A theory of wellbeing contributes to explaining whether this or that state of affairs is a benefit or harm to a particular subject. A natural starting point from which to build such a theory is the subject’s valenced attitudes: I benefit from occurrences I like, desire, value, take a subjective interest in, etc. and am harmed by occurrences I dislike, desire not to happen, disvalue, take a subjective interest against, etc. Call this theory “Unrestricted Wellbeing Subjectivism.” The theory is unrestricted, since no state-of-affair types are excluded; that is, any occurring state of affairs that the subject takes a valenced attitude towards will benefit or harm that subject. There are several reasons philosophers have adduced in favor of restricting wellbeing subjectivism – i.e. in favor of stipulating that some specified types of events are ineligible to affect a subject’s wellbeing. One such source of reasons, against which I will defend unrestricted wellbeing subjectivism, is the problem of self-sacrifice.
The existence of self-sacrifice is difficult for unrestricted wellbeing subjectivism to account for. Consider the following example:
Thomas the Conductor: Thomas has been the conductor of the New York Symphony for many years and his skills are beginning to decline in his old age. He deliberates over whether to retire or continue conducting. He knows he will enjoy conducting more than retirement, but he decides, for the sake of the music, to retire. (He reasons that the music produced by the symphony will be better if the up-and-coming conductor, Rodrigo, takes over.)[1]
Thomas has made a self-sacrificial choice – he values the music even at a cost to his own happiness. On the other hand, he gets the outcome that he most prefers and, according to unrestricted wellbeing subjectivism, the outcome a person most prefers is the one that most benefits that person. But that is counterintuitive: How could it be that Thomas makes a choice that both benefits him more than any other choice and is self-sacrificial? That would violate what Heathwood calls “A Principle about Welfare and Self-Sacrifice: An act is an act of self-sacrifice only if the act fails to be in the agent’s best interest.”[2]
Heathwood’s principle makes intuitive sense, but in what follows I will try to convince you to reject that principle and instead accept that an act can be in the agent’s best interest while also self-sacrificial. My thesis thus follows Connie Rosati’s strategy of debating the definition of self-sacrifice rather than the traditional strategy of revising the definition of wellbeing.[3] Unfortunately, I don’t have the space in this blog post to explain Rosati’s view and the differences it bears to my own.
Here are two cases to get us started:
Fidel the Revolutionary: Fidel joins a militant group of revolutionaries. He suffers many hardships during the war. When the war ends, and because of his many acts of heroism, he becomes the leader of his country and his life goes very well for him. His life goes so well that it counterbalances the harms he faced during the war. Intuitively, Fidel’s decision to join the revolution was both self-sacrificial and, at the end of the day, in his best interest. (Assume that, counterfactually, he would have had a boring life of mediocrity had he not joined the revolution.)
Leah the Heroic Metro Rider: Leah sees that someone has fallen onto the subway tracks and is not able to climb back out. She leaps into action, jumping down and pushing the person back onto the platform, even though she sees the subway train coming and knows there is a significant risk of not having enough time for herself to escape. Luckily, the train’s emergency brakes engage and nobody is harmed (nor would anybody have been harmed even if Lead had not leaped to the rescue). Nonetheless, the person formerly on the tracks, out of gratefulness and admiration, gives Leah $100. Intuitively, Leah’s decision to put herself in harm’s way was self-sacrificial and, in hindsight, in her best interest.
What happened in these cases? What features give rise to the surprising result that a self-sacrificial act can be in a person’s best interests? The answer, I believe, is that self-sacrifice, unlike best interests, is partly determined by a person’s motivations at the time they make their decision. We judge that Fidel and Leah act self-sacrificially because they are motivated to act for the sake of something greater or other than their own wellbeing (a political cause for Fidel, the wellbeing of a stranger for Leah). Whether a state of affairs is in a person’s best interests, in contrast, is determined solely by how well things actually go for a person and not depend on the subject’s motivations.
What kind of motivations am I talking about, exactly? The agent must be motivated by the goal of benefiting some “thing” other than the agent herself. The beneficiary may or may not be a person. Thomas the Conductor sought to benefit “the music.” Consider also Gloria, who is motivated by the goal of benefitting the New York Symphony qua institution. We can imagine Gloria taking the time to philosophize about what, exactly, would count as making the Symphony a better or worse institution (non-instrumentally speaking) – the quality of its music? The wellbeing of its musicians? Its cultural impact? The duration of its existence? – but none of that philosophizing is necessary in order for Gloria to act for the sake of the Symphony. Acting for the sake of the Symphony requires merely that Gloria have a very basic, intuitive sense of what benefits or harms it. Acting for the sake of music requires merely that Thomas have a basic, intuitive sense of what it takes to benefit music. Motivations need not be deliberate, fully-fleshed out, or even philosophically coherent. Motivations are the implicit “folk theories” that a person acts on.
Of course, many of us often act for our own sake or out of behavioral inclination. Consuming chocolate ice cream, for example, involves both the former motivation and the latter “motivation.” Those are the kind of motivations that have nothing to do with self-sacrifice.
Motivations alone cannot be the entire answer, of course. Common sense demands a sacrifice. Consider the following case:
Rodrigo the Conductor: Rodrigo, a youthful, up-and-coming talent who, like Thomas, is motivated to act for the sake of the music (as a final end). Unlike Thomas, however, for Rodrigo to act for the sake of the music is for him to accept the position of conductor that Thomas has just relinquished. Rodrigo is quite eager to conduct the New York Symphony. Even if, counterfactually, he did not take the music as a final end he would still enthusiastically desire this opportunity.
Rodrigo does nothing selfish, but nor does he act self-sacrificially. His motivations are pure enough, sure, but where’s the sacrifice? In addition to the right motivations, then, an act of self-sacrifice requires an opportunity cost, local harm, or risk of harm to the agent. The extent of the local cost will determine the degree of self-sacrifice. Fidel, Leah, Alice, and Thomas all meet these criteria, whereas Rodrigo does not:
| Case | Opportunity Cost, Local Harm, or Risk of Harm? | Motivated to (Non-instrumentally) benefit something other than the agent? | Self-Sacrificial? |
| Fidel the Revolutionary | Yes – Risk of Harm | Yes – Political Cause | Yes |
| Leah the Heroic Metro Rider | Yes – Risk of Harm | Yes – Another Person’s Wellbeing | Yes |
| Alice’s Friday Night | Yes – Opportunity Cost | Yes – Several People’s Wellbeing | Yes |
| Thomas the Conductor | Yes – Local Harm (or Opportunity Cost, depending on the proper theory of harm) | Yes – The Music | Yes |
| Rodrigo the Conductor | No | Yes – The Music | No |
Note that the definition I propose here is external to philosophy of wellbeing; it is compatible with any theory of wellbeing. Of course, this paper is intended to be a defense of Unrestricted Wellbeing Subjectivism, but the definition and its argument stand on their own.
[1] Inspired by the character of Thomas Pembridge of the television show Mozart in the Jungle.
[2] Chris Heathwood, “Preferentism and Self-Sacrifice,” Pacific Philosophical Quarterly 92, no. 1 (2011): 21.
[3] Connie S. Rosati, “Self-Interest and Self-Sacrifice,” Proceedings of the Aristotelian Society 109, no. 1 pt. 3 (2009): 311–325.
______________________________________________________________________________________________________________
Avi Appel is a Philosophy PhD candidate at Cornell University focusing in Ethics.
Hi Avi,
I’m trying to get clear on both the problem and the solution. Why isn’t the problem that we can do what we most desire and yet end up worse off? If we imagine a case like this: someone literally sacrifices their life for another person. It seems intuitively that they are not better off for having done so even if they did what they most desired. But, a critic might charge, on a desire-based subjectivism wouldn’t that make the person well-off (getting what they desired)? I suppose you are raising a different problem. On desire based subjectivism it seems we can’t ever do self-sacrificial acts *if we mean by that those that we both desire to do and those that lower our overall well-being*. I take it that your solution is to say that we can do what we desire and lower our local well-being and that lowering counts as a sacrifice. That seems fair enough as an explanation of how subjectivists can let some self-sacrifice into the picture but then I wonder if you still can’t account for sacrifices that lower our global wellbeing. How can subjectivism say we can do self-sacrificial acts *if we mean by that those that we both desire to do and those that lower our overall well-being*?
Hi Nicole,
The problem of self-sacrifice is that there are too many cases that intuitively involve self-sacrifice but aren’t (prima facie) judged as such by Unrestricted Wellbeing Subjectivism. Thomas the Conductor is one of those and it is very easy to imagine more cases that have the same relevant features as the Thomas the Conductor case.
Regarding the person who sacrifices their own life: I think Heathwood has the right solution to this problem in the paper of his I cite. A Wellbeing Subjectivist can say that although she is getting what she most desires at Time T-1 (the time at which she chooses to kill herself), she isn’t getting as much desire-fulfillments at times T-2, T-3, etc. At every moment after she kills herself (up until the time she would have died anyway) she gets zero desire fulfillment (as do all dead people) when she could have had lots of desire-fulfillment. So getting what she now most wants is not the same as maximizing her global (i.e. lifetime) wellbeing.
Regarding your question about accounting for sacrifices to one’s global wellbeing: Again, some of those cases can be handled in the way that Heathwood proposes in the same paper and using a very similar answer to the one given in the previous paragraph. I call these “Odysseus Choices” after the episode of the Odyssey where Odysseus commands his crew to tie him to the mast. Anytime we make a choice that now fulfills our strongest desires but which we believe will frustrate the strongest desires we will come to have in the future, we make a self-sacrificial Odysseus choice. Albeit, the person who suffers from the choice is yourself-in-the-future. (To clarify: Odysseus was not acting self-sacrificially, but he was making a decision based on his current preferences that he knew would go against what he would soon desire.) So for example, a person might be convinced by his friends and family that life in the army will be so miserable that he will wish he had never enlisted. But in the present his desire to benefit his country outweighs concerns for his future wellbeing. He gets what he now most wants but he does not maximize his lifetime (i.e. global) wellbeing.
Hi Avi,
Thank you for this interesting post. I have been working on the problem of self-sacrifice myself and I agree that the agent’s motive matters here (so does Connie Rosati, in the paper you cite). You are also surely right that there needs to be some harm (or perhaps risk of harm). I am a bit puzzled about how exactly your solution works and I was hoping you could clarify it.
I take it your solution for, say, Thomas’ case is that since he is sacrificing his happiness and doing it for something else, it qualifies as a case of self-sacrifice. In terms of the Unrestricted Wellbeing Subjectivist, this would be considered a voluntary frustration of desires (or preferences, etc.). But the Subjectivist also holds that Thomas is thereby satisfying his greatest desire, outweighing those frustrations, which yields a benefit to Thomas – not a sacrifice. Does your solution involve somehow isolating the frustrations from the satisfactions, so that we can count the loss as a sacrifice, even if the same act also brings a greater benefit? Is this what you mean by “local” harm? But wouldn’t this at best lead to what Overwold (1979) has called a case of cutting one’s losses?
In your response to Nicole, you describe what you call “Odysseus Choices”. But those need not be involved in your cases. Fidel, for example, might most want to join the revolution, and later, when he’s enduring hardships in the war, he might still most want to continue fighting for the revolution. So he is frustrating current desires for a current, stronger, desire, not later strong desires for a current strong one. He is suffering harm, and he is doing so for a greater cause, but it seems to me that a Subjectivist still couldn’t account for the sacrifice, since Fidels’ greatest desires get satisfied each time.
You also say you accept Heathwood’s solution for some kinds of cases, but doesn’t his approach conflict with yours, since he cannot allow for any local-global distinction? (This, btw, is one of the issues I’m most concerned with, and I think there looms a problem for Heathwood).
I would be interested to hear your thoughts on these points, especially if I am misinterpreting you. Thanks!