Hardly mentioned in the massive news coverage of the school-related shootings during the past year is how they ended. Two of the four shootings were stopped by a citizen displaying a gun. In the October 1997 shooting spree at a high school in Pearl, Miss., which left two students dead, an assistant principal retrieved a gun from his car and physically immobilized the shooter while waiting for the police.
More recently, the school-related shooting in Edinboro, Pa., which left one teacher dead, was stopped only after a bystander pointed a shotgun at the shooter when he started to reload his gun. The police did not arrive for another 10 minutes.
Who knows how many lives were saved by these prompt responses?
Anecdotal stories are not sufficient to resolve this debate. Together with my colleague William Landes, I have compiled data on all the multiple-victim public shootings occurring in the U.S. from 1977 to 1995. Included were incidents where at least two people were killed or injured in a public place; to focus on the type of shooting seen in the Ferguson rampage, we excluded gang wars or shootings that were the byproduct of another crime, such as robbery. The U.S. averaged 21 such shootings annually, with an average of 1.8 people killed and 2.7 wounded in each one.
We examined a range of different gun laws, such as waiting periods as well as methods of deterrence, such as the death penalty. However, only one policy was found to reduce deaths and injuries from these shootings: allowing law-abiding citizens to carry concealed handguns. The effect of "shall-issue" concealed handgun laws, which give adults the right to carry concealed handguns if they do not have a criminal record or a history of significant mental illness, was dramatic. Thirty-one states now have such laws. When states passed them during the 19 years we studied, the number of multiple-victim public shootings declined by 84%. Deaths from these shootings plummeted on average by 90%, injuries by 82%. Higher arrest rates and increased use of the death penalty slightly reduced the incidence of these events, but we could not conclusively determine such an effect.
http://www.junkscience.com/news2/lott2.html