Every palliative care physician knows a particular kind of bedside.

The patient can no longer deliberate: a woman in a coma after an intracerebral hemorrhage, a man in the last stage of dementia who no longer recognizes his daughter, a newborn who will never acquire the capacities our ethical vocabulary takes for granted. On accounts that tie personhood to the exercise of reason, such patients stand on uncertain ground: still human, certainly, but persons in some diminished or contested sense (Blumenthal-Barby Reference Blumenthal-Barby2024; Eberl Reference Eberl2024). And yet few clinicians who have stood at that bedside experience the patient this way. Something reaches us there that cannot be reduced to autonomy, to clinical function, or to a data profile. This essay is an attempt to say what it is.

The answer I have arrived at, after several years of working on dignity at the end of life (García Abejas et al. Reference García Abejas, Santos and Mota-Filipe2025; García Abejas Reference García Abejas2026), I call embodied relational dignity. It is less a new theory of dignity than a claim about what a person is: that a person’s being and a person’s relatedness to others are not 2 facts that happen to coexist, but one. To be a person is already to be a claim upon another. Dignity, on this view, is not where clinical ethics begins. It is what we find once we have seen the person correctly.

This matters because every clinical ethics carries an unspoken picture of the person. Autonomy-based ethics sees a rational agent. Some strands of care ethics see a being constituted by recognition. Algorithmic medicine, often without noticing, sees whatever can be extracted as data. Each captures something true, and each, taken alone, loses exactly the patients palliative care exists to serve. If personhood depends on reason, the comatose patient has less of it. If it depends on being recognized, the patient whom no one has time to see on an overwhelmed ward has less of it. If a sufficiently complete record can stand in for the patient, the trace has been mistaken for the one who left it.

Four old answers, read from the bedside, help me say what these pictures miss.

The first is that a person’s worth does not depend on what she can currently do. Aquinas called the person the most perfect thing in all of nature, not because persons are always reasoning, but because each subsists as a whole: body and soul so united that they form one subject, not a mind that happens to use a body (de Aquino Reference de Aquino2003). Reason was never the bearer of dignity; the whole person is, in every state of the body, including the body that no longer deliberates. This is not a pious sentiment but a precise clinical claim. The patient in the coma has less capacity, not less personhood (Frantz et al. Reference Frantz, Rego and Barbas2025).

The second is that no one exists in general. Ortega y Gasset’s line, “I am I and my circumstance” (Gasset Reference Gasset2025), reads differently on a hospital ward. The patient is not a finished self who happens to be lying in bed 12; she is this person, in this illness, at this hour, with this family, in this institution. Her dignity does not depend on her circumstance, but its expression and recognition do. When screens, forms, and scores take the place of listening, what the person is does not change; what changes is whether she can still appear as who she is. Ortega also wrote that human life is a gerund, not a participle, always in the making (Gasset Reference Gasset1982). A case file is a participle; the patient is not.

The third is that none of this can be said calmly. Unamuno refused the abstract man of the philosophers in favor of the man of flesh and bone who is born, suffers and, above all, dies (de Unamuno Reference de Unamuno2021). Suffering is not noise to be filtered out of our concept of dignity; it is where the question of dignity becomes urgent. And, as Unamuno saw, it is our own finitude, felt rather than merely known, that allows another’s finitude to reach us as a summons rather than a fact. Palliative care has a simpler word for this: presence.

The fourth is that relation is not something added to an already complete person. Julián Marías, Ortega’s disciple, described the person as constitutively vocational: someone who is called and must answer, whose life is a project carried out with and before others (Marías Reference Marías1970, Reference Marías1996). At the bedside, this means the patient is not first herself and only secondarily related to those around her. Her dignity neither grows nor shrinks with her relationships, but it becomes manifest precisely in them.

These 4 describe who the person is, but they do not yet explain why, standing at the bedside, I am already obliged before I have finished any assessment. That is what Lévinas adds. For him, the other person reaches us first not as an object to be classified but as a face: a presence that exceeds every category we apply to it and addresses us with a demand before we have decided anything at all (Lévinas Reference Lévinas1972; Lévinas and Nemo Reference Lévinas and Nemo1985). Responsibility does not wait for the other to qualify as a person. It arrives first. The anthropology then explains why the claim was justified all along. Without the face, the person’s being would remain an inert metaphysical fact, waiting for someone to notice it; without an account of who the person is, the face would be a demand without content. Together they describe one thing: a person whose very being is already a claim on someone else.

Three consequences follow for practice. The first concerns consent and communication. Respect for autonomy remains necessary but is not sufficient: time, bodily presence, narrative listening, and attention to the patient’s circumstance are not courtesies added to consent; they are the conditions of meeting the person at all. The second concerns patients who cannot deliberate. They keep full moral standing, not as a concession, but because the vulnerable body before us makes its claim independently of any exercise of reason, past or future. The third concerns artificial intelligence and digital health. No model or record, however complete, can substitute for the person who appears before us. Digital systems in healthcare should be judged by whether they keep data tethered to that living person or quietly take her place.

Return, then, to the bedside. The woman in the coma has not become less of a person because reason has fallen silent; the whole of who she is remains, present in a body that can no longer speak for itself. She has not become less of a person because the ward has reduced her to a bed number and a set of vital signs; her circumstance is not a container to be looked past, but part of what she now is. Her suffering is not incidental to her dignity but the place where it becomes most visible. And she has not stopped calling to those who stand over her. The physician who leans in and listens where there seems to be nothing left to hear has already answered a claim that arrived before any diagnosis was made.

Clinical ethics has long debated dignity without first asking what a person is. I have come to think this is the wrong order. We do not begin with dignity. We begin with the person, and dignity is what we find when we have looked properly.