Litigation Highlight: Fifth Circuit Again Invalidates Federal Drug-User Ban As Applied to Marijuana

On August 28, a Fifth Circuit panel ruled in United States v. Connelly that the federal law banning gun possession by users of illegal drugs is unconstitutional as applied to a marijuana user.  Connelly is the second panel decision from the Fifth Circuit to reach this conclusion (Daniels, which we covered here, was the first and was sent back down by the Supreme Court after it decided Rahimi).  Connelly demonstrates the narrow scope of the majority decision in Rahimi and presents a test case for how much, if at all, Rahimi alters the Bruen test.

The defendant in Connelly was charged with possessing firearms as an unlawful drug user and providing firearms to an unlawful drug user (her husband), both in violation of federal law.  In April 2023, the district court struck down the charges as unconstitutional under Bruen and invalidated § 922(d)(3)—the unlawful transfer charge—on its face in all applications.  The district court relied heavily on the Fifth Circuit’s decision in Rahimi, a decision that was reversed by the Supreme Court this past June. 

On appeal, a Fifth Circuit panel affirmed the district court’s determination that the laws in question could not be constitutionally applied to the defendant in the case but declined to strike them down on their face.  The panel first summarized the Supreme Court’s decision in Rahimi and broke the inquiry into two distinct analytical steps:

First, we ask whether the Second Amendment’s plain text covers an individual’s conduct. Second, we ask “whether the challenged regulation is consistent with the principles that underpin our regulatory tradition.”

After concluding that marijuana users are within “the people” protected by the Second Amendment, the panel turned to the analogical-reasoning step and evaluated three potential historical analogues offered by the government: “(1) laws disarming the mentally ill, (2) laws disarming ‘dangerous’ individuals, and (3) intoxication laws.”  First, the panel found “that marijuana intoxication is . . . most analogous to short-term mental impairment” but that history only evinces a tradition of disarming the mentally ill while they are actually incapacitated.  In other words, “there is no historical justification for disarming sober citizens not presently under an impairing influence.”  Moreover, the panel observed that the unlawful-user prohibition in § 922(g)(3) is “not limited to those judicially determined to be severely mentally ill” and who would pose a danger to others if armed.  Absent any evidence of continuing or permanent impairment due to marijuana use, the panel found that historical mental-illness prohibitions could not support applying the ban on the facts of the case.

Second, the panel considered historical laws targeting “dangerous” individuals such as political dissidents and Catholics.  The panel noted that these laws “were passed during wartime or periods of unprecedented societal upheaval” and had “unique socio-political motivations” beyond merely protecting public safety.  Marijuana users, the judges said, are not considered dangerous because they present a risk of political or religious insurrection.  The panel wrote that a proper analogue might be “repeat alcohol users,” but that “neither Congress nor the states disarmed alcoholics.” 

Third and finally, the panel evaluated laws restricting gun possession by intoxicated individuals and determined that, while “some laws banned carrying weapons while under the influence, none barred gun possession by regular drinkers.”  Founding-era laws, the panel said, primarily “concern[ed] military service [because] intoxicated servicemembers cannot perform their duties while impaired.”  Post-Reconstruction laws were entitled to less weight because they were far in time from the adoption of the Second Amendment, and those laws also were limited to carrying and did not ban gun possession.  Thus, the panel said, while intoxication laws line up along the justification metric, § 922(g)(3) imposes a materially greater burden than historical laws because it targets possession.  In the judges’ view, “[t]he analogical reasoning Bruen and Rahimi [] prescribed cannot stretch that far."  The panel, however, found both § 922(g)(3) and (d)(3) facially constitutional because the laws could operate constitutionally to “ban[] presently intoxicated persons from carrying weapons” and prohibit sales or transfers to such persons.

While the Connelly panel cited Rahimi extensively and did not once mention Daniels (a published Fifth Circuit decision that the Supreme Court GVR-ed after deciding Rahimi), the analysis here looks quite similar to the analysis in Daniels.  There, as here, the panel dismissed certain laws due to the political and social context in which they were enacted, drew a sharp distinction between laws that regulated carrying versus possession, and honed in on “ordinary drunkards” as the most apt historical comparison.  Connelly, then, illustrates the view that Rahimi simply did not change the Bruen test in any substantial way.  The “principle” that Connelly identifies—concern about the public carrying of firearms by those who are actively intoxicated or under the influence of drugs—is, by and large, the same historical justification-and-burden combination that the court relied upon earlier in Daniels. 

What’s more, the Supreme Court rejected irresponsibility as a historically-supported principle for disarmament in Rahimi.  That’s probably the government’s strongest argument in a § 922(g)(3) marijuana case, but it’s not one the government pushed here for obvious reasons.  Parts of Connelly suggest that the judges might be reading Rahimi to institute an even stricter test.  For example, the panel in various places implies that an individualized judicial determination of mental illness or dangerousness might be required to place a law on solid constitutional footing (that type of determination helped support the DVRO ban at issue in Rahimi, but it’s not a part of most group-based gun restrictions).

As another indication of the narrow view of Rahimi that prevails in the Fifth Circuit, on September 12 the panel that had earlier decided Rahimi entered a per curiam opinion on remand affirming Rahimi’s conviction and sentence.  That opinion noted only one substantive way in which the Supreme Court’s decision in Rahimi modified Bruen: by expanding the use of surety laws as a potential analogue to confirm that potentially dangerous individuals can be banned from both possessing and carrying guns.[1]  And Judge James Ho concurred to emphasize his view that the Supreme Court gave short shrift to concerns surrounding mutual restraining orders and that “potential faults” with the federal DVRO gun law at issue in Rahimi can and should be addressed in a future case.

Connelly also comes close to assigning dispositive weight to historical silences.  In her Rahimi concurrence, Justice Barrett wrote that it is “flawed” to “assume[] that founding-era legislatures maximally exercised their power to regulate, thereby adopting a ‘use it or lose it’ view of legislative authority.”  But the panel here seems to do just that, extensively tracing Founding-era alcohol consumption habits and then observing that, “[c]onsidering the ‘extremely high level of alcohol consumption in the early Republic,’ this handful of generally inapposite laws does little to help the government’s position.”  The absence of regulation, however, simply cannot provide much guidance on perceived constitutionality.  Moreover, the panel paints an incomplete picture of Founding-era alcohol use by ignoring the fact that many colonies, states, and municipalities around the time of the Founding prohibited public intoxication and often prosecuted individuals for violating those laws—even if such prosecutions resulted in only minor penalties.  And some states went further: Maine prohibited the sale of all alcoholic beverages except for medicinal purposes in 1851, and other states soon followed suit.  While it’s not clear to me how exactly such broader measures impact the Second Amendment analysis and whether they require a different outcome, the regulation of alcohol use writ large certainly seems relevant enough to surface in judicial decisions on § 922(g)(3).  Yet the panel here doesn’t discuss this aspect of the tradition at all, leaving the impression that alcohol use was not only widespread in early America but almost entirely unregulated.