My friend Ross Levatter sent me a thoughtful email challenging some aspects of my posts (here and here) on the risks we should fear. He gave me permission to post the whole thing. I also shared it with co-author Charley Hooper, who emailed me his thoughts. I’ll answer, and give Charley’s answer, after his letter. Here’s Ross’s letter:

I briefly skimmed your posts on what people are frightened of, where you note that many are frightened about rare events rather than more common causes of death or injury. You write “What should we fear? What threats are most likely to kill us?” Your underlying assumption is that the answer to the latter should strongly inform us in answering the former. But I don’t think that’s right.

The fact is, most adults have made peace with the fact they’re going to die at some point. Most people are not scared of dying per se. They go through the 5 stages, ending with acceptance. You don’t find massive amounts of fear in hospices.

It seems your paper’s underlying assumption is that FEAR of death by X should track LIKELIHOOD of death by X. But I don’t think that’s correct.

What ARE people frightened about?

1. UNEXPECTED deaths. Cardiovascular disease kills lots of people. Cancer kills lots of people. Shark attacks DO NOT kill lots of people. One might be frightened to anticipate dying of something that hardly kills anyone.

2. PAINFUL deaths. Here’s an Anthony Jeselnik joke: “My grandmother died last year. Initially, we all thought she died in the best way possible. What’s the best way possible? Right, in her sleep. But then we had an autopsy and we found grandma died in the worst, most gruesome way possible. During an autopsy.” In fact, numerically, dying in one’s sleep is orders of magnitude more likely than dying during an autopsy (which I doubt has actually ever happened unless you count vivisection.) But it’s hopefully understandable why the thought of the latter is more frightening than the thought of the former.

3. SUDDEN deaths. The *novel* coronavirus infection is very unlikely to kill you, especially if you’re under 70. But unlike dying of cardiovascular disease, this is not a death you’ve given any thought to up until a year ago. So it’s more frightening. A *NEW* way to die.

4. UNFAIR deaths. The odds of dying of a terrorist attack are extremely low. But to most people it seems very UNFAIR to be a perfectly healthy, active individual with decades of life ahead of him in the morning and dead in the afternoon.

I’m sure if I gave it more than casual thought I could come up with other distinctions, but I hope my point is clear. The belief that fear of death from X should track likelihood of death from X is an assumption that is not obviously true—it needs at least to be argued for—and is most likely, IMHO, false.

I basically agree with most of what he said because Ross keyed in on our words “fear” and “afraid.” I think we should have focused not on fear but on what things it makes most sense for a person to invest in preventing. Terrorism? No. Shark attacks? No. Being killed by a policeman while unarmed? No. Living a life to avoid heart attacks or delay them by 5 years? Yes. Being cautious around social groups to avoid COVID-19? Yes.

Charley put it better. He wrote:

His [Ross’s] points are correct and we did use the word “fear,” but what we were getting at isn’t fear as much as being smart about risks. If you want to live a long life, what should you think about and do?

To summarize his points, if we make peace with a manner of death, we no longer fear it. We fear those deaths that are sudden, painful, and unexpected.

We could have used “being smart about fatal risks” instead of “what fatal risks to fear.” But at some point, we start sounding like Star Trek’s Spock.

Now back to me: Consider the hospice point. If you’re in a hospice, then, if it was a good decision for you to be there, you have exhausted the alternative ways of preventing, without great discomfort, the thing or things that you’ll die of. It makes sense both not to fear the thing that will kill you and also not to invest further in preventing it. The time to act would have been much earlier when possibly a change in life style could have given you a couple more years of good life.

Ross then added a short additional point:

BTW, here’s another (I think incorrect) implication of this line of reasoning. Assume “Dying in your sleep” and “Dying from shark attack” are statistically exactly equally likely. Does it then follow that one should be equally frightened at the two prospects? I suspect most people, told they are equally likely, would still more greatly fear dying of shark attack.

I answered Ross as follows:

Yes, I would much rather die in my sleep, as my grandfather did, rather than dying in sheer terror, as his passengers did.

But seriously, folks, it is true that if you  choose between two ways of dying that have equal probabilities, the one that is more painful is the one to avoid.

You could alter our analysis by scaling the numbers, though, and in many cases that won’t matter. For instance, imagine that you would hate being killed by a shark 1,000 times as much as dying in your sleep. The odds of being killed by a shark are still so low that it doesn’t make sense to take account of that in deciding whether to swim in the ocean.