Tracking Mass Shootings in Language

I always thought of the term “mass shooting” as somewhat contemporary. Gun violence has existed for as long as guns have, but the ways in which people talk about firearms has changed a lot over time.

One of the best ways to get a glimpse of the interplay between culture and language is by searching for terms in Google’s ngram, which scours millions of published books dating back to the year 1800 in a matter of seconds. It then charts the frequency with which your search-term appears. It’s not a perfect dataset, and there are a number of problems with the method, but Google ngram can be revealing—and perplexing, too.

For example, I was surprised to see the result of my search for “mass shooting.”

I expected the spike in the 1990s, but I didn’t expect to see the one in the 1940s, with a peak in 1945 that’s higher than today. What accounts for that? World War II? Some sort of issue with the search term or the dataset?

It’s grim subject matter and an unscientific lens through which to examine it, but I’d still like to understand how to interpret this. If you have any good ideas, drop us an email.