Whose Wish Counts When You Forget What You Wanted?
What Happens When Your Mind Changes?
Marta is 72. A few years ago, when her thinking was sharp, she wrote a letter to her doctor. It said: “If I ever get Alzheimer’s disease and can no longer recognize my family, do not give me antibiotics to fight off infections. I would rather die than live that way.” She signed it in front of witnesses. That letter is called an advance directive — a written plan for what you want if you lose the ability to decide for yourself.
Now Marta has moderate Alzheimer’s. She rarely remembers the letter. She spends her days in the garden, humming to the marigolds, and smiles when the nurses bring her tea. She doesn’t seem to suffer. Then she catches pneumonia. Without antibiotics, she might die within days. A doctor holds Marta’s old letter in one hand and looks at Marta humming in the sunroom. The letter says let her go. The woman in front of her seems happy to be alive. Which voice counts: the Marta who planned ahead, or the Marta who is here right now?
Philosophers, lawyers, and families wrestle with questions like this. Your decisions about your own medical care don’t stop mattering just because you get sick — but what if the sickness changes who you are? To think clearly about it, we need to understand the two main tools that surrogates (the people who decide for someone who can’t) are supposed to use.
Two Ways to Decide for Someone Else
When you can’t decide for yourself — maybe because of an accident, an illness, or surgery — someone else steps into your shoes. The legal system and most ethicists say that that person can follow one of two main paths: Substituted Judgment or Best Interests.
Substituted Judgment means the surrogate tries to figure out what you would have chosen. Not what a reasonable person would choose, but you, with your specific quirks, loves, and fears. The idea is to respect your autonomy — your right to run your own life according to your own values. Even if you’ve lost the ability to think through a choice right now, the surrogate can reconstruct the decision you would have made if you were still able. If you left an advance directive, that’s the most direct clue. This is often called precedent autonomy: honoring a choice you made earlier, when you were fully yourself.
Best Interests takes a different angle. The surrogate asks: “What would be good for this patient right now?” Usually, that means things any reasonable person would want — freedom from pain, comfort, the chance to heal or to keep the abilities they still have. The moral idea behind it is beneficence: acting to benefit the person, not just to follow their past wishes. Best interests is used most often when there is little information about the patient’s specific values — for example, with a very young child or someone who has never been able to express personal preferences.
In real-life medicine, many doctors and judges think the two standards work like this: if you were once able to make your own choices (you are formerly competent), Substituted Judgment comes first. Your past autonomous decisions matter most, because respecting autonomy is usually seen as more important than just trying to do good. Only if we can’t figure out what you would have wanted do we fall back on Best Interests. If you were never able to make complex medical choices — what we call never-been competent — Substituted Judgment doesn’t even apply. How can we reconstruct what you would have decided if you’ve never been able to decide something like that? For that group, Best Interests is the only option.
This ordering sounds tidy. But scratch the surface, and cracks appear — especially in cases like Marta’s, where a person’s own mind has shifted over time.
A Quiet Challenge for Those Who Never Chose
The orthodox view says that for never-been competent patients, all you have is Best Interests, understood in a generic way: add up basic goods like avoiding pain, moving around, staying alive. But some philosophers point out that a person who lacks full decision-making capacity can still have a rich inner world of likes, loves, and fears. A child with a mild cognitive disability may not be able to weigh the long-term side effects of chemotherapy, but she may care intensely about being able to dance or to cuddle with her dog. If all we do is check a box labeled “prolong life” or “reduce suffering,” we might miss what actually makes her life meaningful from her own point of view.
In those cases, a careful Best Interests judgment starts to look a lot like Substituted Judgment: you try to reconstruct what matters to this individual uniquely, not what matters to just anyone. Supporters of the orthodox view can accept this. They simply say that Best Interests should be applied in a richly personal way — because any decent theory of well-being says that a person’s good includes the things she cares about, even if she’s never been fully autonomous. The difference is that here you’re not recreating an autonomous choice (she never had autonomy to begin with). You’re just taking seriously what nourishes her as the particular person she is. For infants or those with profound brain damage, however, you really do only have the generic route.
This challenge doesn’t topple the orthodox view. It only reminds us that Best Interests can be far more nuanced than people often assume. The really hard problems lie with the formerly competent — the Martas of the world.
When Your Earlier Self Clashes with Your Later Self
Marta is formerly competent. Her advance directive was clear. According to the orthodox view, her surrogate should refuse antibiotics and let her die — just as she asked. But is that obviously right? Marta today doesn’t seem to agree. She flinches at needles but smiles when the nurse says the medicine might help her feel stronger for the garden. She appears to have an interest in staying alive. The two sets of interests — Marta-before and Marta-now — pull in opposite directions.
This isn’t just a sentimental worry. Philosophers have mapped out exactly how the conflict arises. Before Alzheimer’s, Marta identified with her intellect. She saw a demented life as degrading. That gave her a powerful critical interest (a term we’ll explain shortly) in avoiding that fate. Now, in mid-stage Alzheimer’s, she no longer cares about intellectual independence. Instead, she has simple experiential interests: enjoying music, petting the therapy cat, talking to her sister. And she seems to have a genuine wish to go on living. So whose interests should rule?
One group of thinkers, led by Ronald Dworkin (1931–2013), proposes a threshold of capacity. The idea is that if your mental abilities fall below a certain line, your current interests lose their authority. Only the interests you formed when you still had full decision-making power truly count.
Dworkin builds this on a distinction between two kinds of interests. Experiential interests are about having pleasant experiences — like tasting chocolate, hearing a joke, or feeling the sun. They live and die in the moment. Critical interests, by contrast, are about making your life go well as a whole, according to your deepest values. A grandparent has a critical interest in their grandchildren’s success; a sailor has a critical interest in keeping their wooden boat in fine shape, even after they die. Dworkin says that earlier Marta’s wish to avoid dementia was a critical interest. And critical interests can still be satisfied now, even if Marta can no longer understand them. By honoring her advance directive, we make sure her life, judged as a whole, goes the way she autonomously wanted it to — and that, Dworkin insists, respects her as a person.
But then what is the threshold that strips current interests of their authority? Dworkin ties it to autonomy: the ability to act out of genuine character or convictions. Once you lose that global ability — not just the ability to make one tricky medical decision — you’ve fallen below the line. Marta, confused and unable to live by her own values, has crossed it. Her current smiles, he argues, are not the kind of interests that can override her earlier critical commitments.
Not everyone agrees on where the line lies. The contemporary philosopher Agnieszka Jaworska argues that the core of autonomy is not the ability to enact your values, but simply the ability to hold them — what she calls the capacity to value. If Marta still genuinely values her family ties or her afternoons in the garden, then she still has a stake in her own life that deserves authority, even if she can’t figure out how to act on those values in a complex medical situation. On Jaworska’s view, so long as someone can value, their current interests can override past plans. Thus, if Marta is capable of genuinely caring about her connection to her sister, the interest in staying alive holds authority, and the advance directive may be set aside.
So two versions of the threshold approach can yield opposite answers. Both claim to protect what’s most important — but they pick different bits of the human mind as special. This disagreement has sparked a series of deeper challenges to the whole idea that an earlier self can boss around a later self.
Four Reasons the Past Self Might Not Be in Charge
Philosophers who doubt the threshold approach have offered several powerful objections. None destroys the debate, but each forces us to think harder about why we give any past self the steering wheel.
1. Decisions look forward, not backward. The legal scholar Rebecca Dresser argues that surrogate decisions are about what happens from this moment on. The past cannot be changed. Marta’s earlier desire to avoid dementia cannot be fulfilled in the present — it’s too late for that — so it shouldn’t guide whether she receives antibiotics now. Dworkin’s reply is swift: critical interests don’t care about timing. Honoring her earlier wish still shapes the overall shape of her life, and that shape matters now, even if she doesn’t feel it.
2. If you can still exercise a will, that’s autonomy enough. Philosopher Seana Shiffrin highlights the value of being able to choose small things — to refuse a pill, to pick which shirt to wear. She says that any exercise of will is part of what autonomy protects. If Marta shows preferences, her current will deserves respect, not the will of the past self who is no longer here. A Dworkin-style reply is that exercising simple preferences isn’t the same as guiding your life by deeply held values. That richer autonomy is what makes past critical interests so weighty.
3. Maybe you’re not even the same person anymore. Some thinkers, drawing on the psychological continuity theory of personal identity, suggest that drastic brain changes like Alzheimer’s might destroy the very person you once were. If Marta now is a psychologically distinct individual from the Marta who wrote the letter, then the earlier Marta has no more authority over her than a stranger would. That sounds wild, but it’s a serious idea. Most philosophers respond by saying that personal identity isn’t broken that easily; even with major memory loss, it’s still Marta in the chair. The deeper move, pushed by Jeff McMahan, concerns not identity but prudential concern — the special kind of caring we have for our own future and past.
4. The ties that bind you to your future self can fray. McMahan (21st century) argues that the strength of our concern for our later self should depend on how psychologically connected we are to that self — shared memories, values, plans. In Alzheimer’s, the connections grow thin. Marta-then and Marta-now have very few psychological ties. That means Marta-then had little reason to be deeply invested in Marta-now’s fate — her earlier critical interest might be less urgent than it seems. On McMahan’s view, if the current self is still a person, her current interests win because she’s the one living the life. But if the current self is so deteriorated that she’s no longer a person, McMahan thinks the earlier person’s stronger interests can trump the trivial interests of a nonperson. Others, including Jaworska, argue that when the current nonperson’s interests are substantial — like an active interest in staying alive — they can override even a serious past interest in avoiding dementia.
The neat picture of advance directives and Substituted Judgment always winning has turned into a messy, fascinating puzzle.
Why This Fight Matters to You
Maybe you’re years away from writing an advance directive yourself. But the question crawling beneath all this philosophy is one you’ve already met: What makes you you? Is it your memories? Your values? Your ability to think complex thoughts — or simply your ability to smile when the sun hits your face?
When a grandparent gets sick, or when a friend suffers an accident, families often have to guess what the person would want. They might ask: Would Mom have wanted the surgery? She always said she never wanted to be a burden. But look, she still laughs at her favorite sitcom. The arguments we’ve just explored are being played out, quietly, in hospital rooms every day. They aren’t just about rules for doctors; they’re about how we love people who change in ways they didn’t expect.
The law tends to honor advance directives, treating the earlier self as the real voice of autonomy. But as philosophers have shown, the later self cannot simply be erased. The question of who gets to decide when you can’t is still open — and it depends on deep assumptions about personhood, time, and what it means to care for someone. Next time you watch a film or read a story where a character makes a promise to their future self, pay attention. You might find yourself asking: Would I still be bound by that promise if I became a different person? That’s exactly the kind of puzzle that keeps philosophers up at night — and it might keep you thinking, too.
Think about it
- If you wrote a letter saying you never wanted to be kept alive by machines, but later a brain injury made you forget the letter and you seemed happy, should doctors follow the letter? Why or why not?
- Can you think of a promise you made to your future self (like “I will never stop playing soccer”) that you might want to break if your interests change? Does the “old you” have the right to hold the “new you” to it?
- When you see an older person with dementia who seems content, do you think their earlier self’s wishes to avoid that state should still matter? What would make one self’s happiness more important than the other’s?
