In the first half of the 20th century, hundreds of Americans died each year from lightning strikes. The data is messy, but in the years from about 1920 to the middle of the 1940s, about 400 people were killed by lightning annually.
Last year, 23 did, the fewest on record. Other recent years have had a similarly small number of lightning fatalities, with 28 in 2012, and 26 in 2011, the previous record.
These numbers are all the more remarkable considering how the population of the United States has exploded over the same time period. Measured on a per person basis, the decline in lightning deaths over the last century is staggering, falling from about 3 or 4 annual deaths per million Americans, to fewer than 0.1 in recent years.
Why do so few Americans die from lightning strikes these days?
In the lightning-death literature, one explanation has gained prominence: urbanization. Lightning death rates have declined in step with the rural population, and rural lightning deaths make up a far smaller percent of all lightning deaths (see figure at right). Urban areas afford more protection from lightning. Ergo, urbanization has helped make people safer from lightning. Here's a graph showing this, neat and clean:
But is the move from farms to cities what is driving the decline?
Sure, lightning deaths and the rural population both declined during the 20th century, but so did a lot of other things, for instance, the percent of people living without electricity and plumbing, two infrastructural improvements that also help make your home less vulnerable to lightning. Of course, the development of better infrastructure—what I'll refer to as modernization—is related to urbanization, but it is not limited to urban areas. Over the 20th century, rural infrastructure modernized as well. How can we know how much each is driving the decline in lightning deaths?
There's one number we'd really need, and that's the death rate for the rural population over time. If the rural rate held steady, than urbanization is responsible. If it too dropped, we'd be able to get a glimpse of the relative merits of the urbanization and modernization theories.
Unfortunately, the data just aren't good enough to get at that level of granularity.
I spoke with Ronald Holle, a meteorologist who studies lightning deaths, and he agreed that modernization played a significant role. "Absolutely," he said. Better infrastructure in rural areas—not just improvements to homes and buildings, but improvements to farming equipment too has—made rural regions safer today than they were in the past. Urbanization seems to explain some of the decline, but not all of it.
"Rural activities back then were primarily agriculture, and what we call labor-intensive manual agriculture. Back then, my family—my grandfather and his father before that in Indiana—had a team of horses, and it took them all day to do a 20-acre field." Today, a similar farmer would be inside a fully-enclosed metal-topped vehicle, which offers excellent lightning protection. Agriculture has declined as a percent of total lightning-death-related activities, as the graph below shows, but unfortunately it does not show the per capita lightning-death rate of people engaged in agriculture.
Holle emphasized to me that there's not going to be one "simple, sweeping [explanation for] lightning fatalities." Modernization and urbanization may have each played their parts, but so have better education about lightning safety and improved medical treatments. By one count, in 1959 there were two injuries for every lightning death; by 1994 there were eight; by 1991 the ratio climbed as high as 10:1. Unfortunately, data from the mid-century probably dramatically under-counted lightning-related injuries, so these ratios have less explanatory power than it would at first seem. More significant are the twin factors of better infrastructure and a more urban population.
Death by lightning is perhaps the most cliche way to express the randomness that can befall a person. "You could get struck by lightning" is just another way of saying, hey, anything could happen.
And, of course, anything could happen. Lightning does strike. But its likelihood, and the likelihood of nearly any other seemingly random chance, is not some exogenous, constant factor, but a product of the millions of decisions we make, big and small, that together structure our society and our lives. So, yes, lightning could strike, but it just doesn't strike like it used to.