Teaching Explosives to a Known Would‑Be Violent Offender Is Unprotected Speech: Commentary on United States v. Arthur (4th Cir. 2025)

I. Introduction

In United States v. Christopher Clark Arthur, No. 24‑4306 (4th Cir. Dec. 3, 2025), the Fourth Circuit addressed, for the first time, the facial constitutionality of 18 U.S.C. § 842(p)(2)(B), the federal “bomb‑making instruction” statute, under the First Amendment’s overbreadth doctrine. The court also reviewed the application of the powerful terrorism enhancement in U.S.S.G. § 3A1.4 to a defendant convicted of teaching a confidential informant how to fortify his home with improvised explosive devices (IEDs) to kill federal agents.

Judge Agee, writing for the majority (joined by Judge Young), affirmed Arthur’s convictions and 300‑month sentence. The court held:

• Section 842(p)(2)(B) is not facially overbroad because the speech it targets—teaching or distributing information on explosives to a person known to intend a federal crime of violence—is largely “speech integral to criminal conduct” and therefore unprotected by the First Amendment.
• The district court properly applied (and in any event harmlessly applied) the terrorism enhancement under § 3A1.4 because Arthur’s offense both “involved” and “intended to promote” a “federal crime of terrorism,” namely the murder or attempted murder of federal officers under 18 U.S.C. § 1114.

Chief Judge Gregory dissented. He would have held § 842(p)(2)(B) facially overbroad, emphasizing:

• the extreme breadth of the statute’s defined terms (“explosive,” “destructive device,” “information pertaining to, in whole or in part”),
• the relative ease of proving “knowledge” (including via willful blindness), and
• the novelty and danger of criminalizing truthful, publicly available technical information based on the listener’s criminal intent, rather than the speaker’s own specific intent to facilitate crime.

This decision is a significant data point in the ongoing effort to define the boundary between protected advocacy or technical instruction, and unprotected assistance to violent crime—especially in an era when detailed weapons information is widely available online.

II. Summary of the Opinion

A. Holdings

The Fourth Circuit’s key holdings are:

• No facial overbreadth: 18 U.S.C. § 842(p)(2)(B) survives a First Amendment facial attack. The court characterizes the statute as primarily regulating “speech integral to criminal conduct,” analogizing it to aiding and abetting and rejecting the need to satisfy the Brandenburg incitement standard.
• Knowledge standard is sufficient: The statute’s requirement that the instructor “know” that the recipient intends to use the information to commit a federal crime of violence adequately narrows its reach and keeps it within the unprotected “speech integral to criminal conduct” category, notwithstanding the absence of a shared specific intent to commit the crime.
• Terrorism enhancement: Arthur’s offense both “involved” and “intended to promote” a “federal crime of terrorism” under 18 U.S.C. § 2332b(g)(5) and U.S.S.G. § 3A1.4. His § 842(p)(2)(B) offense necessarily encompassed his knowledge that the informant intended to murder federal officers in violation of § 1114, and he acted with the purpose of helping bring that crime about.
• Alternative variant sentence: Even if the terrorism enhancement were wrongly applied, any error was harmless because the district court expressly announced that it would impose the same 300‑month sentence as a variance based on the 18 U.S.C. § 3553(a) factors, and that sentence would be substantively reasonable even from a lower advisory range.

B. Factual Background in Brief

Arthur ran “Tackleberry Solutions,” a business selling tactical training and manuals to civilians preparing to resist what he cast as “tyrannical” government. After a customer who used his materials was found with pipe bombs, the FBI used a confidential informant, “Buckshot,” to obtain in‑person training.

Knowing Buckshot expected a return ATF visit and wanted to “be ready,” Arthur spent hours instructing him how to create deadly defensive setups—“fatal funnels,” “Spiderweb” explosive traps, Tannerite devices, shotgun‑launched grenades, and improvised detonators—explicitly framed as ways to kill federal law enforcement officers. Arthur was prosecuted for:

• Count 1: teaching explosives to a person, knowing that person intended to use the information to commit a federal crime of violence, in violation of § 842(p)(2)(B); and
• Counts 2–9: various unregistered destructive devices and firearms offenses.

He moved pretrial to dismiss Count 1 as facially overbroad; the district court denied the motion, and a jury convicted on all counts. The district court applied § 3A1.4, found an advisory life sentence (capped by statutory maximums), and then varied downward to 300 months. Arthur appealed the facial constitutionality of § 842(p)(2)(B) and the terrorism enhancement.

III. Detailed Analysis

I. Factual and Procedural Context

The majority sets out the facts partly to illustrate the type of behavior § 842(p)(2)(B) targets. Arthur’s operations were not mere online posting; they included direct, bespoke training. Crucially:

• Buckshot told Arthur specifically that ATF agents had come to his home and that he expected them to return.
• Arthur acknowledged this, framed the situation as a stark choice (“Stand and fight or be … not exactly where you’re supposed to be”), and assumed Buckshot would stay and fight.
• Arthur then provided detailed, step‑by‑step guidance for creating an explosive and firearms‑based ambush environment—electrified fence and dogs, IEDs by doors, Tannerite targets, remote detonators made from light bulbs, a “Spiderweb” with remotely triggered explosives and a sentry gun, and a shotgun converted to launch grenades.
• He explicitly linked these tactics to killing federal officers, referring to some setups as “freaking death box[es]” for agents.

On this record, there was little doubt that Arthur’s own case falls within the core of what Congress aimed at in § 842(p)(2)(B). That posture matters: it underscores that his attack is purely facial, not an as‑applied challenge. He does not claim the statute is unconstitutional as applied to him, but that it prohibits so much protected speech in other applications that it must be struck down altogether.

II. The First Amendment Overbreadth Challenge

A. The Overbreadth Doctrine Framework

The majority correctly starts with the Supreme Court’s articulation of overbreadth in cases like United States v. Hansen, 599 U.S. 762 (2023); United States v. Williams, 553 U.S. 285 (2008); and Virginia v. Hicks, 539 U.S. 113 (2003).

Key features of the doctrine, as the majority recounts:

• Ordinarily, facial challenges require showing there is “no set of circumstances” under which the law is valid (Salerno), and litigants may not assert third‑party rights.
• The overbreadth doctrine is an exception: where a statute “prohibits a substantial amount of protected speech relative to its plainly legitimate sweep,” courts may invalidate the statute facially, even at the behest of a defendant whose own conduct is unprotected.
• Overbreadth is “strong medicine” to be used sparingly and only when unconstitutional applications are both “realistic, not fanciful,” and “substantially disproportionate” to the statute’s legitimate reach (Hansen, Hicks, Ferber).

The analytical steps the Fourth Circuit follows (tracking Miselis):

1. Constrain the statute’s meaning, including by any reasonable narrowing constructions.
2. Identify the extent of protected speech the statute covers, and weigh that against its legitimate applications.
3. If overbroad, consider whether invalid portions are severable; but if not substantially overbroad, reject the facial challenge.

B. Scope and Structure of § 842(p)(2)(B)

Section 842(p)(2)(B) provides that it is unlawful for any person:

to teach or demonstrate to any person the making or use of an explosive, a destructive device, or a weapon of mass destruction, or to distribute to any person, by any means, information pertaining to, in whole or in part, the manufacture or use of an explosive, destructive device, or weapon of mass destruction, knowing that such person intends to use the teaching, demonstration, or information for, or in furtherance of, an activity that constitutes a Federal crime of violence.

The court emphasizes several limiting features:

• The statute is targeted to highly dangerous instrumentalities: “explosives,” “destructive devices,” and “weapons of mass destruction,” all defined elsewhere in Title 18.
• It covers only instructional or informational speech directly about the “making” or “use” of those devices, and only where that information is “distributed to any person” (suggesting a directed transmission, not mere general advocacy).
• Most importantly, it requires that at the time of the teaching or distribution, the speaker knows that the recipient intends to use the information to commit a “Federal crime of violence.” It is not enough that the information could be misused someday, or that misuse is foreseeable; the statute demands knowledge of a specific criminal intent in a specific person (“such person”).

The dissent, by contrast, stresses the breadth inherent in:

• the extremely wide statutory definitions of “explosive” (which can include gasoline, methane, and chemical mixtures with explosive potential), and “destructive device,” which extends to “any combination of parts … designed or intended” to convert a device into a destructive device;
• the phrase “information pertaining to, in whole or in part, the manufacture or use” of such items, which could reach general discussions of combustion, energy, or industrial processes; and
• the fact that liability turns on the listener’s criminal intent, not the speaker’s desire to further a crime.

These competing views of scope set up the core disagreement over overbreadth.

C. Majority’s Use of “Speech Integral to Criminal Conduct”

The majority anchors its First Amendment analysis in the category of “speech integral to criminal conduct,” first articulated in Giboney v. Empire Storage & Ice Co., 336 U.S. 490 (1949), and repeatedly acknowledged by the Court alongside obscenity, defamation, and incitement as an unprotected category (Stevens, 559 U.S. 460).

The reasoning proceeds as follows:

1. Nature of the speech. When someone:
   • teaches another how to manufacture or use explosives or destructive devices, and
   • does so knowing that the recipient intends to use that knowledge to commit a federal crime of violence,
   that instruction is not abstract advocacy; it is the functional equivalent of providing the tools needed to commit the crime. In the court’s view, but for such instruction the recipient would lack the means to carry out the planned offense.
2. Analogy to aiding and abetting. The majority repeatedly treats § 842(p)(2)(B) as akin to aiding and abetting:
   • It points to Rice v. Paladin Enterprises, 128 F.3d 233 (4th Cir. 1997), which upheld civil liability for a publisher of a “Hit Man” manual used in a murder‑for‑hire scheme, emphasizing the line between abstract advocacy and detailed, purposive facilitation of crime.
   • It invokes United States v. Miselis, 972 F.3d 518 (4th Cir. 2020), which described a second class of unprotected speech aside from incitement: “aiding and abetting of criminal conduct,” which may be proscribed “without regard to whether it’s directed [at] and likely to produce imminent lawlessness.”
   • Although aiding and abetting under Rosemond v. United States, 572 U.S. 65 (2014), classically requires the aider to act “with the intent of facilitating the offense’s commission,” the majority holds that for First Amendment purposes, the difference between specific intent and knowledge of the principal’s criminal intent is “a distinction without a difference” because the practical effect is the same: knowing facilitation.
3. Knowledge as the key limiter. The court heavily relies on the statute’s knowledge requirement as the feature that:
   • distinguishes § 842(p)(2)(B) from mere teaching of dangerous subjects to unknown audiences, and
   • keeps its reach mostly within scenarios where the speech is truly “integral” to crime.
   In both the majority’s concrete example (Arthur’s training of Buckshot) and its hypotheticals, it stresses simultaneity: the speaker must know of the recipient’s intent at the time of the teaching or distribution.
4. Rejection of Brandenburg‑style protection. Drawing on Miselis and Williams, the majority underscores that Brandenburg’s protection is confined to “abstract advocacy” not directed to, or likely to produce, imminent lawless action. Here, by contrast, the court sees individualized, operational assistance given to someone already resolved to commit a crime of violence.

Thus, instead of asking whether § 842(p)(2)(B) survives heightened scrutiny for content‑based restrictions, the majority concludes that the First Amendment simply does not apply to most, if not all, conduct the statute criminalizes.

D. Precedents Cited and Their Role

1. Supreme Court First Amendment and Overbreadth Cases

• United States v. Hansen, 599 U.S. 762 (2023): The majority borrows Hansen’s formulation of overbreadth (“realistic, not fanciful; substantially disproportionate”) and its insistence that courts prefer narrowing constructions over facial invalidation. It also notes Hansen’s discussion of the “speech integral to criminal conduct” category, including aiding and abetting, solicitation, and similar crimes.
• United States v. Williams, 553 U.S. 285 (2008): Williams upheld a statute banning offers to provide child pornography, distinguishing offers to engage in illegal transactions from protected mere advocacy. The majority uses Williams for its overbreadth structure and its distinction between abstract advocacy and unlawful solicitation/transactional speech.
• Virginia v. Hicks, 539 U.S. 113 (2003): Supports the high bar for overbreadth and the requirement that challengers show substantial, realistic overbreadth relative to legitimate applications.
• New York v. Ferber, 458 U.S. 747 (1982): Ferber’s instruction to consider narrowing constructions and severability is cited as part of the general overbreadth toolkit.
• United States v. Stevens, 559 U.S. 460 (2010): Recognizes “speech integral to criminal conduct” as one of the “well‑defined and narrowly limited classes of speech” historically unprotected.
• Brandenburg v. Ohio, 395 U.S. 444 (1969): Both majority and dissent recognize Brandenburg’s rule for advocacy‑of‑lawlessness, but the majority insists Brandenburg protects only “abstract advocacy,” not the operational assistance regulated by § 842(p)(2)(B).

2. Fourth Circuit First Amendment Cases

• United States v. Miselis, 972 F.3d 518 (4th Cir. 2020): In Miselis, the court partially invalidated the Anti‑Riot Act under Brandenburg but also recognized a category of unprotected “speech” comprising aiding and abetting and similar conduct. The majority leverages Miselis to classify § 842(p)(2)(B) as “far from” abstract advocacy and squarely within the aiding‑and‑abetting line.
• Rice v. Paladin Enterprises, 128 F.3d 233 (4th Cir. 1997): Rice is central. There, a publisher of a “Hit Man” manual was found civilly liable when it knowingly assisted a murder‑for‑hire. Rice canvassed the “speech integral to criminal conduct” doctrine and suggested (while reserving final judgment) that mere foreseeability of misuse might raise First Amendment questions, but that purposeful facilitation is unprotected. The majority treats § 842(p)(2)(B) as targeting precisely the kind of purposeful facilitation exemplified by Rice.

3. Other Circuit Precedents on Similar Statutes

• Nat’l Mobilization Comm. to End War in Viet Nam v. Foran, 411 F.2d 934 (7th Cir. 1969): Upheld 18 U.S.C. § 231(a)(1), which prohibits teaching the use of firearms or explosives for a civil disorder, emphasizing the narrowing role of the statute’s “knowing or having reason to know or intending” language. The majority cites Foran to support the idea that mens rea requirements narrow teaching‑statutes and can sustain them against facial challenges.
• United States v. Featherston, 461 F.2d 1119 (5th Cir. 1972): Upheld § 231(a) against a vagueness challenge. The majority acknowledges that Featherston is not directly on point (vagueness rather than overbreadth) but considers it broadly supportive of such instructional offenses.

4. Precedents on Mental State and Knowledge

• Flores‑Figueroa v. United States, 556 U.S. 647 (2009): Cited by the majority to show that proving knowledge can be difficult, undercutting the dissent’s notion that a knowledge standard is trivial to satisfy across the board.
• Model Penal Code § 2.02(7): Referenced for the definition of acting “knowingly” with respect to circumstances, reinforcing that § 842(p)(2)(B) demands awareness of the recipient’s criminal intent.

E. The Dissent’s Critique of Scope and Mens Rea

Chief Judge Gregory’s dissent raises a legally sophisticated and policy‑sensitive counterpoint.

1. Breadth of Definitions and “Pertaining to … in Whole or in Part”

The dissent carefully parses the statutory definitions incorporated into § 842(p)(2)(B):

• “Explosive” (via 18 U.S.C. § 844(j)): includes gunpowders, high explosives, blasting materials, detonators, and any chemical compound, mixture, or device that may cause an explosion if ignited or subjected to friction or concussion. Courts have interpreted this broadly (e.g., gasoline, methane).
• “Destructive device” (via 18 U.S.C. § 921(a)(4)): includes specific listed weapons but also “any combination of parts” designed or intended to convert something into a destructive device.

Combined with the phrase “information pertaining to, in whole or in part, the manufacture or use” of such items, the dissent contends that:

• Even a lecture on combustion physics or energy transfer might “in part” pertain to the manufacture or use of an explosive or destructive device.
• Basic engineering or chemistry education, or industrial safety manuals, may be swept in.

Thus, the dissent sees § 842(p)(2)(B) as a broadly phrased, content‑based prohibition on technical and scientific speech.

2. Knowledge vs. Specific Intent and “Willful Blindness”

Judge Gregory’s central normative concern is that § 842(p)(2)(B) requires only knowledge of the listener’s criminal purpose, not a specific intent by the speaker to commit or further the underlying crime.

He notes:

• Under general criminal law, knowledge can be proved through circumstantial evidence and “deliberate ignorance” or “willful blindness” doctrines (citing, e.g., Barnes v. United States, 412 U.S. 837; United States v. Ruhe, 191 F.3d 376), where:
  • a failure to investigate suspicious circumstances can suffice to “impute” knowledge.
• That low threshold, paired with expansive coverage of “information pertaining … in part” to explosives, risks imposing liability on:
  • a university professor lecturing on combustion or physics after receiving a concerning email from a student, or
  • a publisher issuing an explosives safety manual after being warned about a reader’s extremist sympathies.

Even if such prosecutions are improbable in practice, the dissent underscores that First Amendment analysis focuses heavily on chilling effect: the tendency of “ordinary citizens” to “steer wide of the unlawful zone” when the boundaries are unclear or when liability turns on others’ intent (Counterman v. Colorado, 600 U.S. 66 (2023)).

In the dissent’s view, it is constitutionally novel and dangerous to criminalize speech based on the listener’s intent, rather than the speaker’s own “calculated” intent to bring about unlawful conduct, which is required in many other speech‑crime contexts (Hansen).

3. The “Speech Integral to Criminal Conduct” Category

The dissent does not dispute that “speech integral to criminal conduct” is unprotected, but it contends that § 842(p)(2)(B) sits outside that traditional category:

• Historically, that category has applied where speech is the vehicle of the criminal act itself—extortion, fraud, perjury, conspiracy, solicitation, aiding and abetting—with the speaker having specific criminal intent.
• In those cases, the “underlying crime” is sharply defined (e.g., perjury at trial, extortionate demand, solicitation to commit a crime), and the speech “takes the place of” the non‑expressive act (e.g., the threat or false statement).
• By contrast, § 842(p)(2)(B) penalizes dissemination of truthful, generally available technical information (e.g., bomb‑making techniques, chemistry principles), absent any requirement that the speaker desire or intend the resulting violent act.

Judge Gregory points to Florida Star v. B.J.F., 491 U.S. 524 (1989), and related cases to stress judicial reluctance to punish publication of lawful, publicly available information merely because of how it might be used. He also relies on the Department of Justice’s own 1997 report acknowledging that bomb‑making information is widely accessible but that any regulation must “leave untouched” legitimate teaching and publication to remain constitutional.

Accordingly, the dissent concludes that the majority stretches the “speech integral to criminal conduct” category beyond its historical limits, effectively creating a new carve‑out for truthful technical instruction given in the presence of a listener’s criminal intent, even without the speaker’s specific intent.

4. Overbreadth and Chilling Effect

The dissent’s overbreadth analysis emphasizes:

• The statute’s coverage of “socially valuable speech” (technical education, industrial safety, academic research) that touches on explosives or destructive devices.
• The realistic risk that speakers, uncertain about audience intent or fearful of willful‑blindness findings, will self‑censor.
• The insufficiency of prosecutorial assurances that the law will be used “responsibly” (Stevens rejects reliance on “noblesse oblige”).

Where the majority sees realistic applications as predominately unprotected (integral to crime), the dissent sees a substantial universe of borderline situations where speakers will hesitate due to uncertainty about recipients’ future criminal uses, making overbreadth “truly warranted.”

F. Evaluating the Competing Approaches

The divide between majority and dissent in Arthur turns on three interrelated judgments:

1. How tightly does § 842(p)(2)(B) track “speech integral to criminal conduct”?
   The majority, viewing Arthur’s own conduct as paradigmatic, sees the statute as targeting close, direct assistance to violent crime, analogous to aiding and abetting. The dissent is more concerned with edge cases: non‑malicious instruction where the speaker suspects, but does not endorse, a recipient’s bad aims.
2. Is “knowledge of the listener’s intent” a sufficient culpable mental state?
   The majority treats knowledge + targeted technical instruction as enough; the dissent would require a “calculated purpose” to bring about unlawful acts (specific intent), especially where speech is truthful and generally available.
3. How does one balance realistic unconstitutional applications against the statute’s plainly legitimate sweep?
   The majority finds Arthur’s hypotheticals speculative and remote, and stresses congressional flexibility to criminalize speech when it is functionally tantamount to supplying weapons. The dissent, drawing on Counterman, is more sensitive to preemptive self‑censorship by cautious speakers—professors, publishers, trainers—who cannot reliably gauge their audience’s intentions.

As a practical matter, the majority’s opinion gives prosecutors a robust constitutional foundation to bring § 842(p)(2)(B) charges where they can prove:

• directed technical instruction about explosives or destructive devices, and
• contemporaneous knowledge that a recipient intends a federal “crime of violence.”

Future litigation may test how courts handle less obvious records—e.g., generalized online postings, ambiguous messages, or cases where “knowledge” is inferred via willful blindness rather than explicit statements like those in Arthur’s training session.

III. The Terrorism Enhancement under U.S.S.G. § 3A1.4

A. Legal Framework

U.S.S.G. § 3A1.4 provides a dramatic upward adjustment “[i]f the offense is a felony that involved, or was intended to promote, a federal crime of terrorism.” The term “federal crime of terrorism” incorporates 18 U.S.C. § 2332b(g)(5), which requires:

1. commission of one of a specified list of federal offenses (including § 1114—killing or attempting to kill federal officers or employees); and
2. that the offense be “calculated to influence or affect the conduct of government by intimidation or coercion, or to retaliate against government conduct.”

In United States v. Chandia, 514 F.3d 365 (4th Cir. 2008), the Fourth Circuit summarized these elements and held that the “calculated to influence or affect” requirement embeds a specific intent standard. In United States v. Kobito, 994 F.3d 696 (4th Cir. 2021), the court clarified the two alternative § 3A1.4 triggers:

• “Involved” prong: the predicate offense and related conduct must “include a federal crime of terrorism.”
• “Intended to promote” prong: covers situations where the defendant’s offense was committed with the goal or purpose of helping bring about a federal crime of terrorism, even if the defendant did not complete, attempt, or conspire to commit it (citing Mandhai and Graham from other circuits).

B. Application to Arthur’s Conduct

All parties agreed that the crime Arthur knew Buckshot intended—murder or attempted murder of federal law enforcement officers under § 1114—qualifies as a “federal crime of terrorism” when accompanied by the requisite intent to coerce or retaliate against government. The dispute centered on whether Arthur’s own § 842(p)(2)(B) offense:

• “involved” that terrorism offense, or
• was “intended to promote” it.

The Fourth Circuit endorsed the district court’s findings under both prongs:

1. “Involved” prong.
   Because § 842(p)(2)(B) by definition requires that the defendant know the recipient intends to use the information to commit a federal crime of violence, the majority reasons that:
   • Arthur’s crime “necessarily involved” the contemplated § 1114 offense—his instruction and his knowledge of Buckshot’s intent were elements of his § 842(p)(2)(B) violation.
   • His teaching was “integral” to that violent plan: he provided the operational means (IED designs, tactics) by which Buckshot could attack federal officers.
2. “Intended to promote” prong.
   The district court found, and the majority accepts, that:
   • Arthur “help[ed] bring into being” a § 1114 offense by “encouraging and contributing to the murder or attempted murder of federal law enforcement officers.”
   • He did so for the specific purpose of retaliating against and coercing government conduct—fitting § 2332b(g)(5)’s intent standard.
   That suffices under Kobito, Mandhai, and Graham, which hold that direct participation in the terrorist act is not required; purposeful promotion is enough.

Arthur’s argument that his offense did not “include” a completed federal crime of terrorism misreads Kobito’s use of “include.” It is enough that his offense, by its elements and factual context, incorporated knowledge and promotion of a qualifying terrorism offense.

C. Harmless Error and Alternative Variant Sentencing

Anticipating scrutiny of the enhancement, the district court stated explicitly that it would impose the same 300‑month sentence even if it had miscalculated the Guidelines. Under United States v. Mills, 917 F.3d 324 (4th Cir. 2019), a Guidelines error is harmless if:

1. the district court would have reached the same sentence even assuming the error, and
2. that sentence would be substantively reasonable under the correct range.

The Fourth Circuit finds both conditions satisfied:

• The court’s record shows a detailed, individualized § 3553(a) analysis—emphasizing the seriousness of training someone to kill federal agents, Arthur’s lack of remorse, and the need for deterrence—supporting the 300‑month term as an “alternative variant” sentence.
• Even from the lower agreed‑upon range of 168–210 months (without § 3A1.4), an upward variance to 300 months is not an abuse of discretion given the gravity and premeditation of Arthur’s conduct.

Thus, even if a reviewing court later disagreed with the legal interpretation of § 3A1.4, the sentence would stand.

D. Sentencing Significance

The decision confirms that § 3A1.4 can apply not only to direct perpetrators of terrorist acts but also to those whose offenses substantially facilitate or promote such acts, including instructional offenses like § 842(p)(2)(B). Prosecutors can argue that:

• where a defendant’s offense incorporates knowledge of a planned § 2332b(g)(5) qualifying crime and provides the means to carry it out, the “involved” prong is satisfied; and
• a similar record of purposeful encouragement and operational support suffices for the “intended to promote” prong, even if no terrorism offense is actually completed.

District courts, in turn, may insulate their sentences from Guidelines challenges by explicitly announcing alternative variant rationales under § 3553(a), as this court did.

IV. Key Legal Concepts Simplified

1. Facial Overbreadth vs. As‑Applied Challenges

• As‑applied challenge: argues that a statute is unconstitutional as used on the challenger’s specific facts. The court need not decide its validity in other scenarios.
• Facial overbreadth challenge: in the First Amendment context, allows a litigant to argue that a law punishes so much protected speech, relative to its legitimate applications, that it cannot be enforced against anyone, even those clearly within its core. This is an exception to ordinary standing and facial‑challenge rules.
• Overbreadth is reserved for situations where unconstitutional applications are “realistic” and “substantially disproportionate” to the law’s legitimate reach.

2. “Speech Integral to Criminal Conduct”

This unprotected category covers speech that is itself part of the crime:

• extortion demands,
• fraudulent misrepresentations,
• perjured testimony,
• solicitation and conspiracy communications,
• aiding and abetting instructions given with criminal purpose.

The logic: free speech principles do not extend to words used as the mechanism of a crime. In Arthur, the majority places § 842(p)(2)(B) in this category; the dissent argues that expanding the category to truthful technical instruction given with mere knowledge (not intent) is a step too far.

3. Brandenburg Incitement vs. Operational Facilitation

• Brandenburg rule: The state may not forbid advocacy of the use of force or law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to produce such action.
• Abstract advocacy: General statements (“The government is tyrannical, we should resist violently someday”) are protected.
• Operational facilitation: Step‑by‑step instructions tailored to a specific person already committed to a crime (e.g., “Here’s exactly how to rig IEDs around your property to kill federal agents”) can be treated as “speech integral to criminal conduct” and thus outside Brandenburg’s protections, according to the majority.

4. Mens Rea: Knowledge vs. Specific Intent

• Specific intent: The defendant has as his purpose the commission or facilitation of a particular crime.
• Knowledge: The defendant is aware that certain facts exist (e.g., that another person intends to commit a crime), but may not share that purpose.

In many traditional “speech crimes” (solicitation, conspiracy, aiding and abetting), the law requires specific intent. Arthur stands for the proposition (at least in the Fourth Circuit) that for First Amendment purposes, knowledge of another’s criminal intent, combined with direct operational assistance, can be enough to classify speech as unprotected “integral to criminal conduct.”

5. Federal Crime of Terrorism and § 3A1.4

• A “federal crime of terrorism” under § 2332b(g)(5) is:
  • an offense listed in § 2332b(g)(5)(B) (e.g., § 1114 killings of federal officers),
  • committed with a calculation to influence or retaliate against government conduct.
• Section 3A1.4 dramatically increases the offense level and criminal history category if the offense:
  • involved a federal crime of terrorism (i.e., includes or incorporates such a crime), or
  • was intended to promote such a crime (i.e., committed with a purpose of bringing it about).

In Arthur, the court treats the § 842(p)(2)(B) offense as both involving and intended to promote the recipient’s planned § 1114 offense.

V. Potential Impact and Future Directions

A. National Security and Online Instructional Speech

Arthur strengthens the government’s hand in prosecuting those who provide technical guidance on explosives or weapons to particular individuals known to intend violent crimes. Practical implications include:

• Undercover stings and targeted investigations: Law enforcement may more confidently design operations where undercover agents request tailored explosives instruction, then prosecute under § 842(p)(2)(B) if explicit evidence of knowledge of violent plans is elicited.
• Private trainers and “militia” schools: Instructors offering training against “tyrannical government” or law enforcement may face heightened risk if they customize tactics for students who articulate concrete violent plans. The knowledge requirement now has a published construction aligning such cases with unprotected speech.

B. Academic, Technical, and Publishing Contexts

The dissent’s concerns raise real questions about chilling in scientific and technical communities:

• Professors in physics, chemistry, engineering, and security studies regularly teach information that “pertains, in whole or in part,” to explosives and destructive devices. If a student signals extremist views or intentions, the professor might fear that continuing to teach the general curriculum could later be argued to meet the knowledge prong.
• Publishers of engineering texts or military manuals might hesitate to respond to readers expressing dubious intentions, or may adopt screening practices for direct, bespoke advice, to avoid knowledge‑based liability.

The majority’s insistence on contemporaneous knowledge tied to directed teaching moderates, but does not eliminate, this chill. Its narrowing gloss (that § 842(p)(2)(B) is naturally read as covering instruction directed “to any person,” not general public dissemination) may prove important in future cases involving online posting or mass distribution.

C. Relationship Between § 842(p)(2)(A) and (B)

A notable, if subtle, point in the majority’s opinion is its later discussion of § 842(p)(2)(A), which criminalizes similar instruction given with the intent that the information be used for a federal crime of violence, without the “to any person” / “such person” language.

• Subsection (A) appears more suited to broad, undirected dissemination where the speaker intends their published instructions to be used for violent crimes generally.
• Subsection (B), by contrast, is read by the court as focusing on targeted instruction to identifiable persons, requiring knowledge of a recipient’s intent.

This division of labor may guide prosecutors’ charging decisions and courts’ future readings, helping cabin the reach of (B) and possibly mitigating some overbreadth concerns in the online context.

D. Prospect of Further Review

Because the dissent squarely contests the majority’s expansion of the “speech integral to criminal conduct” category and raises serious chilling‑effect concerns, Arthur is a plausible candidate for further appellate review. Issues likely to attract scrutiny include:

• whether knowledge, without specific intent, can be a sufficient mental state to remove speech from First Amendment protection when that speech consists of truthful, publicly available technical information;
• how far Congress may go in criminalizing content‑based teaching statutes before sweeping in too much protected academic and technical speech; and
• the boundary between Brandenburg‑protected advocacy and unprotected “integral” facilitative instruction.

At a minimum, Arthur provides a carefully reasoned, deeply divided appellate treatment of these questions that will likely be cited in future First Amendment and national security cases.

VI. Conclusion

United States v. Arthur is a significant Fourth Circuit decision at the intersection of free speech, national security, and criminal liability for technical instruction. The majority establishes two important propositions:

• Teaching or distributing bomb‑making or explosive‑use information to a person, while knowing that person intends to commit a federal crime of violence, is largely “speech integral to criminal conduct” and lies outside the protection of the First Amendment; § 842(p)(2)(B) is therefore not facially overbroad.
• Such an instructional offense can both “involve” and be “intended to promote” a federal crime of terrorism for purposes of the § 3A1.4 terrorism enhancement, dramatically increasing sentencing exposure.

Chief Judge Gregory’s dissent, however, underscores the cost of this approach: broad statutory definitions and a relatively low mens rea threshold may chill socially valuable scientific, educational, and technical speech, especially when liability hinges on the audience’s intent rather than the speaker’s own purpose.

Future courts will need to navigate these tensions carefully, especially as technology further democratizes access to dangerous knowledge. Arthur marks a clear doctrinal shift toward treating directed technical assistance to known would‑be violent offenders as categorically unprotected speech, even where the speaker does not personally desire the crime to occur. Its lasting significance will lie in how broadly that concept is applied beyond the stark facts of this case.