After the indiscriminate shooting of club-goers was over early Sunday in Orlando, 49 people were dead. When the official death toll came out, media and experts almost immediately declared it the deadliest mass shooting in U.S. history, surpassing the Virginia Tech shooting in 2007 that left 33 people dead.

But was this actually the deadliest mass shooting in U.S. history? The short answer: yes, but it depends on how a mass shooting is defined. That definition has been the subject of some contention.

In the aftermath of the Orlando shooting, some publications called the Wounded Knee Massacre in 1890 the deadliest shooting in American history. There, as many of 300—mostly unarmed—Lakota Sioux men, women, and children were killed at the hands of the U.S. 7th Cavalry Regiment. It remains one of the darkest chapters of U.S. efforts to suppress Native Americans, though the 20 Medals of Honor that were given to members of regiment for their participation in the skirmish have yet to be rescinded. But was it the deadliest shooting incident?

One could even argue the deadliest shooting in a single-day event in the U.S. was at the Battle of Antietam during the Civil War in 1862, in which 3,654 people died.

But these two events—Wounded Knee and Antietam—were military operations and would not qualify as mass shootings, as most experts define it.

“When we talk about mass killing, we talk about it as criminal homicide,” James Alan Fox, a criminology professor at Northeastern University, tells me. “And, generally, killing related to military operations, wartime, etc., is not criminal homicide. It’s homicide, but not criminal homicide.”

Although there are times when American troops kill civilians and are prosecuted for it, like in the My Lai Massacre during the Vietnam War, the actions are still part of war. Mass murder, Fox argues, has criminal intent not related to the military.

But there is no one definition of a mass shooting. The FBI defines it as three or more people killed by a shooter in a single event. The bureau used to define mass shootings as the killing of four or more people, but Congress changed the designation in a 2013 bill.

Mother Jones, in its mass-shooting database, documents instances in which at least four people were killed in a single incident, in a public place, in events that did not involve domestic violence, gang violence, or an armed robbery. USA Today, in its tracker, also includes killings with four or more victims, but includes family killings and robberies.

But other trackers count shootings without deaths. Since 2013, a group of Reddit users have operated the Mass Shooting Tracker, tallying instances when four or more people have been shot, and not necessarily killed, in a single event. One founder told The Trace in 2015 that “the goal is to stop minimizing these acts of violence.” Many media outlets have picked up on this tracker and accepted its definition. It’s how The Washington Post could say the U.S. was averaging one mass shooting a day at one point last year.

And even Fox, the Northeastern University professor, has his own definition. He views mass shootings as when four or more people are killed in a single incident. (He also argues that the Mass Shooting Tracker index leads to “mass confusion” since the vast majority of events on the tracker do not involve deaths, which he says paints a false picture of gun violence in the U.S.)

Most of the mass-shooting databases were created in 2012, when many mass shootings—from Newtown, Connecticut, to Oak Creek, Wisconsin, to Aurora, Colorado—received widespread media coverage. Although the term “mass shootings” has been in our vernacular since, most notably, the 1940s (and readers have shared their theories for why that is to my colleague Adrienne LaFrance), what we now call mass shootings have occurred for decades, even if we didn’t track those incidents the way we do now.

We don’t have the historical context to determine whether or not mass shootings, whatever the definition may be, are on the rise—because there are likely hundreds of mass shootings that were simply not counted. What is clear is that mass killing in different forms, whether through guns, arson, or bombings, have been part of U.S. history.

The Orlando shooting is not the biggest mass killing in U.S. history. That title would go to the Oklahoma City bombing in 1995, where 168 people died. (The 9/11 attacks would not qualify since it was an act of international terrorism, despite the deaths of 2,977 people, not including the attackers.)

And the attack in Orlando may be the deadliest mass shooting in U.S. history, but other shootings were deadliest for specific locations. The 2012 shooting at Sandy Hook Elementary School, for example, where 27 people were killed, is the deadliest mass shooting in a high school or grade school in U.S. history. But that wasn’t the deadliest mass killing at a school in U.S. history. In 1927, 58 people died in a firebombing at the Bath Consolidated School in Michigan.

But what do these titles matter? Would these mass shootings or killings be any less serious or tragic if they weren’t the deadliest? Fox, the professor from Northeastern, says our attention to records is an obsession with consequences.

“Records are made to be broken,” Fox says. “We keep on saying it and there are other people who’d like to have that distinction of being the record-holder.”