The relationship between political violence and political freedom has always been uneasy. On the one hand, movements such as anarchism have promoted the idea of “propaganda of the deed,” the notion that dramatic acts—sometimes violent—can advance revolutionary change more effectively than words alone. Violent action threatens the peace and stability of any political order (this is the intent of violent political actors). On the other hand, governments, in their attempts to suppress domestic violence and secure order, often reach for tools that risk curtailing freedom of thought and expression. Efforts to monitor or preempt radicalism can easily slip into forms of thought control, where dissent itself becomes suspect.
This essay brings these two themes together to show the parallel dangers they represent. The first is the danger of political movements that cross from persuasion to coercion, from “propaganda of the word” to “propaganda of the deed.” The government has several coercive tools to deal with this. (Indeed, as I show in Our Constitution and the Federal Authority to Quell Rebellion (see also Concerning the Powers of The US Constitution—And Those Defying Them), the document that established the American Republic was written to create a strong Executive and give Congress the tools to suppress rebellion and insurrection). The second is the danger of state responses, for example, President Trump’s September 25 National Security Presidential Memorandum 7 (NSPM-7), Countering Domestic Terrorism and Organized Political Violence, that go beyond punishing violent acts to policing associations and beliefs. In both cases, the balance between liberty and order is at risk.
To condemn political violence without also scrutinizing the overreach of state power is to ignore half the problem. As much as I have warned about the problem of domestic violence in America (see, e.g., The Fool Has Come Down Off The Hill. But Who Called on Antifa to Terrorize the Village?), I have also warned about the problem of thought control, whether coming from government or organizations of private power (Fundamental Law Regarding Freedom of Thought and Conscience; Natural Rights, Government, and the Foundations of Human Freedom; Ketanji Brown-Jackson Concerned First Amendment “Hamstrings” State).
I begin with the problem of leftwing excesses. As I have explained in previous essays (see e.g., Tesla and Propaganda of the Deed), the concepts of propaganda of the word and propaganda of the deed come out of nineteenth-century anarchist politics. Anarchists, who reject the authority of the state and law and seek a stateless order rooted in voluntary cooperation and direct democracy, debate how best to spread their ideas—or whether both forms of propaganda should be used in tandem.
Some anarchists believe that persuasion is the most effective path: essays, pamphlets, speeches, and street agitation to win people over to anarchism. This approach is called “propaganda of the word,” rooted in the belief that truth and justice, once clearly communicated, can persuade people to withdraw their consent from oppressive institutions. Others argue that words are not enough. They maintain that dramatic action can inspire change more powerfully than abstract reasoning. This view gives rise to the notion of “propaganda of the deed,” the idea that bold acts—demonstrations, strikes, or even violent action—can serve as a spark to awaken the masses or strike fear in the ruling class. Here, action is a form of communication, carrying symbolic weight meant to shape public opinion or alter political conditions.
Propaganda of the deed can take many forms, some of which are criminal. Peaceful protests and strikes are examples of legitimate actions undertaken to make change. However, direct confrontation with authority or open rebellion, as in uprisings against governments or local officials, is not legitimate from the perspective of even the freest state. Sometimes such action reaches the level of insurrection, where organized groups seek to overthrow the state entirely. In other cases, it takes the form of targeted violence, especially political assassination. Article I, Section 8, Article II, and Article IV authorize the federal government to use military force to quell domestic violence.
Trump’s September 22 Executive Order designating Antifa a domestic terrorist organization is an appropriate step. Some anarchists (and other radicals) justify assassinations as a way to decapitate what they deem oppressive movements or regimes. Here, assassination carries a broader logic: to frighten others from stepping into positions of leadership, thus spreading fear beyond the immediate victim. Such assassinations also fall within the definition of terrorism because they seek to intimidate a larger audience through a symbolic act of violence. Concern for both propaganda of the word and propaganda of the deed is reasonable given that both are bound together by the anarchist conviction that existing structures of power are illegitimate, and that the role of the activist is to awaken society to new forms of life beyond the state or to move others to join them in overthrowing the existing government. However, one must take care to distinguish between opinion (belief), on the one can, and action on the other.
Trump’s NSPM-7, following his order designating Antifa as a terrorist organization, gives sweeping instructions to federal agencies—including Homeland Security, the Justice Department, the Internal Revenue Service, and the Treasury—to adopt a more aggressive, coordinated approach to political violence in the United States. At its core, the memorandum frames terrorism not only as acts of violence but as the culmination of a process beginning with intimidation, radicalization, recruitment, and financial support. It urges officials to intervene earlier, even at the level of “indicia” of potential violence. The question is not whether early intervention is a legitimate government action per se, but what falls within its scope.
“There are common recurrent motivations and indicia uniting this pattern of violent and terroristic activities under the umbrella of self-described ‘anti-fascism.’ These movements portray foundational American principles (e.g., support for law enforcement and border control) as ‘fascist’ to justify and encourage acts of violent revolution.” This is true. So too is this: “This ‘anti-fascist’ lie has become the organizing rallying cry used by domestic terrorists to wage a violent assault against democratic institutions, constitutional rights, and fundamental American liberties.”
The memorandum continues: “Common threads animating this violent conduct include anti-Americanism, anti-capitalism, and anti-Christianity; support for the overthrow of the United States Government; extremism on migration, race, and gender; and hostility towards those who hold traditional American views on family, religion, and morality. As described in the Order of September 22, 2025 (Designating Antifa as a Domestic Terrorist Organization), the groups and entities that perpetuate this extremism have created a movement that embraces and elevates violence to achieve policy outcomes, including justifying additional assassinations. For example, Charlie Kirk’s alleged assassin engraved the bullets used in the murder with so-called ‘anti-fascist’ rhetoric.”
I am concerned about the situation with escalating political violence in America, as previous essays on this platform attest. At the same time, I am concerned about what feels like the specter of thought control in addressing the problem. The First Amendment is precious, and while monitoring associational and financial ties to domestic terrorist organizations may be warranted (with guardrails), identifying “anti-American,” “anti-capitalist,” and “anti-Christian” opinions as precursors to political violence—and therefore subject to government monitoring and surveillance—is not only unwarranted but prohibited by the First Amendment. Readers of Freedom and Reason know that, in the past, I have been critical of capitalism and Christianity. I reserve the right, as a citizen of a free republic, to hold those opinions and change them based on reason and evidence—not out of fear that my views will trigger government surveillance. That is not the country I want to live in.
NSPM-7 suggests government action that goes beyond what is permissible in a free republic: “The United States requires a national strategy to investigate and disrupt networks, entities, and organizations that foment political violence so that law enforcement can intervene in criminal conspiracies before they result in violent political acts. Through this comprehensive strategy, law enforcement will disband and uproot networks, entities, and organizations that promote organized violence, violent intimidation, conspiracies against rights, and other efforts to disrupt the functioning of a democratic society.”
As long as the government focuses on entities, networks, and organizations that incite or promote organized violence, again, within the scope of the law, the memorandum does not run afoul of the First Amendment (though “promote” does not rise to the level of “incite”). However, enlarging the scope of federal action against domestic terrorism—whether Antifa, Islamism, or violent transgenderism—risks intruding not only on free speech rights but also on individuals’ right to privacy. The government is not entitled to know our opinions, and if it does know what we believe (and I am vocal about mine, as are many others), acting in ways that suppress those opinions—or even create a chilling effect—veers into thought control.
Civil libertarians, among whom I count myself, are troubled by language that is overbroad. By defining opposition to capitalism or Christianity as “indicia” of terrorism, the government risks equating protected political expression with criminal suspicion. Thus, Trump’s NSPM-7 blurs the line between legitimate surveillance of violent networks and the monitoring of nonviolent dissent. In effect, NSPM-7 could chill free speech, deter lawful protest, and expose activist organizations to investigation or financial scrutiny. Some civil libertarians are particularly concerned about the directive instructing the IRS to ensure that tax-exempt groups are not “indirectly financing political violence.”
Defenders of NSPM-7 argue that radical ideologies have historically been precursors to violence and that the government must act early to prevent terrorism before it occurs. As suggested above, I understand this reasoning, but there is a danger we must avoid: a kind of “pre-crime” logic, where holding unpopular beliefs becomes a trigger for government scrutiny. The history of counter-subversive efforts in the United States shows how quickly expansive counterterrorism powers can be turned against legitimate political opposition.
For me, the problem with NSPM-7 is less about its stated goal of preventing violence and more about the means it employs. By tying beliefs like anti-capitalism or anti-Christianity to terrorism indicators, the memorandum erodes the boundary between protected dissent and violent acts. Whether enforced expansively or narrowly, Trump’s memorandum reminds the nation of the enduring tension in American legal culture and politics between the First Amendment and national security—the ever-present risk that efforts to suppress violence may also suppress dissent. For this reason, such memoranda should be written with precision to leave no doubt.
Trump’s NSPM-7 is not the first time the US government has issued directives or memoranda that blur the line between criminality and ideology. During the Red Scare of the 1920s and 1950s, federal and state agencies monitored and prosecuted individuals for alleged communist sympathies, often targeting people solely for their associations and beliefs. Similarly, COINTELPRO in the 1960s and 1970s authorized the FBI to surveil, infiltrate, and disrupt activist organizations, including civil rights and anti-war groups, based on ideological grounds rather than criminal conduct (see The Black Panthers: Black Radicalism and the New Left; A Note on Desegregation and the Cold War). More recently, post-9/11 counterterrorism policies expanded surveillance powers to monitor Muslim and other perceived extremist groups, often relying on broad indicators rather than actual criminal activity.
While surveilling associations is justified in certain cases, targeting citizens for their beliefs—short of inciting violence—is a serious concern. When the government seeks to prevent violence, it can slip into regulating political expression and policing thought—something no government should be trusted to do. Indeed, this is why the Founders enshrined the First Amendment. From my perspective, NSPM-7 represents a risk to our civil liberties, creating the potential for the state to criminalize opinions rather than actions. By treating beliefs such as anti-capitalism, anti-Christianity, or anti-Americanism as indicators of terrorism, the directive effectively targets individuals for what they think, not what they do.
Moreover, if NSPM-7 does not specifically clarify Trump’s September 22 order, it focuses too narrowly on antifascist activities. Those concerned with right-wing political violence might reasonably ask why comparable motivations and indicia are omitted—though it might feel unusual to identify pro-American, pro-capitalist, and pro-Christian sentiments as precursors to terrorism. At any rate, this raises concerns about viewpoint discrimination, running directly contrary to the spirit of the First Amendment, which protects dissenting and unpopular ideas precisely because they challenge dominant power structures. One could argue, therefore, that the memorandum does not merely risk viewpoint discrimination but enshrines it into policy, empowering the state to monitor, investigate, and potentially punish individuals for selected ideological positions—a dangerous precedent for civil liberties in a democratic society.
I would feel more confident about NSPM-7 if Trump issued a new or modified version that explicitly ensures American citizens will not be subjected to thought control. Not only because of the concerns I have outlined in this essay, but also because of the boomerang effect. I can imagine the next progressive administration (hopefully there will never be one, but there could be) surveilling me for my opinions regarding gender identity ideology. I can imagine extremists committing violence against those advocating that ideology, while critics like me—being surveilled by the government for our opinions—could be swept up in a government effort to suppress domestic violence against groups advocating genderism. Not that goodwill toward progressives would necessarily prevent them from pursuing totalitarian ends, but at least I would want to have a principle to stand on.
Trump’s memorandum calls to mind Thomas Jefferson, who, in an 1802 letter to the Danbury Baptist Association in Connecticut, emphasized that the legitimate powers of government extend to actions, not opinions. Put another way: the government can regulate people’s outward behavior when it affects others (actions), but it has no rightful authority over what people think or believe (opinions). We must draw a clear distinction between controlling domestic violence, on the one hand, and thought control, on the other. One is free to express any opinion short of inciting violence or demonstrating a clear intention to do violence. If somebody adopts an opinion and acts violently on it, that is not on you. People are responsible for their actions, not their opinions. This does not mean that opinions are unrelated to violence carried out on their basis (which is why we criticize opinions), but it is a matter of lawful culpability and government overreach.
In short, what makes Trump’s NSPM-7 problematic is not the control function of the federal government, but its treatment of ideological markers. As the reader can see, the memorandum lists several beliefs that fall under the broad category of unpatriotic views, from a certain political-ideological standpoint. Though the directive does not declare such views illegal in themselves, it explicitly links them to terrorism indicators, creating a framework where dissenting ideologies may be treated as early warning signs of violence. This risks running afoul of the First Amendment if it interferes with free expression, which I worry that it will.
* * *
Before leaving this essay, I want to emphasize the problem of Antifa and the concerted effort by some in law enforcement at the highest level to obscure the organized character of Antifa. For example, in a 2020 congressional hearing, then-FBI Director Christopher Wray testified that Antifa is not an organization or structured group, but rather an ideology or movement. Wray emphasized this point, stating plainly: “It’s not a group or an organization. It’s a movement or an ideology.” (See Antifa, the Proud Boys, and the Relative Scale of Violent Extremism; Charlie Kirk’s Killer is in Custody and the Specter of Antifa.)
This statement represents a fundamental mischaracterization of the nature of Antifa, and it raises serious questions about either Wray’s competence or his willingness to speak truthfully. Either he knowingly rationalized a false narrative, or he was so ideologically captured that he neglected his responsibility to accurately inform Congress. This characterization was subsequently echoed by President Joe Biden during the first presidential debate with Donald Trump, embedding the claim in public discourse (Buried Lede: Biden Fails to Condemn Antifa at First Presidential Debate).
The troubling reality is that Wray’s framing of Antifa has been accepted uncritically by much of the mainstream media, despite evidence to the contrary. So pervasive is this narrative that it has even seeped into artificial intelligence systems, with platforms such as ChatGPT repeating the same language without scrutiny. When challenged, AI often mirrors this oversimplified framing. An exchange with ChatGPT this morning revealed that it resisted acknowledging the structural reality of Antifa—until confronted with clear analogies to recognized terrorist networks.
I induced the AI bot to provide these examples: Al-Qaeda operates as a loose network of semi-autonomous cells, decentralized leadership, and no formal public membership. ISIS functions through regional “wilayats” with significant operational autonomy, connected by shared ideology rather than a formal chain of command. The Weather Underground historically operated in autonomous cells without centralized rosters to prevent infiltration. It confessed that in none of these cases does decentralization mean the absence of organization. Indeed, it admitted that decentralization is often a deliberate tactic to enhance operational security. It should concern the public that, unless one takes the time and has the know-how to confront AI with its contradictions, its chatbots will regurgitate corporate state propaganda, obscuring the threats to America’s internal security.
Christopher Wray knew all this. How could any FBI director not know this? Many violent movements deliberately obscure their structure to avoid infiltration and ensure operational efficacy. In the case of Antifa, there is ample evidence that collectives operate with internal vetting processes, codes of conduct, and membership mechanisms designed to maintain security. These features are hallmarks of an organized movement. While there may not be a formalized, public hierarchy, Antifa’s structure is real: it consists of chapters, cells, and networks operating nationally and transnationally, united by shared ideology and tactics. Verified instances of arson, assaults, property destruction, and organized confrontation with civilians and police attest to its coordinated activity.
The reality is that Antifa functions as an organization with a hierarchical structure, exclusive membership, and vetting procedures. It is a transnational movement engaged in violent action. Decentralization and secrecy may make providing systematic proof of hierarchy difficult to establish, especially when you are trying to avoid proving the case, but they do not negate the existence of an organization. In fact, they strengthen the claim that Antifa operates deliberately to conceal its structure—precisely because it is organized.
The question then becomes: why did Wray publicly deny this reality in 2020? Was it incompetence, negligence, or a deliberate decision to obscure facts? Wray himself confirmed in Congress that the FBI was investigating individuals identifying as Antifa involved in terrorist acts. Yet he declined to examine the larger structure or implications. This selective acknowledgment suggests not ignorance but deliberate avoidance of the question. This makes Wray an ally in law enforcement to a terrorist organization.
How is an FBI director using his authority to obscure the reality of a domestic terrorist organization allowed to stand? By characterizing Antifa as “not an organization,” Wray effectively diminished the public’s perception of Antifa’s organized capacity for coordinated violence. This mischaracterization undermines accountability and hampers efforts to address politically motivated violence. Documented incidents of assaults on political opponents, destruction of property, and armed confrontations indicate that Antifa’s activities extend beyond spontaneous protest and enter the realm of coordinated extremism. Haul Wray before Congress to interrogate him about his 2020 testimony and other public comments.
If Antifa is indeed a decentralized yet organized entity engaging in systematic violence, then the failure to label it accurately has profound implications for domestic security, public discourse, and democratic transparency. This is not merely an academic disagreement over semantics—it is a question of how America identifies and responds to threats within its own borders. Antifa is a threat to domestic peace and tranquility. The Trump Administration is right to act and I applaud Trump’s September 22 Order to designate Antifa as a domestic terrorist organization.
freedom and reason 