My father is dying.

The hospice nurse believes he has between forty-eight and seventy-two hours left, and my sister thinks I should fly home tonight. I have decided not to go.

This is being treated as evidence of a moral defect rather than an allocation decision.

The ordinary argument is that this is my “last chance” to see him. That description is emotionally forceful but analytically incomplete. Every use of time is a last chance to use those particular hours. The fact that one opportunity is surrounded by relatives in a hospice room does not suspend scarcity.

The Relevant Counterfactual

The trip would require approximately eleven hours door to door: packing, transit to the airport, the flight, ground transportation, and the predictable period during which my family would insist that I sit in a room despite there being no new information to exchange.

The comparison is therefore not between visiting my father and doing nothing.

It is between visiting my father and the best available use of eleven hours.

I currently have a client deliverable due Friday, two hiring decisions outstanding, and a long essay on intergenerational obligation that will be read by several thousand people. These activities affect future income, the quality of my team, and the beliefs of readers who may make better decisions as a result.

Let:

U(V) = F + R + G − T − P − D

F = welfare delivered directly to my father
R = expected improvement in surviving family relationships
G = future grief reduction
T = value of time displaced by travel and attendance
P = productivity loss from disruption and sleep debt
D = decision-quality penalty from acting under social pressure

The family consensus effectively sets F + R + G to infinity. Once “dying father” enters the description, no amount of displaced work is permitted to matter.

I see no reason to accept that discontinuity.

Marginal Presence

My father is receiving morphine and is conscious for short, unpredictable intervals. During our call yesterday, he recognized me for perhaps two minutes, asked whether I was still in Seattle, and fell asleep while I was answering.

There is some probability that my physical presence would comfort him. But the expected duration of that comfort is small, and the causal mechanism is unclear. If he is unconscious, the benefit depends on speculative claims about auditory awareness. If he is awake, a phone call transmits the relevant information—my voice, affection, and confirmation that I remember who he is—at a tiny fraction of the cost.

My sister responds that “being there” is categorically different.

Categories do not create utility.

They summarize it, often badly.

My mother and both sisters are already present. The room therefore has substantial filial coverage. The marginal value of a fourth adult child is lower than the average value of family presence, just as the fourth person carrying a couch contributes less than the first two.

No one has explained why end-of-life attendance should be exempt from diminishing returns.

The Sunk-Relationship Fallacy

Several relatives have invoked everything my father did for me: tuition assistance, Little League, a loan when I was twenty-six, and the year he drove me to physical therapy after my accident.

I am grateful for these things.

They are also sunk.

Past parental investment may generate obligations under a reciprocal moral framework, but it does not establish that repayment should take the form of inefficient last-minute travel. I have already provided years of phone calls, holiday visits, technical support, and assistance comparing Medicare plans. The relationship is not an account in which the full balance becomes due upon terminal diagnosis.

The insistence on a final bedside appearance seems driven partly by narrative completion. People want a last exchange that retroactively organizes the relationship: a confession, forgiveness, one compressed statement of meaning.

This is an aesthetic preference.

My father and I have had decades to communicate. If the relationship contains unresolved material, that is evidence against treating the final semi-conscious interval as a reliable settlement mechanism.

Option Value and Timing

My sister argues that work can be done later, while this visit cannot.

Irreversibility matters, but it does not decide the question. Many low-value opportunities are irreversible. Tonight is also my last chance to work on this exact Thursday before my father dies. The one-way nature of an option increases its value only in proportion to the value of the underlying experience.

There is also a timing problem. If I depart immediately, there is a meaningful probability that he dies while I am in transit, converting the trip into attendance at the aftermath. If I wait for stronger evidence that death is imminent, I increase the probability of missing him entirely. The conventional solution is to travel at the earliest possible moment and absorb all uncertainty personally.

That policy makes sense only if presence has a very high value across every state of the world.

V_now = P(awake) · B_awake + P(unconscious) · B_unconscious
        + P(deceased) · B_bereavement − travel cost − displaced output

My family keeps pointing to the first term. Most of the probability mass appears to be in the other two.

Who Is the Visit For?

The strongest argument for going is not about my father. It is about my mother and sisters.

They are frightened, exhausted, and angry that I am evaluating the trip instead of experiencing it as mandatory. My presence could distribute practical work, provide companionship, and signal that I accept the family’s norms.

These are real benefits. They should be named accurately.

When my sister says, “Dad needs you,” she may mean that she needs me and considers the first formulation more binding. Recasting her preference as my father’s removes it from negotiation.

I have offered to pay for meal delivery, cover an additional night nurse, and remain available by phone. These interventions address fatigue more directly than adding another grieving person to the room. They were rejected because what my family wants is symbolic compliance.

Symbolic goods can have value. They do not become priceless because refusing them is socially legible.

Anticipated Regret

People assume I will regret not going for the rest of my life.

This may be true. But anticipated regret is not an oracle. It is a preference shaped by the same social norms under dispute. If everyone spends the next twenty years telling me that a decent son would have been there, some negative affect is predictable even if the original decision was correct.

We should distinguish regret caused by foregone value from regret manufactured by repeated condemnation.

I also expect to regret missing deadlines, forcing coworkers to absorb my work, and losing momentum on projects that matter to me. These forms of regret receive less moral attention because they do not fit a familiar deathbed narrative.

The asymmetry is suspicious.

Conclusion

I am not visiting my dying father.

I will call at 6:30, when the nurse says he is usually most alert. I have sent my mother money for additional care and given my sister the information she needs to handle his outstanding insurance claim.

Then I will finish the work I had already decided was the best use of this week.

If my father is conscious, he may be disappointed. If he is not, my physical presence cannot repair that fact. Either way, the moral weight assigned to the trip comes largely from convention, salience, and the desire to make death produce a final scene.

I do not think being someone’s son requires surrendering the ordinary arithmetic of time at precisely the moment everyone becomes least willing to perform it.