Increase Care With Vulnerability

In an invented case, a clinician masks the eye region in a patient’s photograph and considers the patient anonymous. The International Committee of Medical Journal Editors says that is inadequate protection of anonymity. Its recommendations state that patients have a right to privacy that should not be violated without informed consent, that consent for publication requires that an identifiable patient be shown the manuscript, and that nonessential identifying details should be omitted. Whether someone can be recognized depends on what readers already know about them, which the author cannot see. The patient can, and that is why the standard puts the manuscript in the patient’s hands.

A uniform standard is the usual way to go wrong. The same casual permission serves for a former senator and for a nine-year-old, or the same ceremony of consent for both, and whichever you pick will be wrong for someone. Another error is the assumption that because the story is yours, the people in it come with it.

The American Anthropological Association’s 2012 principles name harm to dignity and to bodily and material well-being among the most serious harms, especially when research involves vulnerable populations. They say that the duty to do no harm can outweigh the goal of new knowledge and can lead a researcher to discontinue a project. A field that exists to produce knowledge has written into its code that the project itself may have to stop.

The principles also treat consent as something that continues. A memoirist interviews an aunt about a family illness in March, and the aunt’s yes seems to cover that conversation. Once the passage exists, in the words you chose, she is entitled to see it and to say no again. Journalism’s trauma guidance, such as the Dart Center’s advice to show survivors the sections that concern them, works the same way.

Vulnerability has components, and each can be asked about. How large is the gap in power between you and the subject, and how dependent is the person on you, as when access was given in trust? Is capacity diminished by age, illness, trauma or language? Can the person respond at all, or are they absent, incarcerated, dead, or without a platform? How irreversible is the exposure: safety, immigration status, medical information, a child’s identity? Care rises with each, and in practice that means more steps. Consent should be informed and should continue over time; the anthropologists describe it as a dynamic process that goes on past any signature. Show the person the passages that concern them, and omit inessential identifying details. Change details and say that you have; the anthropologists’ code allows pseudonyms and generalization, and the medical editors allow deidentification so long as it does not distort the meaning. Sometimes the step is to wait.

The duty probably rises because of how publishing normally corrects itself. The subject objects, corrects or replies, and that loop is what catches the writer’s errors and excesses. The less able the subject is to answer, the less the loop works, and the more of the answering must happen before publication, performed by the writer.

Consent is not always available, and not always decisive. The subject may be dead, incapable or hostile, and the Society of Professional Journalists asks for heightened sensitivity with juveniles, victims of sex crimes and people unable to give consent. A memoir about a parent with dementia entangles your story with a person who cannot agree, and no rule resolves it; writing the decision down, as the Practice below asks, at least makes it visible.

Ceremony can stand in for care. A signed release is not a substitute for judgment, and consent obtained from someone in no position to refuse is not consent. The anthropologists’ code says to avoid research where consent may not be truly voluntary or informed, and the memoirist’s version of that case is the relative who depends on you for care and cannot easily say no to you. These codes were written for clinicians, researchers and reporters, who usually work inside an institution that can review the decision. A memoirist often has no such body, so the questions in the Practice below have to do that work.

Care has a limit on the other side too. Erasing vulnerable people from the record is a harm of its own, and treating them as unable to speak removes the agency you are trying to protect.

Even drafting is part of the duty. Identifying details about a patient, a child or someone in danger are disclosed to any hosted tool you paste them into (Chapter 6).

Practice. Before publication, write a vulnerability and harm review. Answer in sentences: Who in this Work is least able to respond or to consent? What identifying details are inessential, and can they go? What would be irreversible if I am wrong about this passage? Who else is exposed by implication, such as children or relatives? What would I do if I could not ask them? Then record the changes: omit, alter and disclose, show the passage, delay. A short Work takes an hour. For a book, do one such review for each chapter that has a real person in it, an hour apiece, and stop when each of those chapters has its page. You end up with a dated page, or one per chapter, that you could show an editor.