You have known since you were about four years old that you are going to die. Not suspected. Known, in the way you know that water is wet: you were told, you checked it against the world, and the world agreed. Since then you have watched it happen to other people, some of them people you loved, and each time the evidence arrived intact and unambiguous and applicable to you in every particular. And then you got up the next morning and thought about breakfast.
That should be stranger than it feels. A person told he has a bomb in his house does not go about his week. A person told the bomb will certainly go off, at an unknown hour, within a span he can roughly estimate, ought to be useless for anything else. We are all that person. Almost none of us behave like it, and the ones who do are considered unwell.
Tolstoy noticed the mechanism and put it in the mouth of a dying man. Ivan Ilyich had learned the standard textbook syllogism as a schoolboy: Caius is a man, men are mortal, therefore Caius is mortal. It had always seemed correct as applied to Caius. It had never once seemed to apply to Ivan Ilyich, who had been a boy with a particular mother and a particular smell of leather in a particular room, and who was not Caius and never would be. The logic was flawless and it landed on somebody else. Most of us are still doing that, and most of us do it well enough that we never notice we are doing it.
Freud had a blunter version. In his 1915 essay on war and death, written while the corpses were arriving from the front in numbers nobody had planned for, he argued that at the level where we actually live, the unconscious does not accept its own extinction at all. It has no representation for it. Whenever we try to imagine our own death, he said, we find ourselves still present as an observer, watching. The imagination cannot produce the one image required, because the image requires someone to see it.
Try it now if you like. You will get a funeral, and you will be at it.
The famous answer, and its problem.
The most influential explanation is Ernest Becker’s, from The Denial of Death in 1973. Becker’s argument is that terror of extinction is the engine underneath nearly everything humans build, and that culture is the machinery for converting that terror into something survivable. We cannot have literal immortality, so we manufacture symbolic immortality: children who carry the name, buildings with the name on them, nations, doctrines, a body of work, a god who has promised otherwise. Becker called it the hero-system. Every society, on this reading, is a structured arrangement for letting its members feel they are contributing to something that will outlast them, and the feeling is not a decoration on top of civilization. It is the part holding the weight.
Three psychologists built an experimental program on this, called terror management theory, and it produced one of the most cited findings in social psychology: remind people of their mortality, and they defend their worldview harder. They judge the outsider more harshly, cling tighter to the in-group, punish the transgressor more severely. Hundreds of studies, by one systematic count somewhere between six hundred and eight hundred of them.
Then the replication crisis reached it. In a multi-lab project called Many Labs 4, seventeen labs and about 1,550 participants after exclusions attempted the classic mortality salience effect, deliberately including conditions designed with the original authors’ involvement. They could not reproduce it under any condition. The authors concluded that the original finding was either a false positive or that the conditions required to produce it are not understood or no longer obtain.
The defenders answered, and their answer is not frivolous. Chatard and colleagues pointed out that the replication deviated from its own preregistration by including samples smaller than the plan specified, and that the null result was carried disproportionately by those small studies. Reanalysis restricted to the properly powered sites looked different. The current honest position is that the effect is probably real and much smaller and much less reliable than three decades of literature implied, and that it may replicate something like half the time.
Which leaves the grand theory in an awkward place. Becker’s picture is a magnificent piece of writing and a plausible account of why humans behave as they do. It has not survived measurement as cleanly as its popularity suggests. Hold it as a lens rather than a finding.
The unglamorous answers, which may be the true ones.
The first is that the imagination genuinely cannot do it, exactly as Freud said, and the failure there is structural rather than defensive. Nonexistence has no phenomenal content. There is nothing it is like. Asking a mind to model its own absence is asking a ruler to measure its own length, and the failure is not cowardice.
Epicurus made this into an argument rather than a limitation. Where death is, I am not; where I am, death is not; therefore it is nothing to me, and fearing it is fearing an experience nobody will ever have. Lucretius added the mirror: an eternity elapsed before you were born, and it did not trouble you in the slightest, so why should the symmetrical eternity afterward be different in kind. Thomas Nagel gave the standard modern rebuttal, which is that death is bad not as an experience but as a deprivation, the theft of everything you would otherwise have had, and that the two eternities are not symmetrical because only one of them takes something from you. Nobody has settled it in twenty-three centuries, which itself tells you something about how much of this is argument and how much is temperament.
The second unglamorous answer is that the ordinary day is genuinely absorbing. Not as a distraction from anything. Just absorbing. There is a leak under the sink and a message you have not answered and a person you want to see on Thursday, and the mind is a finite instrument that fills with what is in front of it. Calling this denial may be flattering it. Occupancy is not repression.
The third is temporal discounting, which is not special to death. We are bad at all futures. We undersave, we understretch, we schedule things for a version of ourselves who will apparently be better rested and more virtuous. Mortality is simply the largest item on a list of things we know intellectually and act on hardly at all, and it does not obviously require its own psychology.
Why watching it happen to others does not transfer.
This is the part I find most convincing, and it is the part that has the least literature behind it.
When someone dies, we do not experience it as a demonstration of the general law. We experience it as an event with a cause, and we go looking for the cause immediately and almost involuntarily. He smoked for forty years. She did not catch it early enough. He was eighty-nine and had a good run. There was ice on the road. Every death, given about ninety seconds, acquires a reason, and a reason is a thing that can be avoided by someone who does not have it. The reason is usually true. That is what makes the move so effective. We are not lying to ourselves. We are performing correct causal analysis in a way that quietly converts a universal into a contingency.
Heidegger caught the grammatical form of this. In the everyday way of speaking, he said, death becomes something that happens to one. One dies. The construction is impersonal and it does the whole job in two words. Our obituary language runs on the same fuel: passed, lost his battle, taken from us, at rest. Each phrase moves the event out of the first person and into an administrative middle distance where the subject is nobody in particular.
And the machinery does fail. It fails at four in the morning, usually alone, usually without warning, in the state Philip Larkin wrote about in “Aubade,” where the fact arrives without any of its usual insulation and simply sits there being true. It fails for people with health anxiety and panic disorder more or less continuously. It failed at scale in 2020 for a while. What is notable is not that it fails but how quickly it reseals, how completely, and how embarrassed people are afterward, as though the four a.m. version were the distortion rather than the clear-eyed one.
The part that should worry us.
If Becker was even half right, the arrangement has a cost, and the cost is political.
The machinery that keeps the terror quiet is the same machinery that builds the immortality project, and immortality projects are made of worldviews: the nation, the faith, the movement, the story about who we are that will still be true after we are gone. A worldview carrying that much weight cannot be treated as a set of propositions to be revised in light of evidence. It has to be defended, because what it is holding up is not an opinion. Becker’s darker claim, the one that made the book more than a consolation, was that this is where a great deal of human cruelty comes from. Not from hatred exactly. From people protecting the only structure they have that makes the fact survivable, and finding that other people’s structures are in the way.
I do not know whether that is right, and the experimental evidence for the specific mechanism turns out to be thinner than the citation count implied. But you can look at any political century, including this one, and see something shaped very much like it.
What is left.
Maybe none of this is denial. Maybe it is capacity.
Adults hold true things all the time without being ruled by them. You know the odds on the drive home and you drive home. You know what the doctor said and you make dinner. The capacity to carry a fact without being crushed by it is not a failure of seriousness, and it is not obviously worse than the alternative, which is a life spent staring at a certainty you cannot alter.
What I am fairly sure of is that the four a.m. version and the daylight version are both accurate, and that no arrangement of words has ever gotten someone to hold both at once for very long. The daylight version lets you build things. The four a.m. version tells you why building them matters. We seem to be constructed to alternate, and to be slightly ashamed of whichever one we are not currently in.
I remember confronting this quite early in life as my grandparents began to die, and I suspect some of the mechanisms you describe were operating then — particularly the curious ability to know perfectly well that death applies to me while somehow keeping its personal implications at arm's length.
But my relationship to this has changed considerably over the years, which makes me wonder whether there is another possibility you haven't quite articulated, although perhaps you point towards it when you ask whether living with mortality might be a capacity rather than a denial.
What happens when our relationship to identity itself changes?
From a first-person perspective, everything we ordinarily identify as ourselves — body, thoughts, memories, personality, the story of who I am — arises within awareness and is manifestly impermanent. As that becomes experiential rather than merely conceptual, the eventual disappearance of this particular identity can begin to seem rather less terrifying. That doesn't require the further claim that “I am Awareness” or that something called *me* survives death. I simply don't know.
Which is why I'd also question the statement that death means there is no phenomenal content. We know that we cannot experience our own non-existence, but do we actually know that physical death entails the cessation of experiencing? That seems to me a metaphysical conclusion rather than a phenomenological one.
Practices such as the Headless Way can offer a remarkably direct experience of the apparent nothingness within which everything arises. In that moment, the usual sense of oneself as a bounded entity can become surprisingly difficult to locate. But this is still an experience occurring while embodied, so I would be very wary of extrapolating from it to claims about what happens after physical death. Ramana Maharshi's famous confrontation with his own mortality is a more extreme example of an experience after which fear of death apparently fell away.
So perhaps there is another response to mortality alongside denial and the capacity to carry an unbearable fact: the fact itself may come to look rather different when the identity we fear losing is discovered to be less solid than we assumed.
I grew up with many older and infirm relatives. Death was a fact. I worried about them dying, but not about my own death.
My view is this: I will die and there is nothing I can do about it. Much like other things that I cannot stop from happening, I do not think about it until/unless something comes up that I can do something about. To me, worrying about dying is much like worrying about an asteroid strike that destroys all life in the planet.
I will be alive until I am not. Then, there will either be an afterlife (I like to think so), or not and there will no longer be a me to care about it one way or another.