Queer Theory, Political Economy, and Life Beyond the State
17 Jul 2024 - Otto Vogel
CW: Discussion of school shootings, anti-muslim violence, anti-LGBTQ+ hate, and political violence.
I came across an essay written by Ben Burgis in Jacobin today, titled “On Stochastic Terrorism and Speech as Violence”.1 I think this essay reveals a deep lack of understanding of the framework of “stochastic terrorism”, why it is important, and why there is a surface-level appearance of it being applied “inconsistently”.
Please note, I am not an expert on this topic; I am just an antifascist researcher who finds the concept of “stochastic terrorism” highly useful.
The basic claim of Burgis’ argument is that
In effect, conservatives like Vance are appropriating the idea, long put forward by some liberals, that overheated political rhetoric is itself a form of violence. The theory of “stochastic terrorism” holds that over-the-top rhetoric about a targeted individual or group has the effect of encouraging “lone-wolf” political violence — that is to say, political violence carried out by individuals on their own initiative rather than terrorist organizations — and that this makes the purveyors of the rhetoric responsible for the violence.
Notice here that the theory of “stochastic terrorism” is a distinctly moral or legal one. It connects the actions of one individual to the actions of another individual — the violence person $X$ does is attributed to person $Y$ via the fact person $Y$ purveyed “overheated rhetoric” about the targets of $X$’s violence. Burgis then moves on to claiming that this concept is suspect because nobody uses it consistently:
I’ve never seen anyone consistently apply this theory. Invariably, overheated rhetoric that one finds unfair or irresponsible is blamed for violence against the targets of that rhetoric. Equally white-hot rhetoric that one agrees with, or that is put forward by factions to which one is basically sympathetic, is given a more nuanced treatment, and (appropriate) skepticism is applied to the chain of causation.
So, there is an asymmetry here that we should be suspect of. Stochastic terrorism is used as a linkage between speech of people we disagree with and violence against targets of that rhetoric, but not for groups we agree with. Finally, Burgis argues that we shouldn’t use this framework because it threatens free speech.
In all cases, I’d argue that the “stochastic terrorism” theory dangerously undermines free speech norms by blurring the line between speech and violence. Let’s not go down that road.
The overall picture is that “stochastic terrorism” is a conceptual tool used to attack speech that you disagree with, that attaches the State’s (legitimate?) interest in preventing violence to an (illegitimate?) interest in repressing certain forms of political speech. It is something tied up to, and existing in the shadow of the State, it’s something that directs State interest. To be clear, this is the most reasonable version of this argument. If Burgis is simply saying that we shouldn’t hold personal beliefs that certain forms of speech encourage violence, then that is absurd. Burgis would then be claiming that we should throw out a potentially true theory because it has morally suspect implications.2 We will discuss this below.
The issue is that State prosecution of speech is not what stochastic terrorism is about. This appears, to me, to be the result of some people trying to use stochastic terrorism as a legal argument, in order to legally hold certain people responsible for their speech. This is an issue, however, because stochastic terrorism specifically describes a situation where responsibility is disavowed.
What is stochastic terrorism, then? Stochastic terrorism describes the construction of a media environment where terroristic violence against vulnerable groups is encouraged, condoned and facilitated by figures who have only tenuous relationships to the perpetrators of that violence. That’s a bit abstract, so a metaphor might help.
Imagine you’re cooking popcorn over a fire. The fire is too small to make enough heat for the popcorn to pop, so you and your friends start adding more logs to the fire, stoking the fire, trying to heat up the pot. Eventually, the temperature will get high enough that popcorn starts going pop, pop, pop. Now, can we say that some specific person, or some specific log made the popcorn pop? Not really; instead, each log added fuel that let the fire get hotter. There’s no clean line of causation there. Hell, maybe none of your friends even touched the popcorn. Stochastic terrorism is similar. People “increase the temperature” of political discourse, adding “logs to the fire” by saying things about how action should be taken, about how vile and monsterous their opponents are, until something goes pop. Except, it’s not popcorn. It’s a gun, or a body being crushed by a car.
This sets out a few things that “stochastic terrorism” is not:
Let’s take a look at the example that fully convinced me of the validity of the notion of stochastic terrorism. The 2017 Quebec City Mosque Shooting.
What motivated the attack? According to Amanda Coletta in “Quebec City mosque shooter scoured Twitter for Trump, right-wing figures before attack”:
Bissonnette also appears to have obsessively visited the Twitter accounts of Tucker Carlson and Laura Ingraham, Fox News personalities; David Duke, the former leader of the Ku Klux Klan; Alex Jones of Infowars; conspiracy theorist Mike Cernovich; Richard Spencer, the white nationalist; and senior White House adviser Kellyanne Conway. Bissonnette checked in on the Twitter account of Ben Shapiro, editor in chief of the conservative news site the Daily Wire, 93 times in the month leading up to the shooting.
The court heard that Bissonnette confessed to a social worker several months after the attack that he idolized mass shooters and wished that his own attack had been more deadly. The contents of his laptop reveal that he spent hours looking up mass killers like Dylann Roof, the white supremacist who killed nine churchgoers at the Emanuel African Methodist Episcopal Church in Charleston, S.C., and Marc Lépine, a misogynist who massacred 14 women at Montreal’s Polytechnique engineering school in 1989, according to CTVNews.ca.
[…]
In a video of his police interrogation shown to the court last week, Bissonnette is heard telling officers that his three-minute-long attack was set off by Prime Minister Justin Trudeau’s message of welcome to refugees in the wake of President Trump’s entry ban, which was issued two days before the mosque attack. As chaos spread across airports around the world, Trudeau tweeted: “To those fleeing persecution, terror & war, Canadians will welcome you, regardless of your faith. Diversity is our strength #WelcometoCanada.”
The attacker was clearly motivated by a pervasive anti-immigrant sentiment, something that we know that Tucker Carlson, Laura Ingraham, David Duke, Alex Jones, Mike Cernovich, Richard Spencer, Kelleyanne Conway and Ben Shapiro all engaged in. But that wasn’t enough, we could say. The attacker also idolized other terrorists, like the Ecole Polytechnique attacker and the Charleston Church shooter. Then, he was “set off” by a tweet from Trudeau that welcomed immigrants. If we are positioning this entire thing within the context of a variety of speech acts, then Trudeau’s speech act of tweeting out a welcome message is one of the acts. It is part of the stochastic terroristic network. This isn’t just because it was the instigating moment, but because statements like that are used as ammunition to claim that there is a “great replacement” going on, one that must be stopped through acts of individual violence against immigrants.
But notice how none of those figures can be held to account for what they said. None of Tucker Carlson, Laura Ingraham, David Duke, Alex Jones, Mike Cernovich, Richard Spencer, Kelleyanne Conway or Ben Shapiro were held to account for their anti-immigrant sentiment, none were sued for instigating this attack. Why? Because there wasn’t enough of a case. The connection between speech and violence — which I think is undeniable in this case — isn’t enough to infringe on these figure’s speech rights. And yet, this is one of the textbook examples of stochastic terrorism in my book.
Note how Burgis says that stochastic terrorism involves a targeted “individual or group”. Let’s focus on the individual side of that question. We will catch Burgis in a dilemma. Either, “stochastic terrorism” doesn’t apply to individuals — at which point I don’t think the inconsistency claim applies, unless Burgis wants to provide some examples — or Burgis is against the prosecution of Alex Jones for how he targeted harassment against named individuals in his conspiracism about the Sandy Hook massacre. Now, Burgis has said that Jones loosing his court case was “a good thing. Jones made wildly inflammatory accusations against grieving parents without a shred of evidence — and he did it in order to draw viewers, listeners, and clicks to his media business Infowars and related profit-making ventures. He’s a despicable person.”
But where the question of the court case hinged on defamation, the issue of damages is where we will focus. Jones has to pay over a billion dollars in damages. This isn’t something that is down to the individual speech acts Jones engaged in. What Burgis glosses as “defamation” was tied up in the behaviour of Jones’ viewers. The way that Jones caused so much harm to the victims is by encouraging his audience to harass the parents. Wouldn’t this exactly line up with Burgis’ discussion of stochastic terrorism?
Why is this important? Well, Burgis could say that “well, stochastic terrorism doesn’t apply to individuals.” But then, is the framework still inconsistent? Have there been any examples of left-wing stochastic terrorism, where a variety of left-wing figures had “overheated rhetoric” that led to attacks against groups?
Now as for my thoughts? I don’t think Jones engaged in stochastic terrorism here. I think that an integral part of stochastic terrorism is the question of it being aimed at vulnerable groups. The Quebec City Mosque shooting and the various anti-LGBTQ violence we are currently dealing with are both much larger as well. There’s no single person we can really say is instigating that, directing that. Instead there is a diffuse network of actors, some of them public figures, some of them anonymous posters on fascist messageboards. Maybe we can say stochastic terrorism can target public figures, but that would only be because of the fact that public figures have a different standard of defamation than private individuals, and the different position they have relative to public discussion and the symbolic value we invest in them. But I am more than comfortable to say that stochastic terrorism doesn’t apply to individuals. I am less concerned with political violence that gets publicized and spoken about endlessly — which attacks against public figures will necessarily be — than about mass shootings, bomb threats against hospitals, and generally the encouragement of terrorist violence against vulnerable people.
Finally, I am of the opinion that nobody uses the term “rationality” in political discussions in a fully consistent way. Would Burgis be comfortable letting go of that framework and stop using the term “rationality” or associated terms like “logical”, when discussing politics? I don’t think he would. The question is not if people use the term in an inconsistent manner, is it? The question is if the term is useful or not, if it helps us understand the political realm better. The question of inconsistency of the term “stochastic terrorism” isn’t a good question, because inconsistency of a term — be it socialism, communism, democracy, legality, criminality — doesn’t actually exclude it from discussion. It just means we have to be more careful of using it.
A big part of what I have said above is that “stochastic terrorism” describes a situation where legal responsibility is disavowed, where it becomes impossible for the State to prosecute the speaker. If this is the case, and stochastic terrorism is not something the State can prosecute, then the question of free speech is void. Free speech is a State-endowed right, not something given by whichever God you prefer. But maybe Burgis thinks that our personal beliefs should be subject to this constraint. I think this is absurd.
I take it as given that we should want to have true beliefs about the relationships between speech and political violence. That helps us understand political violence against marginalized groups, identify the major threats and act in ways to protect the vulnerable groups. I think that means that the claim that: “If a version of “stochastic terrorism” is true, then we ought to believe it.” is true. We ought to understand political violence against Black people in order to better prevent it. What we then have is a question of if we should value “freedom of speech” over this understanding. But another person’s freedom of speech is not directly impacted by the beliefs we have about the person. Burgis almost certainly understands this. So shouldn’t Burgis be speaking out against not the theory of “stochastic terrorism” but instead the concrete actions that people take if they believe in this theory? Talking in the abstract, saying that “I believe in the right to freedom of speech” is one thing, talking in the concrete, saying “I don’t think people should be talking about how Ben Shapiro’s tweets might have influenced the Quebec City Mosque shooter.” is another.
Now, Burgis might still reply that the theory of stochastic terrorism makes the somehow apriori invalid move of conflating speech and violence. Except we already think that some kinds of speech can be violent, since things like emotional abuse are a form of violence. Either we can have “non-violent abuse” or some kinds of speech are violent. Now, it’s a separate question as to if the kind of speech we’re talking about when discussing “stochastic terrorism” is violent, but the possibility is still there. We can’t dismiss the claim that “$X$ speech is a form of violence” without some further argument. Violence isn’t merely physical.
That being said, I don’t think that we even need to say that the speech involved in “stochastic terrorism” is necessarily violence to use the framework. Certain speech can incite, can be vile, can be morally abhorrent and something we need to oppose, without being “violent”. What is it about “violence” here that’s so important? Surely Burgis would think that a teacher using racial slurs and berating a Black student is unacceptable. If he doesn’t think “speech is violence”, but he still opposes that, then what is the grounds? What are the grounds that he opposes Jones’ vile speech surrounding Sandy Hook?
There’s also a question I have; suppose that the theory of “stochastic terrorism” is true. That is, certain acts of violence against vulnerable groups are encouraged, facilitated and condoned by certain networks of actors. I don’t think that’s a question of free speech anymore, is it? It’s not like I can prevent that attack by arguing against the speech, or by presenting other speech. It’s entirely possible that my speech could be seen as a further reason to engage in the violent act. It is a place where “the cure for bad speech is good speech” just doesn’t make sense. Events like the 2017 Quebec City mosque shooting raise really complex, difficult questions about speech.
Now, Burgis isn’t required to give a proper definition of “stochastic terrorism”; he thinks the framework is bad. But what Burgis appears to be throwing out is any analysis of how speech can lead towards violence against people. He does this because he thinks that there is a threat that this will infringe on free speech. But I don’t think that’s a genuine threat. If Burgis could point to State involvement that he disapproves of and thinks is due to the framework of “stochastic terrorism”, or if he could provide some examples of inconsistent application of the term, maybe this discussion could actually proceed. Instead, Burgis has given us a vague gesturing towards unnamed inconsistency and an underexplained claim that it might infringe upon speech rights by “conflating speech and violence.”
Instead of engaging in little more than reactive disavowal against a framework he does not understand, Burgis would do well to address complexities and ambiguities thrown up for free speech by the concept of stochastic terrorism. These questions are here. Unfortunately, we are already down that road.
Editorial Note: The essay Burgis posted is a lightly edited excerpt from another peice, The Right Way To Politicize This. I initially replied to that, but seeing as the section I have issues with is now a standalone peice, I took an opportunity to edit and polish this peice up. The publication date has been modified to show this. ↩
There is a lot of nuance in this position for someone influenced by Nietzsche like myself. I am glossing this here, because I am perfectly willing to commit myself in this discussion to saying that truth has its own, autonomous, value. That is to say, we should value truth precisely because it is true, and these ethical claims don’t form a proper argument. If this question comes up again, more nuance can be given. ↩