This essay is me trying to figure out the “edges” of Singer’s argument in Practical Ethics.
I’ve written and rewritten it several times, and it bothers me that I don’t reach a particular conclusion. The essay itself remains at the level of “musings” instead of “worked out, internally consistent philosophical refutation”.
Nevertheless, I want to share my thoughts, so publishing it anyway.
Some specific disclaimers:
I agree with many Singer’s conclusions.
This essay is based on my extension of Singer’s argument. Even though he, to my knowledge, hasn’t explicitly put forth these specific arguments, I believe that they logically follow from those ideas that he has put forth. Nevertheless, I may have misunderstood something and may be arguing against a straw man. If so, please flag it.
My criticism is directed mostly against the “idealized” moral agent which, as far as I understand, Singer accepts as not a real expectation from anyone. That is, there are situations where according to Singer, the right thing to do is to do X, and what people do is not X, and what is reasonable to expect of them is simply to strive for X. I don’t necessarily argue against striving, but I do argue against what is or isn’t right for an agent that doesn’t only strive, but actually does X.
Intro to Practical Ethics
If you’ve read the book, or are otherwise familiar with its arguments, feel free to skip to the next chapter.
Singer claims that you must make ethical decisions based on an equal consideration of interests, and not any other property.
It does not matter what age, race, religion, sex, or species one is – the only thing that matters is one’s capacity to suffer, and one’s capacity to view oneself as a distinct entity, with a past and a future.
Take, for example, eating meat.
It is the human’s interest to feel pleasure from eating a tasty steak. It is the cow’s interest to not be killed.
According to the principle of equal consideration of interests, the cow’s interest to not be killed (nor exposed to factory farming practices) clearly outweighs the human’s interest in eating tasty meat.
There is also a moral ranking here that is based on how refined one’s capacity to suffer is. For example, humans are both sentient and capable of seeing themselves as distinct entities existing over time. Cows are merely sentient.
But if there are some humans who are not sentient nor capable of seeing themselves as distinct entities existing over time (for example, patients in a permanent vegetative state), then they have a lower moral footprint than a sentient cow. The cow still cannot conceive of itself as existing over time (probably), but it can experience suffering, which is more than such a human can.
Therefore, in that case, a cow has a higher moral status, and it would be more wrong to kill that cow than that human.
(Singer explores some edge cases, implications on others and on societal norms; I’m shortening the argument here.)
General moral argument against proximity
Singer claims that proximity is not adequate for moral judgment. If we generalize his argument beyond species, race, religion, nationality, to all markers of proximity, we must come to the conclusion that family is equally excluded from moral protection.
My family members are proximate to me in the sense that we have similar genes, and in the sense that we are one tightly-knit group, irrespective of genes (for example, families with adopted children).
Singer claims that genetic proximity is not a relevant moral factor – he rejects preferential treatment based on species, or race. Therefore, if I extend that line of argument, I cannot provide preferential moral treatment to my family based on their genes.
He also claims that other proximity which is not genetic – such as similarity of religion, or nationality – is equally not a relevant moral factor. Therefore, if I extend that line of argument, I also cannot provide preferential moral treatment to my family based on us being the same group.
Therefore, we must either:
Accept the conclusion that family members should not get any preferential moral treatment from us, or
Make an exception for families, and allow that equal consideration of interests applies in other cases, but not in the case of family.
Thought experiment: burning building
Singer also claims that infants do not have the same moral status as adults. They have no conception of themselves as “a distinct entity existing over time”. They have potential personhood, but Singer claims that potential personhood is not as strong of a moral claim as real personhood.
Here’s a thought experiment:
You apartment building is on fire. You rush in. There’s time to save exactly one person: your 6-month-old baby, or an adult stranger.
If we must not give preferential moral treatment based on proximity, and if infants do not yet possess morally relevant characteristics, then the moral thing to do would be to let your child die in the fire, and save the stranger.
I believe that every moral framework that would have you let your child die so that you can save a stranger’s life is wrong. It must have gotten lost along the way somehow, and it is our task now to find where exactly this framework has gotten lost.
I do not believe that infants actually have the morally relevant characteristics that adults have. And I similarly agree with the premise that future personhood is not as strong a claim to moral status as current personhood.
No, the reason why you should save you child, is that it’s your child, which means that I reject the argument against proximity.
Addressing “roles and expectations”-based counterarguments
A counterargument might be: “you have chosen to have this child and therefore you have a moral obligation to it; it’s different from arbitrary things like nationality or religion.”
We can change the thought experiment to not have your own child in the fire, but your baby brother.
In that case, there is no choice that was made, and you have entered no “contract” that forms a moral obligation of care towards this being; it’s a genetic accident that you had no influence on.
Yet, I argue, it would entail the same effect: if you rush into the building, you should most definitely save your baby brother, and not an adult stranger.
Addressing “favoring family leads to better overall outcomes”
Singer claims that, in aggregate, a society where one is more favorably disposed to one’s family (such as parents being invested in their children) is overall a better society to live in.
This is not because children are more morally valuable than adults, but because the side-effects of behaving that way create a society that is better.
This should mean that parents will invest a lot of time and effort into their children.
But this is a general disposition. It does not mean, in a specific life-or-death situation, that we should ignore the fact that there’s a big difference in infants and adults. If we are to accept “capacity to see oneself as a distinct entity with a past and future” as a moral characteristic that should override proximity-based characteristics, then it seems internally consistent to favor one’s own child in such a situation.
Favoring family even in life or death
We might say: “Favoring family even in life-or-death situations leads to better overall outcomes”.
I personally agree, but then that seems inconsistent, or, at least, selective.
We want equal consideration of interests, but then there’s a particular place that we carve out where equal consideration of interests doesn’t apply as the relevant framework.
Moreover, if we favor family in life and death, family being just one – though very strong – marker of proximity, then that would justify favoring along any other dimension: race, nationality, gender – all things explicitly rejected by Singer as irrelevant moral characteristics.
Where is the boundary between:
“If everyone saves a member of their own family from a fire, even though there’s someone else who deserves help more, that leads to a better overall outcome for society.”
and
“If everyone saves a member of their own race from a fire, even though there’s someone else who deserves help more, that leads to a better overall outcome for society.”
?
One we favor as proper and good; the other is racism.
You could say that family is a “real” relationship; there’s direct care, you have obligations because your child depends on you, and unlike race or religion, it’s not an arbitrary category. But what if the burning building has your cousin that you know nothing about, don’t have any relationship with, and who is effectively a stranger to you?
Even in that case, most people’s moral intuition is to save the cousin, because he is blood.
If we would admit that saving a cousin you know nothing about purely because of genetic proximity is legitimate, than saving based on race is a matter of degree and not category. And saving based on other proximity factors (for example, belonging to the same tribe, or religion) then becomes acceptable too.
Questioning Singer’s theory on its own grounds
Let us assume that to satisfy (the extension of) Singer’s moral framework, we must sacrifice our own child (or baby brother) to save a stranger. Singer’s other argument is that you should keep giving until you reach a point where you start impoverishing yourself.
In that case, Singer’s argument for giving until you go just above poverty falls apart, because why stop at poverty?
Your child is proximate to you: that itself gives it no stronger claim to life. You yourself are even more proximate to yourself.
Therefore, by the same utilitarian calculus by which I should let my child perish in the fire, I should always sacrifice my own life if at least two lives are saved by my sacrifice.
Giving financially saves lives. The difference between giving money and sacrificing your life is a difference of degree: in both cases you are giving something of yourself, your accumulated capacity for change, your “life-force”.
Therefore, whenever I can give money such that I can save at least two lives, I should give that money even if I go into poverty or die.
The argument is that much stronger insomuch as the fact that my giving will almost definitely save more than two lives – cancelling out any objections that I might be killing myself for producing roughly an equal moral outcome.
Therefore, Singer’s argument that we should stop giving someplace where we start entering into poverty picks an arbitrary point. Internally, it favors the survival of the person giving the money.
But if we should be ready to discard the familial obligation to save the life of our not-yet-person child, then we should equally be ready to discard any “familial” obligation to save our own life.
Addressing potential utility generation
You could argue that by continuing living, you could produce more utility overall, and therefore killing yourself to save more people is net harmful, given the fact that you could save much more people in the long run.
But there are two issues here.
One, if we are to keep the internal consistency of the argument, then we should not treat potential utility generation any more favorably than treating potential personhood.
Since Singer claims that potential personhood is not as morally relevant as real personhood, we cannot justify a different treatment for potential utility generation vs. real utility generation.
If we should be ready to sacrifice our potential-person child, then we should be ready to sacrifice our potential future giving.
Two, if we argue for our continued survival on the grounds that we might generate more utility by living longer, that line of argument can extend arbitrarily and we can by the same token argue that we should not so much that it brings us just above the line of poverty, because keeping more money will allow us to live better, potentially generate more money, and therefore generate more utility.
In other words, it proves too much.
Burning building 2
I want to shortly reflect on the burning building thought experiment I introduced.
I would argue that if you rush into the burning building, and see either an infant or adult, both strangers to you, most people’s moral intuition would be to save the infant.
It certainly feels morally correct to me to save a stranger’s baby.
If the choice is between “adult person I know or love” and “stranger’s baby”, that choice is perhaps the most difficult of all. And I am not entirely sure I would pick the adult.
It seems that my moral intuitions are primarily shaped by the maxim of “the strong should protect the weak”. There’s a European moral lineage of chivalry – the notion that you should help those who are helpless, save those who are oppressed, and otherwise seek to be a hero.
Intuitively, morally, I sense that as the right thing to do.
And I would argue that, even on purely consequentialist grounds, being of that particular moral disposition produces overall better outcomes for society.
Disclaimers
This essay is me trying to figure out the “edges” of Singer’s argument in Practical Ethics.
I’ve written and rewritten it several times, and it bothers me that I don’t reach a particular conclusion. The essay itself remains at the level of “musings” instead of “worked out, internally consistent philosophical refutation”.
Nevertheless, I want to share my thoughts, so publishing it anyway.
Some specific disclaimers:
Intro to Practical Ethics
If you’ve read the book, or are otherwise familiar with its arguments, feel free to skip to the next chapter.
Singer claims that you must make ethical decisions based on an equal consideration of interests, and not any other property.
It does not matter what age, race, religion, sex, or species one is – the only thing that matters is one’s capacity to suffer, and one’s capacity to view oneself as a distinct entity, with a past and a future.
Take, for example, eating meat.
It is the human’s interest to feel pleasure from eating a tasty steak. It is the cow’s interest to not be killed.
According to the principle of equal consideration of interests, the cow’s interest to not be killed (nor exposed to factory farming practices) clearly outweighs the human’s interest in eating tasty meat.
There is also a moral ranking here that is based on how refined one’s capacity to suffer is. For example, humans are both sentient and capable of seeing themselves as distinct entities existing over time. Cows are merely sentient.
But if there are some humans who are not sentient nor capable of seeing themselves as distinct entities existing over time (for example, patients in a permanent vegetative state), then they have a lower moral footprint than a sentient cow. The cow still cannot conceive of itself as existing over time (probably), but it can experience suffering, which is more than such a human can.
Therefore, in that case, a cow has a higher moral status, and it would be more wrong to kill that cow than that human.
(Singer explores some edge cases, implications on others and on societal norms; I’m shortening the argument here.)
General moral argument against proximity
Singer claims that proximity is not adequate for moral judgment. If we generalize his argument beyond species, race, religion, nationality, to all markers of proximity, we must come to the conclusion that family is equally excluded from moral protection.
My family members are proximate to me in the sense that we have similar genes, and in the sense that we are one tightly-knit group, irrespective of genes (for example, families with adopted children).
Singer claims that genetic proximity is not a relevant moral factor – he rejects preferential treatment based on species, or race. Therefore, if I extend that line of argument, I cannot provide preferential moral treatment to my family based on their genes.
He also claims that other proximity which is not genetic – such as similarity of religion, or nationality – is equally not a relevant moral factor. Therefore, if I extend that line of argument, I also cannot provide preferential moral treatment to my family based on us being the same group.
Therefore, we must either:
Thought experiment: burning building
Singer also claims that infants do not have the same moral status as adults. They have no conception of themselves as “a distinct entity existing over time”. They have potential personhood, but Singer claims that potential personhood is not as strong of a moral claim as real personhood.
Here’s a thought experiment:
You apartment building is on fire. You rush in. There’s time to save exactly one person: your 6-month-old baby, or an adult stranger.
If we must not give preferential moral treatment based on proximity, and if infants do not yet possess morally relevant characteristics, then the moral thing to do would be to let your child die in the fire, and save the stranger.
I believe that every moral framework that would have you let your child die so that you can save a stranger’s life is wrong. It must have gotten lost along the way somehow, and it is our task now to find where exactly this framework has gotten lost.
I do not believe that infants actually have the morally relevant characteristics that adults have. And I similarly agree with the premise that future personhood is not as strong a claim to moral status as current personhood.
No, the reason why you should save you child, is that it’s your child, which means that I reject the argument against proximity.
Addressing “roles and expectations”-based counterarguments
A counterargument might be: “you have chosen to have this child and therefore you have a moral obligation to it; it’s different from arbitrary things like nationality or religion.”
We can change the thought experiment to not have your own child in the fire, but your baby brother.
In that case, there is no choice that was made, and you have entered no “contract” that forms a moral obligation of care towards this being; it’s a genetic accident that you had no influence on.
Yet, I argue, it would entail the same effect: if you rush into the building, you should most definitely save your baby brother, and not an adult stranger.
Addressing “favoring family leads to better overall outcomes”
Singer claims that, in aggregate, a society where one is more favorably disposed to one’s family (such as parents being invested in their children) is overall a better society to live in.
This is not because children are more morally valuable than adults, but because the side-effects of behaving that way create a society that is better.
This should mean that parents will invest a lot of time and effort into their children.
But this is a general disposition. It does not mean, in a specific life-or-death situation, that we should ignore the fact that there’s a big difference in infants and adults. If we are to accept “capacity to see oneself as a distinct entity with a past and future” as a moral characteristic that should override proximity-based characteristics, then it seems internally consistent to favor one’s own child in such a situation.
Favoring family even in life or death
We might say: “Favoring family even in life-or-death situations leads to better overall outcomes”.
I personally agree, but then that seems inconsistent, or, at least, selective.
We want equal consideration of interests, but then there’s a particular place that we carve out where equal consideration of interests doesn’t apply as the relevant framework.
Moreover, if we favor family in life and death, family being just one – though very strong – marker of proximity, then that would justify favoring along any other dimension: race, nationality, gender – all things explicitly rejected by Singer as irrelevant moral characteristics.
Where is the boundary between:
“If everyone saves a member of their own family from a fire, even though there’s someone else who deserves help more, that leads to a better overall outcome for society.”
and
“If everyone saves a member of their own race from a fire, even though there’s someone else who deserves help more, that leads to a better overall outcome for society.”
?
One we favor as proper and good; the other is racism.
You could say that family is a “real” relationship; there’s direct care, you have obligations because your child depends on you, and unlike race or religion, it’s not an arbitrary category. But what if the burning building has your cousin that you know nothing about, don’t have any relationship with, and who is effectively a stranger to you?
Even in that case, most people’s moral intuition is to save the cousin, because he is blood.
If we would admit that saving a cousin you know nothing about purely because of genetic proximity is legitimate, than saving based on race is a matter of degree and not category. And saving based on other proximity factors (for example, belonging to the same tribe, or religion) then becomes acceptable too.
Questioning Singer’s theory on its own grounds
Let us assume that to satisfy (the extension of) Singer’s moral framework, we must sacrifice our own child (or baby brother) to save a stranger. Singer’s other argument is that you should keep giving until you reach a point where you start impoverishing yourself.
In that case, Singer’s argument for giving until you go just above poverty falls apart, because why stop at poverty?
Your child is proximate to you: that itself gives it no stronger claim to life. You yourself are even more proximate to yourself.
Therefore, by the same utilitarian calculus by which I should let my child perish in the fire, I should always sacrifice my own life if at least two lives are saved by my sacrifice.
Giving financially saves lives. The difference between giving money and sacrificing your life is a difference of degree: in both cases you are giving something of yourself, your accumulated capacity for change, your “life-force”.
Therefore, whenever I can give money such that I can save at least two lives, I should give that money even if I go into poverty or die.
The argument is that much stronger insomuch as the fact that my giving will almost definitely save more than two lives – cancelling out any objections that I might be killing myself for producing roughly an equal moral outcome.
Therefore, Singer’s argument that we should stop giving someplace where we start entering into poverty picks an arbitrary point. Internally, it favors the survival of the person giving the money.
But if we should be ready to discard the familial obligation to save the life of our not-yet-person child, then we should equally be ready to discard any “familial” obligation to save our own life.
Addressing potential utility generation
You could argue that by continuing living, you could produce more utility overall, and therefore killing yourself to save more people is net harmful, given the fact that you could save much more people in the long run.
But there are two issues here.
One, if we are to keep the internal consistency of the argument, then we should not treat potential utility generation any more favorably than treating potential personhood.
Since Singer claims that potential personhood is not as morally relevant as real personhood, we cannot justify a different treatment for potential utility generation vs. real utility generation.
If we should be ready to sacrifice our potential-person child, then we should be ready to sacrifice our potential future giving.
Two, if we argue for our continued survival on the grounds that we might generate more utility by living longer, that line of argument can extend arbitrarily and we can by the same token argue that we should not so much that it brings us just above the line of poverty, because keeping more money will allow us to live better, potentially generate more money, and therefore generate more utility.
In other words, it proves too much.
Burning building 2
I want to shortly reflect on the burning building thought experiment I introduced.
I would argue that if you rush into the burning building, and see either an infant or adult, both strangers to you, most people’s moral intuition would be to save the infant.
It certainly feels morally correct to me to save a stranger’s baby.
If the choice is between “adult person I know or love” and “stranger’s baby”, that choice is perhaps the most difficult of all. And I am not entirely sure I would pick the adult.
It seems that my moral intuitions are primarily shaped by the maxim of “the strong should protect the weak”. There’s a European moral lineage of chivalry – the notion that you should help those who are helpless, save those who are oppressed, and otherwise seek to be a hero.
Intuitively, morally, I sense that as the right thing to do.
And I would argue that, even on purely consequentialist grounds, being of that particular moral disposition produces overall better outcomes for society.