serious standing issues because they only seek injunctive relief that this court lacks the power to issue and Plaintiffs lack an injury in fact. Third, while the Per Curiam order takes a more fact bound approach to the issues presented, the President’s determination under § 12406(3) and § 12406(2) are bolstered by the history and tradition of the early Militia Acts. Fourth, the likely lawfulness of the President’s determination under § 12406 is bolstered by the executive’s proportionate response to the events in Portland. I We held in Newsom v. Trump, 141 F.4th 1032, 1046 (9th Cir. 2025), that presidential determinations under § 12406 are reviewable. Similarly, because