INVESTIGATION: Could Trump Mobilize the Military to Secure the 2026 Midterms?

An Often-Forgotten Founding Principle

To understand why this issue causes such concern among legal experts, we must return to a fundamental constitutional principle. The U.S. Constitution primarily entrusts the states with the administration of federal elections through what legal scholars call the Elections Clause: the states determine the “Times, Places, and Manner”—that is, the dates, locations, and procedures for voting—subject to Congress’s oversight authority. This division of powers is not a technicality. It constitutes a structural safeguard against any excessive concentration of electoral power in the hands of a single federal executive.

This principle explains why neither the Constitution nor any founding law authorizes the president to involve the military in the organization or supervision of U.S. elections. This is not an accidental omission. It is a deliberate choice made by the framers of the Constitution and by Congress throughout history, who sought to keep the federal military apparatus separate from the local and state electoral process. The few laws that address this issue are unequivocal: they prohibit such involvement in virtually all conceivable cases.

A deliberate safeguard against abuse of power

This legal framework did not arise from an excess of theoretical caution. It responds to a very real historical fear: that of an executive branch using armed force to influence the will of the people. The drafters of nineteenth-century federal laws had in mind specific episodes in American history where federal troops had been used—or threatened to be used—to influence contested elections. It is this institutional memory that explains the severity and precision of the laws still in force today.

The result of this framework is a system in which electoral authority remains fragmented among fifty states, each responsible for its own polling places, its own voting machines, and its own certification procedures. This fragmentation, often criticized for its technical inefficiency, is in fact a major safeguard of democracy: it makes it structurally more difficult for any federal executive—whichever one it may be—to exert centralized control over the election results.

The complexity of the U.S. electoral system—its fifty different jurisdictions and disparate rules—is often criticized. But this very complexity is also what makes a centralized power grab much more difficult to carry out.

A law that is nearly a century and a half old

At the heart of this legal case lies a federal law known as the Posse Comitatus Act, codified at 18 U.S.C. § 1385 and enacted in 1878. The text of this law, as confirmed by the official United States Code, is remarkably clear: anyone who, except in cases and under circumstances expressly authorized by the Constitution or an Act of Congress, willfully uses any part of the Army, Navy, Marine Corps, Air Force, or Space Force “as a posse comitatus or otherwise to enforce the laws” is subject to a fine or imprisonment for up to two years.

This law is not merely symbolic or obsolete. It remains fully enforceable and is regularly invoked in federal courts, including in recent cases involving the deployment of the National Guard in domestic law enforcement contexts. The principle it protects is simple to state but fundamental in its implications: the U.S. federal military is not intended to replace civilian police forces or to intervene in the enforcement of domestic laws, except in cases explicitly provided for.

Narrow exceptions, never designed for elections

The law itself provides for exceptions, the best known of which is the Insurrection Act—a set of provisions dating back to 1792 and amended since then—which authorizes the president to deploy federal troops under specific circumstances: at the request of a state facing an insurrection, to enforce a federal court order, or when the enforcement of federal law becomes impossible through ordinary judicial channels. These conditions are deliberately restrictive. They were never designed to cover an election scenario, and no provision allows for a broad interpretation that would authorize the president to invoke the Insurrection Act to oversee an election.

It is telling that even the most ardent supporters of a military deployment around polling places—such as Bannon in his public statements—feel compelled to invoke the Insurrection Act to justify their proposal. This legal reflex, however clumsy, implicitly confirms what legal scholars have documented: without such a legal workaround, the use of the military in an electoral context would be immediately and outright illegal under the Posse Comitatus Act.

There is something almost ironic about the fact that proponents of a military deployment at the polls must themselves seek a legal justification. This proves, better than any argument, that current law already prevents them from doing so.

A Rare Form of Institutional Continuity

The significance of this legal framework is also measured by its longevity. The Posse Comitatus Act and the accompanying election laws have endured for more than 160 years, spanning approximately 80 federal election cycles, without any president ever openly deploying the military to intervene directly in the conduct of a national election. This continuity is no accident. It reflects a deeply rooted institutional consensus—historically shared by presidents across the political spectrum—that the U.S. military must never become a tool for exerting electoral pressure.

This consensus is grounded in specific regulatory texts. The Justice Manual, a reference document used by federal prosecutors at the Department of Justice, specifically governs the application of 18 U.S.C. § 592, which addresses offenses related to the presence of armed forces near polling places. This text is not some obscure internal memo: it structures the way federal prosecutions are initiated in cases of alleged violations of these electoral protections.

The Pentagon directive prohibiting operations at polling places

Complementing this framework is a Department of Defense directive on military support to civilian authorities, updated in 2018. This directive explicitly prohibits Department of Defense personnel, as well as the National Guard operating under Title 32 status, from “conducting operations at polling places.” This directive complements the existing legal framework by specifically targeting the internal military chain of command, independent of federal criminal laws already in force.

The combination of these protections—the Constitution, the Posse Comitatus Act, the guidelines in the Justice Manual, and the Pentagon directive—forms a multi-layered system designed precisely to make any attempt to use the military for electoral purposes unlikely, if not impossible, without simultaneously crossing several legal red lines. It is precisely this accumulation of safeguards that makes the statements reported by Lawfare so troubling: they suggest a willingness to test—or even circumvent—a system designed to withstand this kind of pressure.

Eighty election cycles without the military coming anywhere near the ballot box is no statistical coincidence. It is an institutional discipline built up generation after generation, and one that survives only if every president agrees to respect it.

A legal expert specializing in electoral democracy

Pooja Chaudhuri, senior legal counsel and deputy legal director at the Democracy Defenders Fund—an organization dedicated to defending U.S. democratic institutions—offers additional insight into this issue beyond that provided by Lawfare. Her analysis focuses not only on the formal legality of a military deployment but also on its concrete psychological impact on voters, long before such a scenario even materializes.

According to Chaudhuri, what matters is not just the action itself, but “the deterrent effect on voters”—a phenomenon of civic discouragement. If voters hear that the immigration agency ICE might be deployed near polling places, or that the rules for mail-in voting might change as the election approaches, many will simply give up on voting, regardless of the actual legality of these threats. This is a mechanism of preemptive intimidation that does not even require the actual implementation of a military deployment to produce its effects.

Mixed-status families: the first potential victims

Chaudhuri highlights a particularly sensitive issue: mixed-status families—that is, households where some members are full U.S. citizens and others are not—may choose to abstain from voting out of fear of consequences for the non-citizen members of the household, even though these U.S. citizens have a perfectly legal and indisputable right to vote. This dynamic transforms a theoretical threat into a concrete barrier to voter participation, without any law needing to be formally violated to produce this effect.

This is where one of the most troubling paradoxes of this issue lies: even though the U.S. legal framework makes a military deployment at the polls largely illegal, the mere public rhetoric suggesting this possibility—voiced by figures close to those in power—can be enough to suppress voting among certain populations, without any concrete military action ever having to be taken. The threat, in and of itself, becomes a political tool.

This is perhaps the most insidious aspect of this entire affair: you don’t even need to send a single soldier to discourage voters. All you have to do is hint at it, and the damage is already done before Election Day.

A direct accusation made by Common Cause

Dan Vicuña, senior policy director for voting and fair representation at Common Cause—a long-standing U.S. advocacy organization dedicated to civil rights and electoral transparency—offers one of the most direct critiques of this issue. According to Vicuña, the accumulation of these signals—statements about the military, rhetoric about fraud, and reluctance to clearly commit against a deployment—reflects a specific political objective rather than a legitimate concern for electoral integrity.

Vicuña summarizes his interpretation of events as follows: “What all of this boils down to is a desire to avoid any accountability to voters during the midterm elections, to ensure—or predetermine—the outcome of a midterm election that he believes will go badly for him.” This phrasing frames the debate not in terms of national security or electoral integrity, but in terms of personal political calculation—a serious accusation that deserves to be reported accurately, with full attribution.

A thread stretching back to 2020

Vicuña also links this issue to a broader historical context: “From the great lie of the 2020 election to inciting a violent insurrection to overturn a free and fair election, he has no respect for democratic norms… all of this is solely about his own power and his own ego.” This statement, also directly attributed, places current concerns about the 2026 midterms within a documented historical continuum: the 2020 disputes, the subsequent events widely covered by the American press, and now this new phase of tension surrounding the potential use of the military.

It is important to note that these remarks represent the opinion expressed by a leader of an organization dedicated to defending the right to vote, and not a fact established by a court. But their informational value remains: they illustrate how experts in election law interpret the trajectory of recent events, based on facts they themselves consider to be well-documented.

Accusing a president of seeking to predetermine an election outcome is a serious allegation. But it comes from an electoral law expert, not from an anonymous user on social media, and that difference matters greatly in how we should interpret it.

A documented escalation in rhetoric over several months

Steve Bannon’s remarks warrant close scrutiny, as they are not an isolated slip of the tongue but rather a repeated escalation of rhetoric over several months. According to reports corroborated by several U.S. media outlets, Bannon first mentioned deploying ICE agents around polling places, before going further by explicitly calling for the use of active-duty military units under the Insurrection Act. This progression—from federal immigration police to airborne troops of the regular army—illustrates a gradual normalization of an idea that, if articulated all at once, would likely have provoked more immediate rejection.

This strategy of gradual normalization is a well-documented phenomenon in political discourse analysis: an extreme proposal becomes progressively more acceptable in public debate when it is preceded by intermediate proposals that seem, by comparison, more moderate. The fact that Bannon first mentioned ICE before explicitly naming the airborne divisions is likely no rhetorical coincidence.

The White House: Between Denial and Ambiguity

In response to these statements, the White House’s official reaction has been mixed. Officials have, at times, described the idea of such a deployment as “categorically false,” according to reports in the American press. But this occasional denial has not prevented the president himself from making ambiguous remarks about the role of the states in election administration, at one point referring to them as mere “agents of the federal government,” an interpretation that directly contradicts the constitutional framework described above, in which it is precisely the states—and not the federal government—that hold primary authority over the organization of elections.

This contradiction between occasional official denials and the president’s own statements fuels precisely the kind of uncertainty that legal scholars like Chaudhuri identify as problematic: regardless of what is ultimately implemented, the ambiguity surrounding these intentions is already creating a measurable climate of anxiety among voters and local election officials.

Saying one thing and then its opposite is not a communication strategy; it is an admission of unease. When the White House denies one day what the president seems to confirm the next, the ambiguity itself becomes the message.

Los Angeles: A Real-World Test of the Posse Comitatus Act

Although not directly related to the elections, a recent court case sheds significant light on this issue. A federal judge ruled that the federalization of the National Guard in Los Angeles to support immigration enforcement operations violated the Posse Comitatus Act. This decision, handed down by a federal district judge, rejected the administration’s arguments that the president possesses an inherent authority to protect federal functions, independent of existing statutory restrictions.

This case is relevant to the electoral context because the same legal arguments—concerning an alleged inherent presidential power beyond the control of the courts—could theoretically be invoked in an electoral context if an administration sought to justify a military deployment around polling places under the guise of protecting a “federal function.” The federal court’s rejection of this argument in the Los Angeles case is therefore an encouraging sign for those who fear this logic might be extended to the electoral sphere, but the appeal of this decision to higher courts leaves significant uncertainty.

Sections 592 and 593: Little-Known Criminal Protections

Beyond the Posse Comitatus Act, two other provisions of the federal criminal code—18 U.S.C. §§ 592 and 593—specifically criminalize the deployment of troops near polling places or military interference with voting, the conduct of elections, or election officials. These provisions apply to both career military personnel and members of the National Guard called to federal duty, and another criminal provision makes it a crime for a civilian official—such as a member of the President’s cabinet—to aid or abet a violation of these protections.

These laws, largely unknown to the general public, form a second line of legal defense—distinct from the Posse Comitatus Act but just as binding. Their very existence demonstrates the extent to which U.S. lawmakers, as early as the nineteenth century, had anticipated and specifically sought to prevent this type of scenario involving military interference in elections—long before anyone could have imagined the current political circumstances.

What strikes me, as I delve into these old statutes, is their almost prophetic precision. Lawmakers from another century had already foreseen that one day, someone would be tempted to use the military to influence an election, and they had taken care to draft the law accordingly.

Electoral stakes viewed as existential by the ruling majority

To understand why these scenarios are taken seriously by recognized legal experts, we must place the debate within its immediate political context. The 2026 midterm elections represent a major challenge for the current administration: the Republican majority in the House of Representatives—and potentially in the Senate—is at stake, against a backdrop in which several national polls indicate difficulties for the ruling party. It is precisely this type of context—where the loss of legislative power appears to be a credible possibility—that experts identify as conducive to temptations to circumvent the usual rules.

This interpretation is not limited to the commentators cited in this report. It has also been articulated by former legal advisors to previous administrations—of varying political leanings—who have warned in public op-eds that the president might not accept unfavorable election results without a fight, and that he might seek to use the considerable powers of the presidency, or even the armed forces, to challenge a verdict he finds unpalatable.

A 2020 Precedent That Weighs Heavily on the Analysis

The shadow of the 2020 election looms over this entire issue. The documented attempts to challenge the results of that election—including draft executive orders aimed at having the National Guard seize voting machines—constitute a precedent that legal experts cannot ignore when assessing the likelihood of similar scenarios in 2026. This is not abstract speculation: legal advisors close to the president reportedly drafted executive orders to that effect at the time, although these drafts were never implemented.

It is this continuity between the intentions expressed in 2020 and the more recent statements documented by Lawfare that transforms a series of isolated remarks into a coherent trajectory in the eyes of experts. An isolated event could be dismissed as an excessive statement with no follow-through. A documented pattern repeated over several years, however, warrants serious and ongoing scrutiny.

A politician may say something regrettable once and apologize. But when the same idea resurfaces, year after year, in different forms but with the same objective, it becomes difficult to dismiss it as a mere slip of the tongue.

Seizure of ballots before they are certified

Among the scenarios documented by the most rigorous legal and journalistic analyses is the possibility that the federal government might attempt to seize ballots, voting machines, or voting data in key jurisdictions before all votes have been counted. This scenario is based on a concrete precedent: an FBI raid aimed at seizing election documents from the 2020 election—including ballots and voter registration data—in a county in Georgia. Although this action does not directly concern the 2026 midterms, experts say it demonstrates the federal government’s willingness to intervene in local election matters in a manner considered unusual.

Such a scenario, if repeated on a larger scale and at a critical moment in the vote count, could, according to some analysts, compromise the chain of custody for ballots, making it more difficult to clearly certify results in certain contested districts. This type of maneuver does not necessarily require a massive military deployment: it can rely on court orders obtained under conditions deemed politically motivated by legal observers.

Deterrence through the Presence of Federal Forces

A second scenario, more directly related to the central theme of this report, involves the deployment of federal agents or military personnel near polling places located in areas with large immigrant populations, with the explicit or implicit aim of discouraging voter turnout. This scenario does not require a massive armed intervention or the use of force: the mere visible presence of federal forces, combined with alarmist public rhetoric, may be sufficient to produce the desired effect, as Chaudhuri’s analysis of the deterrent effect highlights.

It is essential to note that most experts consulted in the American press, while taking these scenarios seriously, believe that their full implementation would face significant institutional resistance: federal courts, local election officials—including in Republican-led states—and possibly even within the military hierarchy itself, which has historically been reluctant to engage in missions perceived as politically compromising.

What reassures me somewhat in this otherwise bleak situation is that even the most concerned experts do not predict guaranteed success for these maneuvers. U.S. institutions are not defenseless, but neither are they invincible.

Senators Speak of a “Red Alert” Moment

Several Democratic lawmakers have publicly expressed their concern over these developments, sometimes in very direct terms. Some have described the situation as a “red alert” for American democracy, while others have warned that the administration might seek to turn certain cities into “militarized zones” as the election approaches. These statements, although coming from politicians on the opposing side of the current administration, are based on documented facts rather than mere rhetorical accusations, which justifies their inclusion in a factual analysis.

State-level elected officials, notably Democratic governors, have also expressed similar concerns regarding the prolonged maintenance of National Guard deployments in certain cities beyond the dates initially announced—deployments that, according to these officials, would coincide in a troubling manner with the 2026 election calendar.

Secretaries of State Preparing Legal Responses

Faced with these uncertainties, state election officials—including those in states led by both major parties—have begun to plan concrete legal responses in the event of an attempt at federal interference in the electoral process. Senators serving on committees overseeing federal elections have organized meetings with election experts to “stress test” various threat scenarios, explicitly including the deployment of federal agents near polling places and the seizure of ballots by law enforcement.

This preventive mobilization, although political in nature and therefore subject to partisan interpretation, reflects a climate in which legal preparedness for a scenario of military or federal interference is no longer considered a marginal possibility but rather a serious enough threat to warrant concrete simulations by election officials and their legal advisors.

When senators organize electoral crisis simulations with experts, it is no longer paranoia. It is risk management—the same logic that would be applied to a natural disaster or a cyberattack.

Not all scenarios are created equal

It would be dishonest to present only the most alarming version of this issue. Several scholars and analysts, while acknowledging the gravity of the documented statements, believe that the likelihood of these scenarios being fully and successfully implemented remains limited, due to the many institutional obstacles already mentioned. A political science professor has thus pointed out that federal courts have already restricted the administration’s ability to deploy National Guard troops against the will of local and state authorities, which would significantly limit the options available for a widespread electoral deployment.

This same analysis also points out that a visible show of military force during an election period could have the opposite effect of what is intended: rather than discouraging voter turnout, it could instead further mobilize voters opposed to the current administration—a phenomenon observed in other comparable political contexts throughout recent democratic history.

Uncertainty Itself as a Risk Factor

Nevertheless, even the most measured analysts acknowledge one point in common with their more alarmist colleagues: the uncertainty surrounding the administration’s true intentions constitutes in itself a risk factor for the credibility of the electoral process. A professor specializing in constitutional law summarized this tension by noting that the legal definitions of key terms such as “insurrection” or “unlawful obstruction” remain deliberately broad in existing statutes, leaving considerable room for interpretation by the current executive branch—an ambiguity that has never been tested on the scale of a full national election.

It is precisely this legal gray area—rather than absolute certainty one way or the other—that best defines the current state of the debate among experts in American electoral and constitutional law. Neither total alarmism nor complete denial does justice to the actual complexity of the documented situation.

I far prefer this uncomfortable gray area to a fabricated certainty. Saying that we do not know exactly what will happen, while precisely documenting what has been said and written, seems to me more honest than any categorical prediction.

Institutions Put to the Test but Not Yet Broken

Beyond the specific case of the 2026 midterms, this report sheds light on a broader phenomenon observed for several years in the United States: the repeated testing of institutional norms that, historically, had never needed to be formally codified because they were universally respected through an implicit political consensus. The use of the military in an electoral context is one such norm that seemed so self-evident that it had almost never needed to be tested in court in the country’s modern history.

The very fact that renowned legal scholars, such as the authors of the Lawfare analysis, or leaders of voting rights organizations like Chaudhuri and Vicuña, feel compelled to document and publicly explain laws that are more than a century old is, in itself, a troubling indicator. Until recently, these laws had never needed to be explained to the general public because compliance with them was taken for granted.

The Role of Citizen and Journalistic Vigilance

This observation does not necessarily lead to a dire prognosis. Rather, it underscores the importance of continuous vigilance on the part of journalists, legal professionals, election officials, and citizens themselves to ensure that these legal safeguards remain effective in the face of potential attempts to circumvent them. This is precisely the role played by specialized publications such as Lawfare, or organizations such as Common Cause and the Democracy Defenders Fund, by methodically documenting every statement, every court decision, and every sign that might indicate a deviation from the norm.

This vigilance is not a matter of ideological bias. It constitutes an essential function of the fourth estate and of organized civil society in any functioning democracy: to bring to light what might otherwise remain hidden in the shadows of informal statements and carefully maintained ambiguities.

A democracy does not always die from a spectacular coup d’état. Sometimes it erodes slowly, through laws that are gradually disregarded, until the day they are openly violated no longer comes as a surprise to anyone. It is this slow drift that must be monitored.

What the Facts Establish—and What They Do Not Establish

At the conclusion of this report, it is important to summarize precisely what the documented facts actually allow us to assert. Recognized legal experts, writing in a reputable specialized journal, believe that the president might seek to use military or paramilitary means to influence the 2026 midterm elections. This assessment is based on documented statements by the president himself and his former strategic advisor, as well as the absence of a clear denial from his Secretary of Defense. It is also based on a direct precedent: the documented attempts to challenge the 2020 election through extraordinary means.

What the facts do not establish, however, is certainty that such a scenario will actually materialize, nor in what precise form. The U.S. legal framework—including the Posse Comitatus Act, the properly interpreted Insurrection Act, Sections 592 and 593 of the Criminal Code, the Justice Manual, and the Pentagon directive—constitutes a set of robust safeguards, recently tested and upheld by at least one federal court decision in a comparable context. These safeguards do not guarantee that abuses will never occur, but they make such a scenario legally perilous for anyone who might attempt to implement it.

Why This Vigilance Must Remain Fact-Based and Not Conspiracy-Theory-Driven

This report has been deliberately constructed using attributed quotes, verifiable legal texts, and documented legal precedents, while systematically avoiding any unverifiable speculation. It is this methodological rigor that distinguishes serious journalistic investigation from a conspiracy theory: every claim here can be traced back to its source, every quote includes an identifiable name and position, and every legal reference points to a publicly accessible text.

American democracy has survived eighty election cycles without direct military intervention in the voting process. Nothing in this report guarantees that this record will hold indefinitely, but neither is there any evidence that it will be broken in 2026. What is certain, however, is that the question itself deserves to be asked openly, rigorously documented, and monitored with the same factual vigilance that has, until now, made it possible to faithfully report every warning sign identified by experts whose expertise is not seriously contested.

I conclude this report without any certainty about what will happen in November, and that is exactly what intellectual honesty demands here. But I also conclude convinced that ignoring these documented warning signs would be a far more serious mistake than reporting them with caution.

Signed, Maxime Marquette, columnist