What if Trigger Warnings Don’t Work?

Earlier this year, Brandeis University’s Prevention, Advocacy, and Resource Center released a “Suggested Language List,” developed by “students who have been impacted by violence and students who have sought out advanced training for intervening in potentially violent situations.” The students’ purpose, they wrote, was “to remove language that may hurt those who have experienced violence from our everyday use.” They proposed avoiding the idioms “killing it,” “take a stab at,” and “beating a dead horse.” I was struck that one of the phrases they recommended avoiding was “trigger warning,” and that the proffered explanation was sensible: “ ‘warning’ can signify that something is imminent or guaranteed to happen, which may cause additional stress about the content to be covered. We can also never guarantee that someone will not be triggered during a conversation or training; people’s triggers vary widely.”

Trigger warnings started to appear frequently on feminist Web sites in the early two thousands, as a way to warn readers of fraught topics like sexual assault, child abuse, and suicide, on the theory that providing warnings would reduce the risk of readers experiencing symptoms of post-traumatic stress disorder, or P.T.S.D. Their use steadily increased online, particularly on social media. College students who were accustomed to seeing trigger warnings on the Internet began asking their instructors to provide them in class. In 2014, Oberlin College produced a trigger-warning policy as part of its Sexual Offense Resource Guide, advising faculty members to “understand triggers, avoid unnecessary triggers, and provide trigger warnings.” It claimed that a trigger, defined as something that “recalls a traumatic event to an individual,” would “almost always disrupt a student’s learning and may make some students feel unsafe in your classroom.” For example, “Chinua Achebe’s Things Fall Apart is a triumph of literature that everyone in the world should read. However, it may trigger readers who have experienced racism, colonialism, religious persecution, violence, suicide, and more.” Oberlin dropped the policy after receiving pushback from faculty, some of whom argued that the list of triggers was potentially endless.

Yet many academics embraced the use of trigger warnings. The philosopher Kate Manne explained, in a 2015 Times Op-Ed, that “the point is not to enable—let alone encourage—students to skip these readings or our subsequent class discussion (both of which are mandatory in my courses, absent a formal exemption). Rather, it is to allow those who are sensitive to these subjects to prepare themselves for reading about them, and better manage their reactions.” She wrote that exposing students to triggering material without trigger warnings seemed “akin to occasionally throwing a spider at an arachnophobe,” which would impede rather than enable the rational state of mind needed for learning. By 2016, an NPR poll of eight hundred college and university teachers showed that half of those surveyed had used trigger warnings in their teaching. Since then, trigger warnings have become culturally mainstream well beyond classrooms: last month, the Globe Theatre, in London, forewarned its audiences of “upsetting” themes in “Romeo and Juliet,” including suicide and drug use.

Whatever individual instructors might do in their courses, universities have not typically adopted official policies on trigger warnings. But the University of Michigan does provide its teachers a guide to trigger warnings within its resources on “planning for inclusive classrooms.” The guide urges instructors to design course content “with common triggers in mind” and offers examples of “tags” that teachers might provide on syllabi, including “death or dying,” “pregnancy/childbirth,” “miscarriages/abortion,” “blood,” “animal cruelty or animal death,” and “eating disorders, body hatred, and fat phobia.” The university tells teachers that “it is appropriate” to say to students: “If you have concerns about encountering anything specific in the course material that I have not already tagged and would like me to provide warnings, please come see me or send me an email. I will do my best to flag any requested triggers for you in advance.” When I read this, I pictured instructors attempting to comply with this advice by keeping color-coded tabs on individual students’ triggers in their teaching notes. In the event that teachers “miss flagging content that a student may identify as triggering,” they are told to “apologize sincerely to the student, assure them that you will try to do better, and ask for any clarification.”

As a law professor teaching criminal, constitutional, and family law—subjects that involve topics such as homicide, sexual assault, racial discrimination, guns, domestic violence, abortion, divorce, and child abuse—I know from experience that many students have endured very challenging life and family experiences that may not be apparent to others. As a result, my introduction to any course includes a statement that it will delve into many of the most controversial and difficult issues in our society, ones that may personally affect the lives of people in the class, and that all discussions must be conducted with respect for one another. I don’t frame my statements as addressing triggers, and I don’t flag particular readings or discussions, apart from the fact that a course unit may already have a heading: “Homicide,” “Sexual Assault,” “Segregation,” or “Divorce,” for example. (Of course, any student with a disability—including mental illnesses such as P.T.S.D.—may seek appropriate accommodations through the school’s disability office.) Since trigger warnings began to appear on syllabi, I’ve been troubled by uncertainty about whether students benefit from them. Like the Brandeis students, I’ve wondered, Is warning students that they are about to be traumatized or re-traumatized likely to decrease or increase the stress they feel?

Because trigger warnings involve assumptions about emotional reactions, particularly with respect to P.T.S.D., psychology researchers have begun to study whether trigger warnings are in fact beneficial. The results of around a dozen psychological studies, published between 2018 and 2021, are remarkably consistent, and they differ from conventional wisdom: they find that trigger warnings do not seem to lessen negative reactions to disturbing material in students, trauma survivors, or those diagnosed with P.T.S.D. Indeed, some studies suggest that the opposite may be true. The first one, conducted at Harvard by Benjamin Bellet, a Ph.D. candidate, Payton Jones, who completed his Ph.D. in 2021, and Richard McNally, a psychology professor and the author of “Remembering Trauma,” found that, among people who said they believe that words can cause harm, those who received trigger warnings reported greater anxiety in response to disturbing literary passages than those who did not. (The study found that, among those who do not strongly believe words can cause harm, trigger warnings did not significantly increase anxiety.) Most of the flurry of studies that followed found that trigger warnings had no meaningful effect, but two of them found that individuals who received trigger warnings experienced more distress than those who did not. Yet another study suggested that trigger warnings may prolong the distress of negative memories. A large study by Jones, Bellet, and McNally found that trigger warnings reinforced the belief on the part of trauma survivors that trauma was central (rather than incidental or peripheral) to their identity. The reason that effect may be concerning is that trauma researchers have previously established that a belief that trauma is central to one’s identity predicts more severe P.T.S.D.; Bellet called this “one of the most well documented relationships in traumatology.” The perverse consequence of trigger warnings, then, may be to harm the people they are intended to protect.

In other respects, trigger warnings seem to have less impact than their critics have feared. Some opponents of trigger warnings seem to suppose that they are a way for students to demand that they not encounter ideas that challenge their beliefs, particularly on social-justice issues. That opposition is part of broader worries about teachers “coddling” students, cultivating their fragility, or shielding them from discussions that might expand their minds. Trigger-warning studies, however, have revealed that giving trigger warnings does not seem to result in recipients choosing to avoid the material. Instead, the warned individuals tended to forge ahead.

If those suffering from P.T.S.D. were responding to trigger warnings by opting out of reading or discussing the flagged content, then, as McNally has pointed out, that would be concerning from a mental-health point of view, because the clinical consensus is that avoiding triggers worsens P.T.S.D. As McNally has written, treatment of P.T.S.D. patients involves “systematic exposure to traumatic memories until their capacity to trigger distress diminishes.” But, of course, teachers should not undertake to therapize or treat their students—nor should they act in ways that are known to be counter-therapeutic if they can avoid it. Since there isn’t evidence that trigger warnings help, and there is now some evidence that they might even increase anxiety, McNally, Jones, and Bellet do not recommend the use of trigger warnings. As Jones put it, “From a clinical lens, you should never do anything that doesn’t work, period, even if it doesn’t do harm. If it’s not actively helping, encouraging its use would essentially be engaging in clinical pseudoscience.”

One alternative, suggested by the Brandeis list, is to replace “trigger warning” with “content note.” Would that be an improvement? Bellet told me that “what really matters is not so much what you’re calling it as what you’re insinuating about a reaction.” Jones added that “humans are pretty smart about language, so, when you say ‘content note,’ if what you really mean is trigger warning, people are going to figure that out very quickly, and then the advantage of using a different word is gone.” Moreover, as the Brandeis students note, a person’s triggers are idiosyncratic and hard to predict; it may be false assurance to think that a list of stock triggers identified by administrators, faculty, or students reflects the sensitivities of individual trauma survivors. For example, a person who suffers from P.T.S.D. might be triggered not by discussions of events similar to their experience but rather by a smell or a sound.

If, in the face of scientific findings, one lets go of the view that trigger warnings foster mental health or facilitate learning, what reasons might remain for using them? It may well be that, at this point, trigger warnings have developed a spinoff cultural meaning that departs from the aim of providing psychological aid to those who suffer from trauma. A trigger warning might really function as a signal to the subset of students who are looking for it that the teacher is sensitive to their concerns—or at least compliant with their requests—regardless of psychological benefit or harm. The choice to send such a signal is of course part of a teacher’s academic freedom. But it is important to undertake it with the understanding that signalling compassion for students and trauma survivors in this particular way may be at cross purposes with helping them, whether psychologically or pedagogically.

As the scientific consensus on trigger warnings develops further, it’s conceivable that universities might even begin to worry about liability arising from their myriad instructors acting in ways that are known not to help—and possibly to harm—students’ mental health. Perhaps what is called for is a more neutral and humble stance, in which instructors don’t approach pedagogy as if it were an adjunct of psychological care. We could accept uncertainty about the psyches of individuals, disclaim the conceit of exerting control over triggers, and avoid confident assumptions about anyone’s trauma. That might help move teaching in a direction that attempts first to do no harm.