A few weeks ago, we addressed the Executive Order on Transgender Service and our concerns regarding the treatment of transgender service members; the role of military lawyers; the misunderstanding of likely military risks and unit cohesion; as well as the availability of state guards as an alternative mode of service. We follow up here by discussing the latest developments including the new Action Memo from the Department of Defense (DoD) and the Equal Protection aspects of the Talbott v. Trump and Shilling v. United States litigation challenging the Executive Order. We begin with a quick overview of Judge Ana Reyes’s and to a lesser extent Judge Benjamin Settle’s latest rulings as well as the newly issued DoD Action Memo.
As for analysis, first, we address Judge Reyes’s and Judge Settle’s approach to military deference and what they view as constituting (or not) a reasoned justification deserving of deference. Second, we discuss these judges’ response to the DoD’s new guidance and what that might mean for a possible future executive order excluding individuals with gender dysphoria from military service. We also speak briefly about how the DoD’s response to Judge Reyes’s ruling fits into a larger pattern of how the administration is responding to judges and others who criticize the administration’s actions in the security space.
In Part II of this post, we consider what the new Action Memo would mean for mental health access in the military more generally were the injunction to be lifted or more broadly were the order deemed constitutional. Finally, we conclude with what lessons other possible litigants might take from these rulings as regards the Veterans Administration and the invocation of the Insurrection Act and Martial Law.
I. Background: The Granting of the Preliminary Injunctions and the March 21 Action Memo
In an excoriating 79-page opinion, Judge Reyes granted a nationwide preliminary injunction. On March 27, Judge Settle did the same, noting, “The government’s arguments are not persuasive and it is not an especially close question on this record.” They both concluded the Executive Order was unlikely to survive judicial review on Equal Protection grounds. After articulating similar deficiencies in the limited reasoning offered for the ban, both concluded plaintiffs were likely to succeed on the merits under the following three (or four for Judge Reyes) rationales:
- Transgender individuals are members of a quasi suspect class: they face discrimination, contribute to society, are part of a discrete group with immutable characteristics, and constitute a minority lacking political power. The Executive Order would thus likely fail the relevant intermediate scrutiny standard of review.
- The exclusion of transgender individuals on the basis of gender dysphoria is discrimination on the basis of sex. The Executive Order would thus likely fail the relevant intermediate scrutiny standard review.
- The government could not articulate a relationship between the Executive Order and legitimate state interests in military readiness, unit cohesion, and cost reduction. The Executive Order would thus also likely fail the much more forgiving rational basis standard of review.
- For Judge Reyes, the Executive Order was also so infused with animus as to be inexplicable on other grounds, providing another reason the Executive Order would likely fail rational basis standard of review. Judge Settle left open the question of animus and did not rely on it for his preliminary injunction grant.
II. Judge Reyes and Judge Settle: Findings of Fact and Examination of Evidence
A. What evidence and reasoning the government offers is not sufficient for deference or to satisfy relevant levels of scrutiny
Both Judge Reyes and Judge Settle acknowledged cost, military readiness including deployability, and unit cohesion to be legitimate state interests. Both agreed that deference to military reasoning is appropriate when there is reasoned, professional analysis by the Executive Branch on matters within military expertise. But neither would defer if the judgment did not make sense or ran contrary to the legitimate state interests. Both concluded the material considered by Department of Defense in constructing its policies to implement the Executive Order to be insufficient to justify deference on any offered rationale. In this section, we review their assessment of the four pieces of evidence informing the DoD Policy: (1) the Mattis Policy limiting transgender service under the first Trump administration, (2) the 2021 review by the DOD Psychological Health Center, (3) a 2025 medical literature review conducted by the Office of the Assistant Secretary of Defense for Health Care, and (4) a review of cost data.
1. The Mattis Policy
The judges were both respectful of the Mattis Policy, but neither found it a source allowing them to defer to the military’s judgment. Both discussed approvingly the significant time and expertise that informed the transgender service policy during the first Trump administration. They noted the policy was generated only after convening a panel of experts and consultation with transgender service members and military and civilian medical professionals with expertise in care and treatment of gender dysphoria. But Judge Reyes found it important that the findings were “limited by ‘uncertainty’’’ and what she described as the earlier stages of medical literature on this topic. While the conclusions of such a panel might have been owed deference by the courts in 2018, she emphasized both that the Mattis Policy was then informed by available military practice and experience (i.e., transgender service in the Obama administration) and the best available medical data at the time, and that, in contrast, the Hegseth Policy was not informed by either the most recent available military practice and experience (i.e., open transgender service under the Biden administration) or by the newly available “overwhelming [medical] conclu[sion] that gender dysphoria is highly treatable.” Judge Settle emphasized an additional point—that the Mattis Policy acknowledged that the honoring the existing commitment to then currently serving transgender service members “including the substantial investment it has made in them” outweighed the risks and thus allowed them to continue serving in their preferred gender, change their gender marker, and receive medically necessary gender affirming care. The Hegseth Policy made no such allowance nor offered a reasoned explanation for why it failed to do so.
2. AMSARA (the DoD’s Psychological Health Center of Excellence and the Accession Medical Standards Analysis and Research Activity)
The judicial assessment of the 2021 AMSARA is a similar story to the Mattis Policy. Judge Reyes and Judge Settle questioned neither the validity nor the relevance of the AMSARA. Yet both concluded that the DoD Action Memo inappropriately cherry picked two of its findings to reach the conclusion that transgender service members affect military readiness via deployment rates and unduly burden other service members. While the DoD cited the AMSARA finding that rate of disability evaluation was significantly higher for transgender service members than other service members, both judges noted that evaluation rates were not relevant as that information was not presented in its full context. Judge Settle pointed out that the AMSARA noted that “members of the transgender community are encouraged (and in many cases required)” to be evaluated more than cisgender peer.” And moreover, the actual relevant findings included that the rates of transgender service members who experienced disabling conditions was comparable to all service members evaluated for such. In other words, the military cannot rest its assessment on evaluation rate differences that were either required or encouraged particularly when the AMSARA also concluded that actual disability rates did not differ.
The second finding cited by the DoD relates to deployability—a criteria both judges accepted as a legitimate state interest. The DoD noted that the AMSARA found “nearly 40% of Service members with gender dysphoria in observed cohort were non-deployable over a 24 month period.” Neither contested this finding, but instead both observed that the AMSARA itself described the finding as limited as it had no data from cisgender service members to serve as a baseline. In other words, the AMSARA itself was unable to draw any conclusions as to whether the non-deployability rate of the transgender cohort was meaningfully different than from the overall non-deployability rate. Both judges seemed to signal that valid evidence of higher than cisgender deployability rates for transgender troops would be the kind of evidence calling for judicial deferral.
3. 2025 Assistance Secretary of Defense for Health Affairs Medical Literature Review
Third verse same as the first and second. The DoD cited the 2025 medical literature review for findings related to high rates of suicidal ideation and attempted suicides for transgender individuals. It referenced: the 13 times higher suicide attempt rate for transgender individuals as compared to cisgender individuals; the two times greater likelihood of transgender individuals receiving a psychiatric diagnosis; and the low-to-moderate evidence for the success of transgender mental health and gender affirming care. Again, both judges accepted the legitimate state interest in the mental health of soldiers as it affects various aspects of military readiness. But both judges questioned the relevance of the evidence cited given that none of the studies in the medical literature review was specific to military data and the military already engages in screening for suicide and other mental health conditions. In other words, while the studies may be respectable, valid studies and the review a competent and accurate one, the review tells us nothing specific about the rates of suicidal ideation and attempt rates for those who have been or would be in the future otherwise deemed fit for medical service.
And perhaps equally important, both judges pointed out that many of the overall conclusions of the literature review run contrary to the Hegseth Policy. For instance, the literature review demonstrates that access to gender affirming care and protections against discrimination mitigate the suicide risk for transgender individuals. In addition, Judge Reyes notes that every meta study (i.e., study of studies) in the literature review found gender affirming treatment, both hormones and surgeries, improve psychological well-being, mental health, and quality of life. She concluded that the Action Memo summary was “misleading” and thus the review “contradicts, rather than supports, the conclusions the Action memo draw from it.”
4. Cost
The Action Memo cited 2015-2024 data to conclude that the DoD spent approximately 52 million dollars providing gender dysphoria care. This averages out to approximately 5.2 million a year. Judge Reyes and Settle did not dispute these costs, but yet again questioned whether such costs could justify a ban in the absence of additional information. The DoD offered no baseline assessment of why such costs were important to any of its goals or even why such costs were high compared to other medical costs. Judge Reyes noted that the military might have suggested the costs ought to be compared to cisgender soldiers rather than to the overall budget, but they failed to do so. She suggested that the DoD must put the cost in context. For example, how did it conclude 5.2 million a year for gender affirming care is unjustified while 41 million a year for Viagra remains untouched? And moreover, the DoD must account for both sides of the ledger. The DoD cannot simply count the costs of gender affirming care without also counting the costs of discharging and replacing currently serving transgender service members.
B. What is Missing
Both judges agreed that the record was lacking in the kind of reasoning supported by evidence necessary for deference. Reyes repeatedly stressed the speed of the order and the ensuing implementing policy at the expense of consultation with uniformed military leaders, the absence of a detailed study group, the limited analysis of existing evidence, and the absence of a variety of kinds of data. Judge Settle went so far as to describe the DoD’s ultimate position as lacking “any evidence” to support the military’s new judgment.
Both emphasized that the most important reasoning and evidence pertain to the analysis of open transgender service from 2021-2024. In the policies and in the litigation, the DoD offered no evidence of a decline in unit cohesion, recruiting, and military readiness attributable to open transgender service. In the policies and in the litigation, it offered no instance of a single transgender soldier lacking in integrity, honor, selflessness, or discipline.
Nor did the DoD meaningfully dispute the evidence introduced by plaintiffs. That evidence included testimony from someone responsible for reviewing gender transition requests and resolving issues relating to open transgender service who observed “no negative impact” and no complaints relating to unit cohesion or military readiness from allowing transgender service even as she was “responsible for resolving” issues relating to the Austin policy permitting open transgender service. She testified that the temporary non-deployability of some transgender service members was “no different than the myriad medical reasons that any service member might become temporarily non-deployable.” Another military personnel responsible for implementing and administering the policy testified he received no complaints and the Austin policy did not require ‘any significant changes to the DoD health care system.” Neither judge thought that the government’s rebuttal to this evidence was sufficient for deference. They did not find compelling the government’s declaration from the current assistant secretary of defense for manpower and reserve affairs who suggested that the declarants were too far from individualized command to get service member complaints on readiness, unit cohesion and deployment. Both seemed to expect some affirmative evidence of problems from the government when none was forthcoming.
C. The Trump 2.0 Policy is Meaningfully Different than all the Policies that Preceded It
Judge Reyes and Judge Settle both assessed the history of transgender service policies. In so doing, each discussed what made the evidence in those policies worthy of deference. Reyes emphasized the RAND report issued under the Obama administration’s working group commissioned the National Defense Research Institute which is “not a fly-by-night operation. It is ‘a federally funded research and development center sponsored by the Office of the Secretary of Defense, the Joint Staff, the Unified Combatant Commands, the Navy, the Marine Corps, the defense agencies, and the defense Intelligence Community.” Because she could determine who did the assessment and that those individuals were part of a credible operation, she could credit the expertise and the credibility of such an assessment. In discussing the first Trump administration’s transgender ban as implemented by the Mattis Policy, Judge Reyes noted the working group convened by Secretary Mattis and the 2018 DoD report that emphasized the “uncertainty surrounding the view of medical professionals.” She explained why the D.C. Circuit found a significant change from the 2017 Memo in the Trump I administration to the 2018 policy informed by Mattis’s study and memorandum. Similarly, the Biden Executive Order allowing transgender service relied on a DoD study, congressional testimony from multiple high ranking officials across the service branches declaring no knowledge of issues of “unit cohesion, disciplinary problems, or issues of morale” from open transgender service, statement from the former surgeon general on medical fitness, a working group that surveyed “all available scholarly evidence” and consulted with “medical experts, personnel experts, readiness experts, health insurance companies, civilian employers, and commanders whose units included transgender service members.” In contrast, both Reyes and Settle noted that they could not even ascertain who made the assessment that a transgender military ban was needed, much less who if anyone those individuals consulted in making such a determination.
D. The EO and Implementation is an unconstitutional transgender ban not remedied by the new Action memo
Judge Reyes’ grant of the preliminary injunction opened with a nod to the larger context in which the Trump administration is challenging the lawfulness of judicial actions. She wrote, “The Court does not issue this preliminary injunction lightly. Judicial overreach is no less pernicious than executive overreach. But the coordinate branches must, ‘by their mutual relations, be the means of keeping each other in their proper places.’” To that end, she noted that the administration could have and may in the future craft a lawful policy limiting individuals with gender dysphoria from serving in the military, the door is open for another, better reasoned policy. For now, her job is to assess the ban in front of her. In numerous places, she pointed out the kind of evidence and reasoned decision-making that would oblige a judge to defer to the executive branch’s determination and how its absence allowed for an alternate conclusion.
After Reyes’ grant of the preliminary injunction, but prior to Settle’s opinion, the Department of Defense issued a March 21 memorandum to provide additional guidance on implementation of the relevant Executive Order. The memorandum did the following:
- It defined the phrase “exhibit symptoms with gender dysphoria” as referring to the diagnostic criteria in the DSM of Mental Disorders and explained that the language would only apply individuals who “exhibit such symptoms as would be sufficient to constitute a diagnosis.”
- It clarified that the primary method for identifying service members with a current diagnosis or history of, or exhibit symptoms consistent with gender dysphoria is through medical records.
- It directed unit commanders to require service members to self-report whether they have a current diagnosis of or history with, or exhibit symptoms consistent with, gender dysphoria during their annual periodic health assessment. Commanders were then to categorize such individuals as non-deployable.
- It instructed the processing of individuals identified under the processes outlined above for involuntary separation, unless they qualified for a waiver as detailed in previous documents.
The March 21 memorandum neither reiterated the language that “adoption of a gender identity inconsistent with an individual’s sex conflicts with a soldier’s commitment to an honorable, truthful, and disciplined lifestyle, even in one’s personal life” from the executive order nor renounced it. Nor did it introduce additional arguments, additional studies, or any other additional evidence in support of the Executive Order. The DoD then filed a motion to dissolve, or stay pending the appeal, the preliminary injunction as it believed the March 21 memorandum presents a significant change in fact.
Neither judge believed the Action Memo provided a significant change in fact that would alter the plaintiffs’ likely success on the merits. Judge Reyes issued an additional ruling denying the DoD’s motion. (For what it is worth, a third Judge Christine O’Hearn adopted Reyes’s reasoning in issuing a temporary restraining order for two individuals seeking to enjoin involuntary military hearings.) Both Reyes and Settle were unconvinced by the DoD’s attempt to distinguish between a policy aimed at individuals with gender dysphoria and transgender individuals. As they both pointed out, only transgender individuals suffer from gender dysphoria even if not all transgender individuals suffer from gender dysphoria. Though the implementing memo only used the term “gender dysphoria,” and never the term “transgender,” the combination of the expansiveness of the policy and virtual impossibility of satisfying the waiver conditions meant the policy is a de facto transgender ban despite the government attorney’s protestations to the contrary. Neither judge found the waiver provisions to cut against any of their findings and were skeptical that any transgender service member could in fact qualify for such a waiver. Nor did the new memo cut against Reyes’s conclusions on animus.
What would provide a significant change in fact would likely have to be a serious consideration of open transgender service under the Biden administration. Both judges rejected arguments that it was the plaintiffs must affirmatively provide such a study in order to deny the military deference. Rather, both judges seemed to imply that military deference cannot be given in the absence of assessment of the most relevant, most timely information. The change in fact underpinning this case is not the Action Memo but the success story of open transgender service uncontested by the DoD in building the record here.
E. Our Observations
These complaints about a lack of a detailed argument supported by recognizable evidence are emblematic of complaints about the second Trump administration writ large. To take just two headline-dominating examples, concerns about DOGE and cuts to various federal agencies as well as objections to visa cancelations and removals to an El Salvador prison, are rooted at least in part in concerns that the administration seems to believe that Executive Power is not just expansive, but also requires no sustained reasoned justification that can be questioned in court.
That said, unlike some other examples, the DoD has recognized the court’s authority and has abided by the injunction. As far as we know, none of the implementation procedures either in the early guidance or in the March 21 memorandum has been utilized post injunction. The Action Memo explicitly recognizes that it is not to be implemented unless the injunction is stayed or lifted entirely.
While adhering to the court’s authority, the administration is simultaneously following a familiar playbook used to pushback against any perceived challenges to its inherent authority: discrediting and attacking the credibility of the supposed challenger. Consider the administration’s attacks on Judge Reyes. In response to Reyes’s denial of reconsideration, Secretary of Defense Hegseth posted,
Since “Judge” Reyes is now a top military planner, she/they can report to Fort Benning at 0600 to instruct our Army Rangers on how to execute High Value Target Raids…after that, Commander Reyes can dispatch to Fort Bragg to train our Green Berets on counterinsurgency warfare.
First, note the use of scare quotes around “Judge” as if to suggest she is not in fact a duly appointed federal judge confirmed by the Senate. Second, note the disparagement of her willingness to impose her judgment at all. Who is she to make any determination about the military as she is not qualified to instruct on high value target raids or on counterinsurgency warfare? But one does not need to be a military expert in order to conduct equal protection review, and Judge Reyes did in fact defer to military experts. She repeatedly discussed the importance of the military testimony in favor of open transgender service and lamented the unwillingness of the DoD to offer any military testimony of its own on the actual consequences of open service under the Biden administration. Third, the gratuitous and inaccurate labeling of Judge Reyes as “she/they” (Judge Reyes uses she/her pronouns) is another way to try to undermine her judgment. The implicit suggestion is that Judge Reyes is biased because she is a member of the same group who is implicated by the transgender ban—remember the ban itself requires the use of she/her pronouns for persons assigned female at birth, or as the ban states “pronoun usage…must reflect a Service member’s sex.” Or worse still, that she is undeserving of respect at all because of the alleged use of she/they pronouns. In addition, the Department of Justice has also lodged an ethics complaint against Judge Reyes regarding her interactions with the Defendant’s counsel. For analysis of this complaint, we recommend this synopsis.
We have seen a similar administration response to others who do not defer to the administration’s judgment. Think of President Trump’s call to impeach Judge James Boasberg after he ruled the administration could not invoke the Alien Enemies Act to deport individuals without a hearing. Or think of the attacks on Jeffrey Goldberg of The Atlantic. Despite the accuracy of Goldberg’s reporting as confirmed by the release of the text chain, the administration made several targeted and personal attacks against Goldberg. Secretary Hegseth suggested that Goldberg had a history of reporting on hoaxes and was a “deceitful and highly discredited so-called journalist.” President Trump referred to Goldberg as a “total sleazebag.” What they have not done is outline how the administration will fully comply with security protocol so as to avoid future lapses.
Given this playbook, we find it unlikely the administration will engage in the thoughtful, deliberative, and slow-moving fact-finding process required by Reyes and Settle. While the history of the Muslim ban under the first Trump administration, as outlined by Judge Reyes, does provide a pathway on how to move from an unconstitutional to constitutional Executive Order by engaging in reasoned deliberation, we find it unlikely that such a deliberative process would result in a record supportive of banning transgender members. Rather we worry the administration will continue to publicly make attempts to discredit and attack these judges, transgender service members, and any public effort—whether by attorneys, journalists, or concerned citizens—to oppose the administration’s efforts to ban transgender members. Perhaps the administration is trying to build public support in case it does eventually refuse to follow the injunction or to place pressure on more politically sensitive or attuned circuit court judges or Supreme Court justices who will eventually review the injunction on appeal.
Please stay tuned for our next post on the impact this litigation will have on mental health care of service personnel and on future challenges to the administration’s control, oversight, and deployment of the military.
The views expressed in this post are those of the authors and not of the Department of Defense or any military department.
