In the interest of precision, I think it is good to be “outrageous” in the sense that, if you think something is true, it is often fine to say it even if it’s unpopular. But I think a tone of outrage can often be counterproductive, both in (1) alienating audiences and (2) causing frustration, which can cloud one’s judgment. (1), for example, is why I think in many contexts, it might be better to frame EA as an opportunity rather than an obligation (certainly not all – I suspect the “drowning child” thought experiment, for instance, has been very persuasive overall!).
"I think in many contexts, it might be better to frame EA as an opportunity rather than an obligation (certainly not all – I suspect the “drowning child” thought experiment, for instance, has been very persuasive overall!)."
First, I was referring to the “drowning child” argument in terms of its persuasiveness, rather than truth (even though I think it’s plausibly true, as I’ll explain in a moment).
Second, believing EA is an obligation, in terms of actual truth, doesn’t require accepting the drowning child argument. “Doing a reasonable amount of good is a moral obligation people have” seems like common sense, and most effective altruists don’t do all they could to be good. See Richard Chappell here (https://www.goodthoughts.blog/p/beneficentrism).
Third, I think I disagree with most of your objections:
(1) “It cannot be philosophically relevant whether a principle is uncontroversial” seems wrong to me. I agree that it being uncontroversial does not entail its truth, but it is probably evidence of its truth, as a widely-held intuition is probably more reliable than a less widely-held one.
(2) You argue that it seems wrong that people have an obligation to engage in such self-sacrifice that we are in almost as bad a condition ourselves. I will note that Singer‘s argument, as presented, does not suggest this. Even if you take the fully utilitarian stance that you ought to give until the “point of marginal utility,” as Singer calls it, that’s probably still much better as a life than the life of people in extreme poverty, because if you‘re suffering a lot, you‘re probably not going to be earning as much money or as good at your job, which reduces the amount of good that you do. But Singer’s paper doesn’t even defend reaching that point. It seems like lots of things might be of comparable moral importance. “Comparable” doesn’t mean “almost exactly the same.” But also, I’m not sure your reductio does work. It seems plausible to me that, in a world where some people are lucky while others are in immense suffering, there is a duty, that goes further than just supererogation, to engage in a lot of self-sacrifice. I’d also note that this intuition you point to is plausibly unreliable, since people, in general, have status quo bias, bias in favor of believing they’re not behaving immorally, and biases in favor of helping their own interests, which might cloud their judgment in arriving at this intuition.
(3) You say that the obligation to save a drowning child is explained by a “local contract” by occupying or voluntarily entering a neighborhood. I don’t find this plausible for two reasons. First, this is an obligation that would apply in any neighborhood you choose to enter, meaning there’s no neighborhood where you can be free of this particular stipulation in the hypothetical contract. This, then, runs into the same problem as most social contract arguments that, say, try to justify the state: the contract wasn’t really signed by you voluntarily, as even if you voluntarily entered that particular neighborhood, it’s a rule that exists in any neighborhood you step into, so you didn’t consent to the rule. Second, a neighborhood seems like an arbitrary geographical boundary. You could draw the boundary anywhere, including the “world.” To illustrate why it’s arbitrary, imagine you have a particular boundary for a neighborhood. There must be some point in the neighborhood that is very close to and visible from a different neighborhood. Does that mean it’s fine to walk past a drowning child a few meters away as long as they’re not “in the neighborhood”? But maybe you instead define a neighborhood according to some radius of distance from you. But I still think that seems arbitrary.
(4) You say that the rules would never “include an obligation to assist people in an area of general and sustained emergency, such as a famine or deadly disease; as that would keep people away and result in less assistance.” I think this conflates the truth of what people have a moral duty to do and how we should, in practical terms, convey that message and create social norms. I’m all for a social norm that asks for much less than the standard implied by the “drowning child” argument (e.g., the Giving What We Can pledge settles at 10% of your income). I think that is a separate question from what the actual truth of the matter is.
(5) On the supposed paradox, I’m not really bothered by this. In general, I think whether we judge a person to be a “good person” or a “bad person” is not a question of moral facts, but rather of social engineering. I think there’s a fact of the matter to whether an action is right or wrong, but no fact of the matter to whether a person is a “good person” or “bad person” or obvious way to define it: it seems like a social engineering question. For example, a utilitarian doesn’t have an easy way to balance right or wrong actions in a single life against each other, as figuring out if a person is “good” requires considering other questions, like whether they have good intent. Mostly, I just don’t think there is a moral fact for whether a person is good, and I don’t think people should be satisfied by being “extremely moral”: doing immoral things is bad regardless of your current “endowment” of overall good-ness. Here’s one argument for why my system is preferable: your paradox applies to calling *any* omission immoral at all, because there’s just way more metaphysically possible things someone could do that are bad for other people than good, so if you accept that it is ever possible to judge an omission as right or wrong – even if you accept a doing/allowing distinction – then you run into your paradox. It seems like a better solution than “rejecting evaluating omissions entirely” is “recognize that judging whether an action is right or wrong is a clearer problem than judging whether a person is good or bad.” Example: seems like a rich person donating $100 is less virtuous than a poor person donating $100, even if they end up having the same outcome. Most utilitarians would just say whether a person is “virtuous” isn’t really a moral fact in quite the same way.
(6) I don’t see why free markets being good refutes the argument. It seems plausible to me that trying to campaign for free trade is an important and just moral cause. If your argument is true, Singer would just say “cool, so then, you have an obligation to donate to advocacy organizations that might work at removing trade restrictions.” I think you’re not thinking at the margin: we live in a world with unfree trade and substantial poverty, and no individual can make it all on their own, so it seems like there’s a role for helping others voluntarily anyway.
As a general point, I think there’s a difference between personal morality and whether coercion to make you fulfill your personal obligations is justified: It might be wrong for governments or neighborhoods to force people to meet their moral duties, but they might still have a moral duty. So I think it‘s possible to be a libertarian and still accept Singer’s argument.
Thanks for the detailed responses, and apologies for the extreme delay.
If the “drowning child” argument had “persuasiveness” (interpreted here as persuading more readers than not), then presumably the persuaded people would adjust their behaviour accordingly and embrace Effective Altruism (EA). Rather, most readers accept the moral obligation with a drowning child in front of them, but they don’t accept that this extends to allegedly analogous situations worldwide in the way that the Singer article argues. They can’t say exactly what’s wrong with the argument, but it manifestly doesn’t persuade them.
It is not credible that “Doing a reasonable amount of good is a moral obligation people have” is “common sense”. All that most people relevantly do is give very modest donations to charity every now and then. And it is not likely that most of them even regard this as a “moral obligation”.
“Beneficentrism: The view that promoting the general welfare is deeply important, and should be amongst one’s central life projects. … Beneficentrism strikes me as impossible to deny while retaining basic moral decency.”
The general welfare is deeply important. But it doesn’t follow that intentionally “promoting the general welfare is deeply important”, let alone that it “should be amongst one’s central life projects”. The invisible hand of the free market tends to promote it quite efficiently, although some additional charity may do more good than harm. As for “basic moral decency”, that would mainly seem to require people to refrain from committing crimes against people and their legitimate property. If we are to add anything to this, then maybe it is that we should—unless it is too onerous for us—give any necessary immediate assistance to people in our immediate vicinity if they are in obvious difficulties (and no one else is already assisting them). It is far beyond “basic moral decency” that we should be “promoting the general welfare”—let alone of the entire world.
(1) In philosophy, which is what the Singer article is engaged in, an appeal to what is “uncontroversial” is at least dubious; if not a logical fallacy. It seems that at least part of the purpose of philosophy is to challenge what is “uncontroversial”. If we are in a hurry to make some decision, then we might prefer to opt for what is “uncontroversial” or a “widely-held intuition”. But then we are not engaged in doing philosophy. But that remark was, in any case, an aside.
(2) Singer’s article says that we are to “prevent what is bad … when we can do it without sacrificing anything that is, from the moral point of view, comparably important”. It is hard to see how people can avoid self-sacrifice on this principle. If a relatively well-off person buys anything whereby the price of it could have done more good if sent to a charity for those in dire need, then he would appear to be flouting that principle.
On the contrary, “comparably important” morally, does appear to mean “almost exactly the same” level of moral importance.
The word “supererogation” just means going beyond what duty requires. So, “a duty, that goes further than just supererogation” is incoherent as it stands.
As there is no quotation provided, it is not clear to what “this intuition you point to” refers.
(3) First, there seems to be no inherent problem with the idea of a world of neighbourhoods with contractual obligations to help in emergencies if one enters those neighbourhoods. That does not make any such contract nonconsensual. But, of course, the world is not composed entirely of such neighbourhoods.
Second, what is conventional is not thereby necessarily “arbitrary”. But “neighbourhoods” in the relevant sense should be defined by private property. Therefore, there will be a property line that one crosses and thereby enters a different neighbourhood.
(4) It is the impracticality of a particularly onerous “obligation to assist” contractual rule that is what makes such a rule unlikely. This impracticality point has nothing to do with “the truth of what people have a moral duty to do and how we should, in practical terms, convey that message and create social norms”. Therefore, it can’t “conflate” them.
(5) The paradox is not about morally judging people or what they ought to be called. It is about the incoherence of the Singer article’s biased account of morality: only taking into account alleged obligations to do good things while completely ignoring logically parallel obligations not to do bad things.
The long sentence beginning “Here’s one argument” is too unclear to make definite sense of it. So is the rest of the paragraph. It does not help that non-quotations are often put within quotation marks. Some of them are apparently intended to be paraphrases; but they are not accurate paraphrases.
(6) The efficiency of free markets in promoting welfare refutes the Singer article’s implied idea that the main solution to “major evils” is, what is now called, EA.
To make it a point of criticism that, “it seems like there’s a role for helping others voluntarily anyway”, ignores the reply to the Singer article where it is also explicitly affirmed that, “Into the foreseeable future there will always be room for charity that can do real good around the world”.
Libertarianism usually involves affirming that it is always moral for people to do whatever they like with their own person and property (as long as this does not infringe the similar liberty of other people, of course). It appears to be inconsistent with this to assert in addition, ‘But if they don’t practice EA, then they are immoral’. Therefore, it doesn’t seem “possible to be a libertarian and still accept Singer’s argument”. But then the Singer article is inherently paradoxical for everyone, not just for libertarians.
This is important and true! But the tricky question is whether to say important and true things in a way that is *gratuitously* outrageous, to bait more engagement.
For example, in my post on 'genetic reproductive freedom' - https://www.goodthoughts.blog/p/genetic-reproductive-freedom - I was careful to distinguish technologically-aided genetic reproductive *freedom* with coercive eugenics (i.e., technologically-aided reproductive *coercion*). It would have been more provocative to frame the post as "(liberal) eugenics is good!" Maybe that would have prompted more engagement (as hate bait). But that would come at the cost of harming understanding. And I think there are generally good reasons to prioritize promoting better understanding over mere engagement.
That said, there can be real tradeoffs here, and *some* degree of provocative framing for important issues can often be reasonable to help draw attention to them, even if it results in some extra misunderstanding. Maybe we should aim for whatever maximizes "importance-weighted understanding" in one's audience? (As per https://www.goodthoughts.blog/p/moral-misdirection )
Could be fun to explore average vs total views of epistemic do-gooding. (Seems like totalism has got to be the right view in this domain: it's clearer better to have a positive epistemic effect on a larger audience!)
In the interest of precision, I think it is good to be “outrageous” in the sense that, if you think something is true, it is often fine to say it even if it’s unpopular. But I think a tone of outrage can often be counterproductive, both in (1) alienating audiences and (2) causing frustration, which can cloud one’s judgment. (1), for example, is why I think in many contexts, it might be better to frame EA as an opportunity rather than an obligation (certainly not all – I suspect the “drowning child” thought experiment, for instance, has been very persuasive overall!).
"I think in many contexts, it might be better to frame EA as an opportunity rather than an obligation (certainly not all – I suspect the “drowning child” thought experiment, for instance, has been very persuasive overall!)."
It is only an opportunity, at best. The "drowning child" thought experiment is hopelessly confused: https://jclester.substack.com/p/peter-singers-famine-affluence-and
Thanks for the reply! I’d make three notes.
First, I was referring to the “drowning child” argument in terms of its persuasiveness, rather than truth (even though I think it’s plausibly true, as I’ll explain in a moment).
Second, believing EA is an obligation, in terms of actual truth, doesn’t require accepting the drowning child argument. “Doing a reasonable amount of good is a moral obligation people have” seems like common sense, and most effective altruists don’t do all they could to be good. See Richard Chappell here (https://www.goodthoughts.blog/p/beneficentrism).
Third, I think I disagree with most of your objections:
(1) “It cannot be philosophically relevant whether a principle is uncontroversial” seems wrong to me. I agree that it being uncontroversial does not entail its truth, but it is probably evidence of its truth, as a widely-held intuition is probably more reliable than a less widely-held one.
(2) You argue that it seems wrong that people have an obligation to engage in such self-sacrifice that we are in almost as bad a condition ourselves. I will note that Singer‘s argument, as presented, does not suggest this. Even if you take the fully utilitarian stance that you ought to give until the “point of marginal utility,” as Singer calls it, that’s probably still much better as a life than the life of people in extreme poverty, because if you‘re suffering a lot, you‘re probably not going to be earning as much money or as good at your job, which reduces the amount of good that you do. But Singer’s paper doesn’t even defend reaching that point. It seems like lots of things might be of comparable moral importance. “Comparable” doesn’t mean “almost exactly the same.” But also, I’m not sure your reductio does work. It seems plausible to me that, in a world where some people are lucky while others are in immense suffering, there is a duty, that goes further than just supererogation, to engage in a lot of self-sacrifice. I’d also note that this intuition you point to is plausibly unreliable, since people, in general, have status quo bias, bias in favor of believing they’re not behaving immorally, and biases in favor of helping their own interests, which might cloud their judgment in arriving at this intuition.
(3) You say that the obligation to save a drowning child is explained by a “local contract” by occupying or voluntarily entering a neighborhood. I don’t find this plausible for two reasons. First, this is an obligation that would apply in any neighborhood you choose to enter, meaning there’s no neighborhood where you can be free of this particular stipulation in the hypothetical contract. This, then, runs into the same problem as most social contract arguments that, say, try to justify the state: the contract wasn’t really signed by you voluntarily, as even if you voluntarily entered that particular neighborhood, it’s a rule that exists in any neighborhood you step into, so you didn’t consent to the rule. Second, a neighborhood seems like an arbitrary geographical boundary. You could draw the boundary anywhere, including the “world.” To illustrate why it’s arbitrary, imagine you have a particular boundary for a neighborhood. There must be some point in the neighborhood that is very close to and visible from a different neighborhood. Does that mean it’s fine to walk past a drowning child a few meters away as long as they’re not “in the neighborhood”? But maybe you instead define a neighborhood according to some radius of distance from you. But I still think that seems arbitrary.
(4) You say that the rules would never “include an obligation to assist people in an area of general and sustained emergency, such as a famine or deadly disease; as that would keep people away and result in less assistance.” I think this conflates the truth of what people have a moral duty to do and how we should, in practical terms, convey that message and create social norms. I’m all for a social norm that asks for much less than the standard implied by the “drowning child” argument (e.g., the Giving What We Can pledge settles at 10% of your income). I think that is a separate question from what the actual truth of the matter is.
(5) On the supposed paradox, I’m not really bothered by this. In general, I think whether we judge a person to be a “good person” or a “bad person” is not a question of moral facts, but rather of social engineering. I think there’s a fact of the matter to whether an action is right or wrong, but no fact of the matter to whether a person is a “good person” or “bad person” or obvious way to define it: it seems like a social engineering question. For example, a utilitarian doesn’t have an easy way to balance right or wrong actions in a single life against each other, as figuring out if a person is “good” requires considering other questions, like whether they have good intent. Mostly, I just don’t think there is a moral fact for whether a person is good, and I don’t think people should be satisfied by being “extremely moral”: doing immoral things is bad regardless of your current “endowment” of overall good-ness. Here’s one argument for why my system is preferable: your paradox applies to calling *any* omission immoral at all, because there’s just way more metaphysically possible things someone could do that are bad for other people than good, so if you accept that it is ever possible to judge an omission as right or wrong – even if you accept a doing/allowing distinction – then you run into your paradox. It seems like a better solution than “rejecting evaluating omissions entirely” is “recognize that judging whether an action is right or wrong is a clearer problem than judging whether a person is good or bad.” Example: seems like a rich person donating $100 is less virtuous than a poor person donating $100, even if they end up having the same outcome. Most utilitarians would just say whether a person is “virtuous” isn’t really a moral fact in quite the same way.
(6) I don’t see why free markets being good refutes the argument. It seems plausible to me that trying to campaign for free trade is an important and just moral cause. If your argument is true, Singer would just say “cool, so then, you have an obligation to donate to advocacy organizations that might work at removing trade restrictions.” I think you’re not thinking at the margin: we live in a world with unfree trade and substantial poverty, and no individual can make it all on their own, so it seems like there’s a role for helping others voluntarily anyway.
As a general point, I think there’s a difference between personal morality and whether coercion to make you fulfill your personal obligations is justified: It might be wrong for governments or neighborhoods to force people to meet their moral duties, but they might still have a moral duty. So I think it‘s possible to be a libertarian and still accept Singer’s argument.
Thanks for the detailed responses, and apologies for the extreme delay.
If the “drowning child” argument had “persuasiveness” (interpreted here as persuading more readers than not), then presumably the persuaded people would adjust their behaviour accordingly and embrace Effective Altruism (EA). Rather, most readers accept the moral obligation with a drowning child in front of them, but they don’t accept that this extends to allegedly analogous situations worldwide in the way that the Singer article argues. They can’t say exactly what’s wrong with the argument, but it manifestly doesn’t persuade them.
It is not credible that “Doing a reasonable amount of good is a moral obligation people have” is “common sense”. All that most people relevantly do is give very modest donations to charity every now and then. And it is not likely that most of them even regard this as a “moral obligation”.
“Beneficentrism: The view that promoting the general welfare is deeply important, and should be amongst one’s central life projects. … Beneficentrism strikes me as impossible to deny while retaining basic moral decency.”
The general welfare is deeply important. But it doesn’t follow that intentionally “promoting the general welfare is deeply important”, let alone that it “should be amongst one’s central life projects”. The invisible hand of the free market tends to promote it quite efficiently, although some additional charity may do more good than harm. As for “basic moral decency”, that would mainly seem to require people to refrain from committing crimes against people and their legitimate property. If we are to add anything to this, then maybe it is that we should—unless it is too onerous for us—give any necessary immediate assistance to people in our immediate vicinity if they are in obvious difficulties (and no one else is already assisting them). It is far beyond “basic moral decency” that we should be “promoting the general welfare”—let alone of the entire world.
(1) In philosophy, which is what the Singer article is engaged in, an appeal to what is “uncontroversial” is at least dubious; if not a logical fallacy. It seems that at least part of the purpose of philosophy is to challenge what is “uncontroversial”. If we are in a hurry to make some decision, then we might prefer to opt for what is “uncontroversial” or a “widely-held intuition”. But then we are not engaged in doing philosophy. But that remark was, in any case, an aside.
(2) Singer’s article says that we are to “prevent what is bad … when we can do it without sacrificing anything that is, from the moral point of view, comparably important”. It is hard to see how people can avoid self-sacrifice on this principle. If a relatively well-off person buys anything whereby the price of it could have done more good if sent to a charity for those in dire need, then he would appear to be flouting that principle.
On the contrary, “comparably important” morally, does appear to mean “almost exactly the same” level of moral importance.
The word “supererogation” just means going beyond what duty requires. So, “a duty, that goes further than just supererogation” is incoherent as it stands.
As there is no quotation provided, it is not clear to what “this intuition you point to” refers.
(3) First, there seems to be no inherent problem with the idea of a world of neighbourhoods with contractual obligations to help in emergencies if one enters those neighbourhoods. That does not make any such contract nonconsensual. But, of course, the world is not composed entirely of such neighbourhoods.
Second, what is conventional is not thereby necessarily “arbitrary”. But “neighbourhoods” in the relevant sense should be defined by private property. Therefore, there will be a property line that one crosses and thereby enters a different neighbourhood.
(4) It is the impracticality of a particularly onerous “obligation to assist” contractual rule that is what makes such a rule unlikely. This impracticality point has nothing to do with “the truth of what people have a moral duty to do and how we should, in practical terms, convey that message and create social norms”. Therefore, it can’t “conflate” them.
(5) The paradox is not about morally judging people or what they ought to be called. It is about the incoherence of the Singer article’s biased account of morality: only taking into account alleged obligations to do good things while completely ignoring logically parallel obligations not to do bad things.
The long sentence beginning “Here’s one argument” is too unclear to make definite sense of it. So is the rest of the paragraph. It does not help that non-quotations are often put within quotation marks. Some of them are apparently intended to be paraphrases; but they are not accurate paraphrases.
(6) The efficiency of free markets in promoting welfare refutes the Singer article’s implied idea that the main solution to “major evils” is, what is now called, EA.
To make it a point of criticism that, “it seems like there’s a role for helping others voluntarily anyway”, ignores the reply to the Singer article where it is also explicitly affirmed that, “Into the foreseeable future there will always be room for charity that can do real good around the world”.
Libertarianism usually involves affirming that it is always moral for people to do whatever they like with their own person and property (as long as this does not infringe the similar liberty of other people, of course). It appears to be inconsistent with this to assert in addition, ‘But if they don’t practice EA, then they are immoral’. Therefore, it doesn’t seem “possible to be a libertarian and still accept Singer’s argument”. But then the Singer article is inherently paradoxical for everyone, not just for libertarians.
This is important and true! But the tricky question is whether to say important and true things in a way that is *gratuitously* outrageous, to bait more engagement.
For example, in my post on 'genetic reproductive freedom' - https://www.goodthoughts.blog/p/genetic-reproductive-freedom - I was careful to distinguish technologically-aided genetic reproductive *freedom* with coercive eugenics (i.e., technologically-aided reproductive *coercion*). It would have been more provocative to frame the post as "(liberal) eugenics is good!" Maybe that would have prompted more engagement (as hate bait). But that would come at the cost of harming understanding. And I think there are generally good reasons to prioritize promoting better understanding over mere engagement.
That said, there can be real tradeoffs here, and *some* degree of provocative framing for important issues can often be reasonable to help draw attention to them, even if it results in some extra misunderstanding. Maybe we should aim for whatever maximizes "importance-weighted understanding" in one's audience? (As per https://www.goodthoughts.blog/p/moral-misdirection )
Could be fun to explore average vs total views of epistemic do-gooding. (Seems like totalism has got to be the right view in this domain: it's clearer better to have a positive epistemic effect on a larger audience!)