The killings of 20 first-grade students and six adult staff members at the Sandy Hook Elementary School in Newtown, Connecticut, on December 14, 2012, obviously constitutes a terrible, newsworthy event. But the news coverage did more than report the details of what happened at Newtown. It also sought to classify this incident as an instance of a larger problem. The initial news reports described what had happened as “the second deadliest school shooting,” “another mass shooting,” and a “mass killing” (all in stories in the next day’s New York Times) and as “the second deadliest shooting event in U.S. history” (The Washington Post). The Post’s website ranked the 12 “Deadliest U.S. shootings” (the earliest case on their list occurred in 1949), while the Mother Jones website added Newtown to its page “A Guide to Mass Shootings in America” (which included only cases from 1982 to 2012).
It may seem self-evident that the killings at Sandy Hook Elementary ought to be classified as a shooting event, or as a school shooting or a mass shooting. Of course we classify events into categories that make sense to us, and it is easy to take familiar categories for granted. We learn of terrible crimes and we are accustomed to commentators talking about incidents as instances. But the ways we make sense of the world—the terms we use to describe that world—are created by people, and they are continually evolving, so that specific categories come into and fall out of favor. In fact, in recent decades, Americans have understood events like the Newtown killings in a variety of ways.
In 1966, Charles Whitman killed 14 people and wounded 32 others by shooting from a 28th floor observation deck on the University of Texas campus in Austin. (In addition, before coming to the campus, he killed his wife and mother.) Whitman had been an Eagle Scout and a Marine; commentators at the time puzzled that an apparently respectable young man had committed such a terrible crime. The Whitman shootings occurred less than three weeks after Richard Speck had murdered eight student nurses by stabbing or strangling them. Reporters linked the two cases, and also mentioned other killers, such as Charles Starkweather and the “Son of Sam,” in articles about mass slayings, mass killings, or multiple killings.
In the 1960s and 1970s, then, it was understood that the key feature of these cases was a high body count. These early discussions of mass murder lumped together cases that varied along what would come to be seen as important dimensions:
Time: Did the killings occur more or less simultaneously, or did they extend over several days, months, or years?
Place: Did the killings occur in a single location, or in a variety of places?
Method: How were the victims killed?
By the early 1980s, the Federal Bureau of Investigation promoted the distinction between mass murder and serial murder. The Bureau had a new databank—the Violent Criminal Apprehension Program, or VICAP—that could help law enforcement identify similar crimes that had occurred in other jurisdictions. But in the aftermath of revelations about the FBI’s surveillance of the civil rights movement, an effort to expand the bureau’s domestic data collection invited suspicion and resistance. The FBI used the serial murderer menace—and particularly the idea that serial killers might be nomadic, able to kill in different jurisdictions without the authorities ever recognizing that crimes in different places might be linked—to justify the VICAP program. That set the stage for Clarice Starling and all the other heroic FBI agents who began pitting their wits against serial murderers in crime fiction and movies. “Son of Sam” would no longer be classified with Charles Whitman.
Charles Starkweather, who killed several people during a 1958 crime spree was also moved into a separate category. The bureau developed the concept of spree murders, a series of killings in different places over a period of time, often occurring as a fugitive tries to stay ahead of the law. Mass murder was now understood to involve, not just several victims, but killings that occurred in more or less the same place, at more or less the same time.
But how many victims are needed to make a mass murder? Obviously, whatever line is drawn will be arbitrary. Some analysts have argued for including any incident with three or more homicides; most favor four or more as the standard. The lower the minimum number of victims, the more incidents will be counted as mass murders. Setting the bar at 33 would exclude all of the killings on The Washington Post’s list of American incidents (which presents the 32 murders at Virginia Tech as the deadliest U.S. shooting), although other events, such as the 77 killings by Anders Breivik in Norway in 2011, would still count.
Journalists notice patterns, so similarities between cases invite the creation of new categories. For example, in 1986, a postal worker killed 14 postal employees; then, in 1991, there were two more incidents involving former postal workers killing employees at post offices. This led to the expression going postal. Eventually, after further incidents in 1993, the Postal Service responded with a program to improve their workplace and prevent violence. Some criminologists began writing about workplace violence, although this category was defined as including any violence in a workplace, not just mass murders. Under that definition, a large share of workplace violence involved robberies. (According to one analysis, the three most common sites for workplace violence were taxicabs, liquor stores, and gas stations—all isolated settings likely to have cash on hand.)
At the end of the 1990s, attention shifted to schools. During the 1997–98 academic year, there were heavily publicized incidents in West Paducah, Kentucky; Jonesboro, Arkansas; and Springfield, Oregon. The Jonesboro story made the cover of Time, which featured a photo of one of the shooters as a young child wearing camouflage and holding a rifle, with the caption “Armed & Dangerous.” Thus, the expression school shooting was already familiar a year before the April 1999 killings at Columbine. Advocates and academics began compiling databases of school violence, although the results were surprising: The average number of deaths per year fell, from 48 during the period from the fall of 1992 through the spring of 1997, to 32 during the period spanning September 1997 through the end of the school year in 2001, even though Columbine and the other best-publicized cases occurred during the latter period. In spite of commentators declaring that the nation was experiencing a wave or epidemic of school shooting, the evidence suggested that violent deaths in schools were declining.
It isn’t entirely clear what counts as school violence. The Journal of School Violence began publication in 2002, but its pages tend to be filled with articles about bullying, sexual harassment, prevention and response programs, and other topics that define violence broadly. The killings at Virginia Tech (2007) and Northern Illinois University (2008) were referred to as school shootings, indicating that colleges could count as schools.
Defining the problems in terms of its setting, such as workplace violence or school violence, suggests that steps need to be taken to protect people in those settings. Thus, the Postal Service devised a program to address workers’ frustrations and to prevent violence, just as, post-Columbine, many schools sought to beef up security measures. The response of the National Rifle Association following the killings at Newtown—recommending that more schools be staffed with armed security personnel—implied that the problem was violence in schools.
But the Newtown shootings followed a number of other high-profile shooting incidents outside of schools, including the 2011 shooting of Rep. Gabrielle Giffords (D–Ariz.) and some 2012 incidents at a movie theater near Denver, a Sikh temple outside Milwaukee, and a Minneapolis factory, settings with little in common. Gun control advocates, of course, had viewed school shootings as shootings, rather than tragedies in schools. They had been calling for a new federal “assault weapons” ban ever since the earlier ban expired in 2004, and the Colorado movie theater shootings had involved an assault weapon, so some of the commentary following that incident emphasized the nature of the gun. But the weapons used in other incidents, including the Giffords shooting, were pistols, leading most gun control advocates to emphasize that there had been a wave of mass shootings: that this was a gun problem that needed to be addressed. Many liberals had been critical of the Obama administration’s reluctance to draw attention to gun control as an issue during the 2012 presidential campaign. But the killings at Sandy Hook Elementary occurred after the election, and pro–gun control advocates and commentators now became even more vocal in drawing attention to mass shootings. Since the Newtown killings also involved an assault rifle, some advocates called for limiting magazine capacities or otherwise restricting assault weapons.
Yet there were competing interpretations. Coverage of several recent violent events—including the shootings at Virginia Tech, the Giffords rally, the Colorado movie theater, and Newtown—emphasized that the shooters had been treated for mental illness. Therefore, there were calls for more thorough background checks for gun purchases, checks that would make it harder for mentally disturbed individuals to gain access to firearms.
In the aftermath of the Newtown killings, there were interesting parallels between the comments of those advocating on behalf of gun owners and those advocating on behalf of the mentally ill. Gun advocates argued that there are hundreds of millions of guns, but that very few of those guns (or gun owners) are involved in mass shootings. Similarly, mental health advocates pointed out that tens of millions of individuals have mental problems, but very few of them turn to violence, let alone mass murder. Both arguments are, of course, true, because mass shootings are rare.
According to The New Republic, there were 70 mass shootings between 1982 and 2012, leaving 543 people dead. The magazine does not say whether this death toll includes the shooters, who often—but not always—also die. But let us assume that only the shooters’ victims were counted. (Amy Sullivan, the New Republic article’s author, says she believes that is the case.) That works out to about 2.3 incidents and 18 victims’ deaths per year. Last year was an unusually bad year, with 68 people killed in seven mass shootings—a terrible toll, to be sure. But in the context of some 2.5 million deaths from all causes last year, mass shootings, while dramatic, are simply not a major cause of death. And because these events are quite rare, and their number fluctuates from year to year, it is difficult to determine a clear trend. The horrors of 2012 made it easy from some commentators to claim that mass shootings were on the rise, but should there be fewer cases in 2013, it is unlikely that people will note that the problem is growing smaller.
My point is that it is possible to characterize Newtown as an instance of a lot of different social problems: as a mass shooting; as a school shooting; as mass murder; as workplace violence (remember the staff members who were killed were at work); as a crime involving an assault rifle; as a case of a mentally ill person committing acts of violence; and so on. We expect journalists to have a sort of sociological imagination, to help us understand incidents as instances. And we can understand why advocates for gun control, mental health, or other causes might favor particular labels, but we need to appreciate there is no One True Classification, that the categories we use are merely tools that may help us better understand what happening in our society.