Like many of us, my morning routine includes anxiously scrolling through the news as soon as I wake up. Lately, I’ve noticed a curious trend in how certain events are described.

See, for example, this The Washington Post headline from December 2021: “A Black teen died in custody while being restrained facedown. Now the prone position is again under fire.” Or this NBC News headline from May 2021: “Baby boy killed during attempted arrest in Mississippi, police say.” Or even this New York Post headline from June 2020: “Ohio protester dies two days after exposure to tear gas, pepper spray.” 

All three of these headlines have something in common: a missing subject — namely, the police. 

Pepper spray and tear gas don’t deploy themselves, and no one dies from “being restrained” without someone actively restraining them. But by failing to acknowledge this, these headlines deflect responsibility and separate police officers from the harm caused by their actions. This reflects a broader issue: the use of passive voice and intentionally vague language to obscure violence. 

In the early 2000s, political scientist William Schneider coined the term “the past exonerative tense” to refer to when people in power acknowledge harm but try to avoid blame. The classic example of this is when politicians admit that “mistakes were made.” For example, after the Watergate scandal, Richard Nixon’s press secretary conceded that “mistakes were made” when the White House attacked reporters. A decade later, Ronald Reagan similarly acknowledged that “mistakes were made” when the U.S. government sold weapons to Iran to fund rebel groups in Nicaragua. Mistakes were made, but who made them? They did. 

This rhetoric, as we can see, continues today, from news headlines to police department press releases. 

Pepper spray and tear gas don’t deploy themselves, and no one dies from “being restrained” without someone actively restraining them. But by failing to acknowledge this, these headlines deflect responsibility and separate police officers from the harm caused by their actions.

In December 2021, a Los Angeles police officer fatally shot a 14-year-old girl through a department store dressing room wall when confronting an unarmed suspect. 

The LAPD first reported this through a series of tweets, one of which read, “Update: As ofcrs contacted the suspect and OIS occurred, one of the officer’s rounds penetrated a wall that was behind the suspect, beyond that wall was a dressing room.  Officers searched the dressing room and found a 14 year old female victim who was struck by gunfire.” (The term “OIS” stands for “officer-involved shooting.”)

Strangely absent from this account is what caused the officer’s bullet to penetrate the wall and why the victim “was struck” by gunfire — of course, the officer himself. News organizations then began to parrot this language. A New York Times headline, for example, referred to the “Officer Whose Bullet Killed a 14-Year-Old Girl.”

This may seem insignificant at first, especially when confronting the broader problem of police violence. But it raises important questions about why reports of police violence are often so deferential to the police’s version of events. The phrase “officer-involved shooting” has become so commonplace that many people don’t think twice when they see it. Despite the fact that most of us now know what it really means, the phrase obscures a crucial piece of information — how exactly was the officer “involved”? 

Police officers are actively aware of this, too. In December 2021, the Baltimore Sun published an article with the headline, “Baltimore Police: Suspect in stable condition after being shot by officers in Northwest Baltimore.” Despite this seemingly unremarkable description, the Baltimore City Fraternal Order of Police — the Baltimore police union — strongly disapproved in a tweet: “More anti-police headlines from @baltimoresun.” 

Evasive and accountability-avoidant language isn’t limited to discussions of police misconduct. For example, according to a widely criticized set of guidelines from the Centers for Disease Control and Prevention in 2016, “Drinking too much can have many risks for women,” including “injuries/violence” and “unintended pregnancy.” The guidelines conveniently ignore the fact that the presence of someone else, often a man, is necessary for either unexpected violence or unintended pregnancy to occur. Instead, women are preemptively blamed. 

And headlines about sexual violence and sexual harassment frequently erase the actions of perpetrators and focus on what was done to the victim. In 2018, researchers found that news reports about campus sexual assault often use the passive voice to describe the actions of alleged abusers. But this has real consequences: A 1995 study found that when reports alleging violence against women are written in passive voice, male readers might minimize the harm experienced by survivors and attribute blame to the survivor instead of the perpetrator. 

A similar responsibility-avoidant phenomenon can be seen in coverage of military conflict, which deflects human agency onto nonhuman objects. For example, this Los Angeles Times headline from 2021: “Family says 7 children were killed in Kabul drone strike; U.S. is investigating.” Or, this sentence from a 2011 WIRED article about military drones known as Predators: “from April 21 to 9 a.m. Central European Time today, the Predators have launched 145 strikes, according to Pentagon spokesman George Little.” In both cases, responsibility for strikes and deaths is placed on the drones themselves, despite the fact that drones obviously can’t fire themselves — someone must have made the intentional decision to do so. The U.S. military’s active role in these decisions is hidden. 

All of these examples, while different, reflect a similar rhetorical approach: the tendency to hide human agency and place responsibility for harm onto objects — the bullets, the drinks, the drones — instead of onto the people who weaponize them.  

The use of vague, sanitized language to describe current events encourages a muted and distanced response to these stories. They might have elicited a more outraged response had they been described differently; for example, “Police officer kills 14-year-old girl.” 

All of these examples, while different, reflect a similar rhetorical approach: the tendency to hide human agency and place responsibility for harm onto objects — the bullets, the drinks, the drones — instead of onto the people who weaponize them.  

It’s become almost cliche to refer to things as Orwellian. But this might be a case where the comparison is necessary. In his 1946 essay “Politics and the English Language,” George Orwell criticizes the fact that language consists “largely of euphemism, question-begging and sheer cloudy vagueness.” Nowhere is this more evident than in the way we talk about and understand modern-day social issues — police brutality, sexual assault, war and more. 

By distorting reality and diverting attention away from those who commit harm, language can be used to maintain unequal power dynamics. The more I become aware of this, the harder it is to forget it. I often can’t browse the daily news without coming across at least one headline with a mysteriously absent perpetrator, a “protestor killed” or “suspect shot” without any mention of who caused this to happen. And while this may not appear to be the most pressing issue, it’s more important than it might seem at first. Linguistic choices unconsciously and constantly shape how we think and what we believe. Real change can’t happen unless we stop framing violence, intentionally or not, in ways that conceal the actions of perpetrators rather than highlighting them.