Why Do Sharks Eat So Many People?

“Dad?” my son asked as we got out of the car. “Why do sharks eat so many people?”

I assume he’d seen something about sharks at school, but the question still came out of nowhere. Thirty seconds before, he’d been telling me about his school’s Mardi Gras party and asking me if I felt better (because I’ve been sick, which is why this article still isn’t looking at the fossils he found). So, I put my mind to it. “I don’t think they do,” I answer.

“They don’t eat people?” he responds, sounding skeptical.

“Well, they can,” I concede. “But they don’t eat people all that often.”

“Why not?” he asks, and I swear he sounds disappointed.

“Well, a lot of them are kind of small. And we aren’t the kind of things they normally hunt.” I’m trying to remember shark facts, now. “Most of them are probably just wondering what we are, and people get scared of them and think they’re attacking.”

“Why do people get scared?” he asks.

I shrug. “Why did you get scared of them, when we went to the aquariam?”

“Because they look mean!” Then he grabs my hand. “Since you’re not feeling good, can we take a break and play Star Wars?”

Do sharks eat a lot of people?

This is one of those questions where I really hope I told my son the right thing, because I was working from half-remembered articles I’d read years ago, and memory is a fickle, tricky thing. Fortunately, it turns out that the University of Florida maintains the International Shark Attack File (ISAF), which they describe as “the longest running database on shark attacks, has a long-term scientifically documented database containing information on all known shark attacks, and is the only globally-comprehensive, scientific shark attack database in the world”.

Before we dive into the numbers, though, it’ll be important to define two terms the ISAF uses: provoked attacks and unprovoked attacks.

  • An unprovoked attack is edfined as “incidents where an attack on a live human occurs in the shark’s natural habitat with no human provocation of the shark.”
  • A provoked attack, on the other hand, usually occurs “when a human initiates physical contact with a shark, e.g. a diver bitten after grabbing a shark, attacks on spearfishers and those feeding sharks, bites occurring while unhooking or removing a shark from a fishing net, etc.”

The ISAF reports that 2016 was a pretty typical year, with a total of 150 alleged shark attacks. Out of those 150 attacks, here’s how the results broke out:

  • 81 were classified as unprovoked attacks (53 total in the United States, 43 of which were in Hawaii).
  • 37 were classified as provoked attacks.
  • 12 were sharks biting boats (aka “boat attacks”).
  • 1 was a shark eating part of an already dead human cadaver.
  • 12 had insufficient evidence to prove a shark attack.
  • 7 were determined to be attacks by other marine animals (including a barracuda and an eel).
  • 5 were determined to be abiotic injuries (that is, an environmental injury – scraping coral or rock, for instance).

So, in the fairly typical year of 2016, there were 118 total shark attacks on humans (131 if you count the boat attacks and the scavenging). 2015, by contract, hit 98 unprovoked shark attacks- the highest yearly total on record. On average, shark attacks result in 6 to 8 fatalities per year (depending on what period of time you average out), but 2016 only had 4 shark attack fatalities. Statistically, surfers are most likely to be attacked 958% of the total), followed by recreational swimmers and waders (32.1% of the attacks).

That’s not a lot of attacks, is it?

No, not really. In fact, there’s a whole lot of animals that are much more likely to kill you. According to the BBC, venomous snakes kill an estimated 50,000 people each year, rabid dogs kill an average 25,000 people each year, crocodiles kill an estimated 1,000 humans per year, and hippos kills an estimated 500 people per year. All of which makes the paltry 6-8 shark kills each year pretty tame.


Still, it’s probably best not to do this unless you know what you’re doing.

The ISAF provides some other interesting comparisons. The odds of dying from a shark attack are 1 in 3,748,067. Fireworks are about 11 times more lethal (odds of death: 1 in 340,733), sun and heat exposure are 273 times more lethal (odds of death: 1 in 13,729), and the flu is 59,664 times more likely to kill you (odds of death: 1 in 63).

Why do shark attacks get so much attention, then?

Rarity, in my opinion. Rarity and shock/

See, we notice things that appear out of the ordinary. According to the CDC, heart disease is the leading cause of death in the United States (614,348 deaths in 2014), followed by chronic lower respiratory diseases (147,101 deaths), accidents (136,053 deaths), stroke (133,103 deaths), Alzheimer’s disease (93,541 deaths), diabetes (76,488 deaths), and influenza and pneumonia (55,227 deaths). Most of those get no attention at all, unless they’re really spectacular or they happen to someone famous. But death by animal attack? That’s unusual, particularly in the United States. And particularly because we like to think we’re outside the food chain. Homo sapiens sapiens is, realistically, the ultimate alpha predator and one of the dominant environmental forces on the planet. It’s unsettling to be reminded that we’re still animals, and that we can still become prey.

Also, sharks inhabit an alien environment that we can only meaningfully visit with the benefit of technology. A wolf or bear attack happens on dry land, so the surroundings aren’t inherently hostile to us. But sharks? A shark could kill us with their own environment, even if the bite isn’t fatal. So, they seem frightening.

Just in case I am one of the unlucky ones, any tips?

Actually, yes. Here’s what the ISAF says: “If one is attacked by a shark, we advise a proactive response. Hitting a shark on the nose, ideally with an inanimate object, usually results in the shark temporarily curtailing its attack. One should try to get out of the water at this time. If this is not possible, repeated blows to the snout may offer a temporary reprieve, but the result is likely to become increasingly less effective. If a shark actually bites, we suggest clawing at its eyes and gill openings, two sensitive areas. One should not act passively if under attack as sharks respect size and power. “


When will you go to Heaven?

As I may have mentioned before, my son is five. And, like most of the five-year-olds I’ve met, he’s interested in and puzzled about death. Not in a morbid, death-obsessed way. He just doesn’t understand it, and he keeps asking questions because he’s curious. So one day – it’s been a while, so I don’t remember the exact circumstances – he looks up and asks “When will you go to Heaven, daddy?”

No matter what, that sort of thing catches you off guard. But I just shrugged and tried not to get super emotional as I look at him and say “not for a long, long time”. He looks at me and says “I’ll come visit you, so you don’t get lonely.” At that point, feeling myself tear up, I hugged him and changed the subject. Soon enough, we were chattering away about superheros. Really, it wasn’t a conversation I was prepared to have.

But awkward questions, often enough, are also great questions. For an answer to this one – how long am I likely to live – I turned to the Actuarial Life Tables of the United States Social Security Administration. To begin with, an actuarial table (also known as a life table) is a statistical tool used to predict average life expectancies. And at the time I’m writing this (at the age of 43.87 years old), the Social Security Administration expects me to live another 35.71 years (to an age of 79.58). The table also indicates that I have a 5.4844% chance of dying before I reach 80.

I’ll take those odds. I play D&D, so I know that a 1-in-20 chance of failure (or, in this case, death) isn’t too bad. And besides, if I make it to 80 the SSA predicts I have another 8.13 years in me. That gets me to the age of 88. At 88, I have a 13.7126% chance of dying before age 89 – still not too shabby. That’s only a little worse than 1-in-8 chance of death. And if I don’t die, the SSA predicts another 4.65 years of life. So, call my age 93 at that point.

At age 93, I’ve got a 22.4931% chance of death. That’s about 2-in-9 odds of dying. And they predict another 3.19 years of life, getting me to 96. 96 gets me to a 28.7218% chance of death (about 1-in-3), and another 2.61 years if I “roll well”. That puts me at 98, and my odds of death don’t change too terribly (32.4599%) – with another 2.33 years if I’m still alive. And that makes me 100.

HOw does that compare to my family? Hmmm… my dad died of cancer at the age of 41, and it still gives me chills that I’ve outlived him. My paternal grandfather died of cancer in his mid-sixties, and my maternal grandfather lived to his late eighties. So, honestly, that 80 to 88 range looks perfectly doable. But, well, like I said: I play D&D. I’ll risk the odds, and plan to max out the SSA table at age 119.