The Non-Identity Problem Part 2: Objections to the First Premise
Did anyone harm Pebbles?
Premise 1: Wilma’s choice to conceive Pebbles (the blind child) does not make Pebbles worse off
This article will discuss potential objections to the first premise, following Chapter 2 from David Boonin’s The Non-Identity Problem and the Ethics of Future Persons. You must read part 1 to understand this piece.
As a brief refresher, Wilma refused to take a harmless pill for two weeks. As a result, she gave birth to a blind child (Pebbles) rather than a sighted one (Rocks). Pebbles will live worse life than Rocks, but not one so bad that Pebbles will regret her existence. Most people hold the intuition that Wilma committed an immoral act by conceiving Pebbles rather than Rocks. Yet, a premise-by-premise analysis finds no clear reason to condemn Wilma’s decision.
Note that Boonin’s book delves much deeper than I could in a 2,000-word article. I will address the key points, but you should remain aware that the book dives much deeper into certain arguments.
This article will contain three sections. The first will run through three less interesting objections. The second features the de re/de dicto distinction, and the third discusses a potential asymmetry between pain and pleasure.
I also have one final caveat. I will borrow Boonin’s terminology of “pain points” and “pleasure points.” A thought experiment might say “child one experiences 2,000 pleasure points, while child two experiences only 1,500.” This terminology does not refer to vulgar utilitarianism. “Pleasure” merely refers to the things that make life worth living, while “pain” refers to the things that we’d rather live without. In other words, “pleasure” doesn’t necessarily refer to the hedonistic sort. In some contexts, it could include fulfilling experiences that one doesn’t enjoy at the moment. The same applies to “pain,” which encapsulates any negative experience that one would rather live without. I use this terminology because it’s easier to compare numbers than to devise a qualitative system like “good” vs “really good” vs “really GOSH DARN ROOTIN’ TOOTIN’ good.”
Misc. Objections
Incoherence
The incoherence objection argues that we can’t compare non-existence to existence, so it’s incoherent to claim that Pebbles’ life is better off for having existed. This objection fails for two reasons. First, it doesn’t seem to make sense. Most of us prefer to have existed, so it doesn’t seem like we struggle to make this comparison. People also make a similar comparison when they wish they would have been born later or would have rather been unconscious for some horrific experience. Second, this objection doesn’t solve the non-identity problem. If it’s incoherent to argue that Pebbles’ life is better off, it must also be incoherent to say it’s worse off.
Metaphysics
The metaphysical objection argues that Pebbles and Rocks are the same person. Like the one above, this one fails for the same two reasons: 1) it doesn’t seem true and 2) it wouldn’t matter if it were. As Boonin notes, the idea of Pebbles and Rocks being the same person leads to some bizarre conclusions. Imagine if Wilma faced the choice between a sighted child and blind twins. In that case, would both twins be the same person as Rocks? Also, what if Wilma needed to take the pills for two years rather than two months? She could then give birth to Rocks and Pebbles, thereby giving birth to the same person twice. The objection also lacks relevance since almost everyone objects to Wilma’s actions without considering Rocks and Pebbles to be the same person.
Continuing vs Creating
The final miscellaneous objection argues for a moral difference between continuing an extant life and creating a new one. This objection states that there can be cases in which life is worth continuing, but not worth creating. Here, we see the same issues as the previous two arguments: implausibility and irrelevance. For the implausibility piece, it’s not clear why any difference should exist. If events are worth experiencing, I don’t see why it should matter if those events occur for new people or extant ones. As for irrelevance doesn’t seem to apply to Wilma’s conundrum. Wilma’s choice involves creating people, so continuation doesn’t apply.
De Re vs De Dicto
If I say “NBA Finals MVP,” I might refer to Steph Curry, but I might also be referring to the active holder of that title. The former would be the de re sense of “NBA Finals MVP,” while the latter represents the de dicto sense. Since I can’t remember Latin terms, I’ll refer to de re as Specific Person (SP) and de dicto as Person Category (PC). If Curry’s wife says “my husband is the Finals MVP,” she’s probably not planning on swapping husbands after every season, so she’s using the term in the SP sense. If a basketball analyst says “the NBA finals MVP usually brings his team to the playoffs in the following season,” that analyst uses the PC sense.
Thus far, I’ve been speaking of Wilma’s actions in the SP sense. As a result of Wilma’s actions, “her child” is Pebbles instead of Rocks. If we view the phrase “her child” in the PC sense (I’ll write this as [her_child]), Wilma made [her_child] worse off, since [her_child] is now blind. This would contradict premise one and solve the non-identity problem.
For this solution to work, we’d need some reason to prefer the PC sense over the SP one. Unfortunately, the PC interpretation can lead to some odd results. Let’s say I have the choice between donating to the median child in Somalia (one of the poorest countries) and Luxembourg (one of the wealthiest). The Somali child currently lives a life of +30 pleasure points while the Luxembourgish one lives a life of +80 pleasure points. The donation would raise the Somali child’s life to +40, while it would raise the Luxembourgish one to +81. It seems like, in this case, I ought to donate to the Somali child. Yet, the DD interpretation produces the opposite conclusion. If I donate to the Somali child, [my_donation_recipient] sits at +40, compared to +81 for the kid in Luxembourg. According to the PC argument, I would make [my_donation_recipient] 41 points worse by donating to the poorer child. Thus, the PC interpretation can’t hold in all cases.
Still, this doesn’t refute the PC argument. Maybe the SP reading works for my Somali child’s thought experiment while the PC argument works in other circumstances. My previous paragraph only proves that the PC interpretation fails in some cases. To address the non-identity problem, we’d need a thought experiment with the same morally relevant factors as Wilma’s case. Boonin discusses Caspar Hare’s safety officer (SO) thought experiment as one potential candidate.
The Safety Officer
In this example, the titular SO implements reforms to reduce automobile injuries. Her efforts reduced the number of accidents from 100 to 70, and she reduced the severity of each accident from 50 pain points to 40 pain points. It seems that our SO acted in the right. Imagine, though, that her actions changed the identity of the accident victims. Without her regulations, 100 people born on even-numbered days would have gotten hurt. With them, 70 people born on odd-numbered days suffered injuries. That doesn’t seem to change our evaluation, even though every accident victim is worse off with the safety regulations than they would have been without those regulations.
Hare argues that our conclusion results from a PC interpretation. The regulations improved things for [the_accident_victims], even if the 70 odd-numbered people suffered a worse fate. Meanwhile, an SP interpretation seems to imply that the safety regulations made people worse off. The 70 accident victims would have avoided an accident without the regulations, so each accident victim was made worse off by the regulations. We would therefore conclude that the safety officer acted in the wrong. That conclusion is ridiculous, so the PC interpretation seems superior.
Next, Hare notes the similarities between this case and Wilma’s. In both instances, an individual accepts a role with a special responsibility towards certain people. The SO accepts responsibility toward accident victims, while Wilma does so for her child. Both cases also, he argues, present non-identity issues, where an SP interpretation couldn’t guide one’s behavior.
I’ve stuffed a lot of information here, so let me lay out the argument in a more familiar format:
Premise 1 (the interpretation premise): The SO thought experiment requires a PC interpretation
Premise 2 (the relevancy premise): If we accept a PC interpretation for the SO thought experiment, we must accept a PC interpretation for the blind child thought experiment
Conclusion: The blind child thought experiment requires a PC interpretation
The Interpretation Premise
Premise one states that we need PC reasoning to achieve the correct result for the SO. If think of the case in SP terms, Hare argues, we would condemn the SO and that seems absurd. However, I don’t think our intuition regarding the safety officer relies on a PC interpretation. I see a more plausible reason to believe that the officer acted in the right: she has a duty to reduce the overall amount of injury. She owes this duty for the same reason firefights must extinguish fires or police ought to protect people from crime. If people accept a duty, they have to fulfill it. Hence, we don’t need to consider the SP/PC distinction.
Even if we accept the SP/PC framework, I think we can obtain the right result with an SP interpretation. One could view the safety officer’s stakeholders as the general public in the SP sense. In other words, we can see the officer as having a duty to protect the currently-living people that are driving cars or will drive them in the future. Thus, we lack a reason to prefer a PC interpretation in this case.
The Relevancy Premise
Boonin argues that, even if we accept the Interpretation Premise, the SO case doesn’t apply to Wilma’s decision. Recall that Hare noted two parallels between the safety officer and Wilma:
Parallel 1: Both agreed to a special obligation towards a person or group of people
Parallel 2: Both involve decisions where a person cannot make SP decisions
Boonin imagines a doctor that can choose between accepting sick patients and healthy ones. This thought experiment shares parallel #1 since a doctor owes a special obligation to his patient. It also shares parallel #2. Both the sick and healthy patients would be made better off if the doctor cares for them, so SP considerations can’t drive his decision-making.
What should the doctor do? Under a PC interpretation, the doctor must make [his_patients] better off. Let’s assume that the doctor could lower the sick patient’s pain from 100 points to 80 points, and he could lower the healthy patient’s pain from 10 points to 9 points. If he accepts the healthy patients, [his_patients] will experience 9 paint points. If he accepts the sick ones, [his_patients] will sit at 80 pain points. Thus, choosing the healthy ones over the sick ones would lower the pain of [his_patients] from 80 to 9. The doctor would therefore be obligated to accept the healthy ones! This mirrors my Somali child thought experiment at the top of the section. Since a duty to accept the healthy patients (or donate to the Luxembourgish children) seems less plausible than the Implausible Conclusion, we can reject a de re/de dicto solution to the non-identity problem.
Pain/Pleasure Asymmetry
The last argument, from David Benatar, relies on an asymmetry between pain and pleasure. Causing someone’s pain is always bad while depriving a potential person of pleasure is not. Imagine two couples: the Cursed and the Blessed. If the Cursed Couple conceives, they will give birth to the Cursed Child. This child will experience constant pain, such that she would be better off having not existed. The Blessed Couple, on the other hand, would conceive the Blessed Child, and this kid will live a wonderful life. It seems that it would be morally wrong for the Cursed Couple to conceive while it would not be morally wrong for the Blessed Couple to decide against conceiving. Benatar holds that pain/pleasure asymmetry provides the best solution for the Cursed Couple/Blessed Couple thought experiment. The Cursed Couple commits an immoral act since they cause pain. The Blessed Couple merely foregoes a child’s pleasure, and the absence of pleasure isn’t bad if there’s no extant person to deprive it of. The pain/pleasure asymmetry would contradict premise one.
As Boonin notes, it’s not clear that a pain/pleasure asymmetry accounts for our different evaluations of the Blessed and Cursed Couple. There’s a more plausible reason to accept that Blessed Couple need not conceive: no one has an obligation to conceive1 a child, regardless of how good that child’s life will be. Furthermore, Boonin notes an uncomfortable implication of the pain/pleasure asymmetry argument. Any life, even the one of the Blessed Child, will experience some amount of pain. The Blessed Couple, therefore, faces two choices
Have the child, providing this child some pain (though this pain is outweighed by the child’s pleasure)
Don’t have the child, forgoing the child’s pleasure
Using the asymmetry argument, #1 would contain some moral wrong, given that all pain is wrong. Yet, #2 wouldn’t contain any wrongness, as the absence of pleasure isn’t wrong when there’s no one to experience that absence. As a result, it would be worse for even the Blessed Couple to conceive! This implication seems more implausible than the Implausible Conclusion, so we must reject it. Benatar could add an exception for cases where the pleasure outweighs the pain during the child’s life. That would be more plausible, but it wouldn’t solve the non-identity problem, since Pebbles’ life also contains more pleasure than pain.
Boonin discusses some additional formulations of the pain/pleasure asymmetry argument, but I won’t detail them here. Instead, I’ll try to provide a general idea of why asymmetry arguments fail. Rights only apply to existing people. Potential people have no right to exist, regardless of any pain or pleasure considerations. If we granted rights to potential people, we would all violate these when we decide not to have another child. One may make a potential person worse off by not conceiving them, but, if that person has no right to exist, then no wrong has been committed.
Some people believe in an obligation to have children for economic, religious, or personal reasons. Even then, it’s unlikely anyone believes that couples must conceive every possible child that would live a meaningful life.
Other note: the header image is not relevant, but it was one of the first results when I searched for “de re de dicto” among Creative Commons images. It looks pretty cool, so I stuck with it. Credit to this guy: https://www.flickr.com/photos/tonz/9481504464/
I’m really enjoying this series, but I’ve gotta say — a lot of Boonin’s arguments here seem like hand-waving. For example, the idea that “no one has an obligation to conceive a child, regardless of how good that child’s life will be” seems to me to be obviously false. Of course if you CAN create a good life, you should, and saying “no one is under an obligation” seems to me to be simply defining the problem away.
Of course Boonin’s position accords with our moral intuitions, and I understand that is the point of the exercise. But at some point we should recognize that our moral intuitions aren’t mathematically logical — they are rough-hewn heuristics arising from various evolutionary pressures. Expecting to be able to reason our way to an explanation that makes them all Hall into a neat logical pattern seems like a fool’s errand.
The reason we reject the ethical call to have as many children as possible is not that nonexistent people don’t have claims, but that we’re selfish. It’s the same reason we reject the claims on our actions of Effective Altruism, or Peter Singer, or ethical vegetarianism. Matt Yglesias has a good post today about how even in the middle of a genuine energy crisis, affluent people are unlikely to turn down the heat absent government rationing. At a certain point our own comfort and desire for our lives to turn out a certain way simply trumps any moral claim.
Those selfish desires may even have a rational evolutionary explanation — in the ancestral environment pleasures were rare and a signal that we were engaged in pro-survival behavior. It makes sense that we would prioritize them above all. But I don’t think that means that any individual choice we make by following our selfish moral heuristics is objectively or rationally justifiable.
There’s another, deeper problem, too. We are imbued with a deep survival instinct, such that, as you note, many people would intuitively agree that life is worth living. But… is it? Certainly almost all of us would agree that some lives seem not worth living, and on a sheer pain/pleasure ratio I’d guess MOST lives are at best a wash, and more likely a bad investment altogether. Yet we’re stuck with our survival instinct, aren’t we?
Given all this, I think we can’t really hand-wave away a lot of the objections by saying things like “it seems implausible” or “most people would agree….” Most people are driven to mad, illogical conclusions by a tangle of competing animal-level heuristics that don’t exist to make us happy or moral, but just to get us to the next generation.