What are your thoughts on pursuing a PhD compared to 3-4 more years of work experience? How much career capital is a formal PhD degree actually worth?
This answer might depend a lot on the context, so:
I have a degree in pure mathematics and that is also where I would potentially pursue a PhD. I have high grades, but not from a top university.
I am living in Germany and am willing, but not eager, to move to another country in order to maximise my impact.
I am considering Earning to Give as my most impactful option (though I am far from certain yet). If I do not pursue a PhD, my next step would most likely be trying to asses my fit for quantitative trading: https://80000hours.org/career-reviews/quantitative-trading/ (To make matters more complicated, it is mentioned that you can enter the field with graduate degrees, but some roles do require a PhD. So if you are knowledgeable in this area, some advice would be especially welcome.)
I have neither a concrete plan nor a scholarship for my potential PhD (yet) and would change my university anyways. So it might also be a consideration to try some jobs and internships and pursue a PhD later on if worthwhile. However I am under the impression that returning into academia is more difficult than staying there.
These are valid concerns and I wouldn't really argue against them. I at least partially share your scepticism about meaningful longterm impacts, after all. (Though if anything, then preventing our extinction in the next centuries seems like the most promising candidate for longterm change - but of course it is also questionable wether we can actually predictably decrease the X-risk probability within this timeframe, let alone on a larger scale.)
I guess I am just very pessimistic about our ability (and willingness) to do something meaningful about wild animal welfare (especially for insects) within the next few (hundred) years. Aside of in- or decreasing their numbers, which is not a consideration, as long as we don't even know (with some certainty) if their lifes are net-positive or net-negative. So it seems to me as if the only possibility for a substantial change with a predictable outcome in the area of wild animal suffering/welfare might happen only if we survive at least a few hundred more years and have 1. the time to investigate the most crucial questions and 2. a shift within political/ethical considerations which would allow to implement larger-scale programs solely for the purposes of wild animal welfare. I have no idea if the second condition will ever be met, though...
Nevertheless, given the massive uncertainty about the solvabilty of wild animal suffering and X-risk prevention, I still think that the fight for farmed animals (especially the more overlooked ones like fish and shrimps) might actually be one or maybe even the top priority right now. At least it might be the most impactful action where we actually know that we have a positive impact at all... (And who knows, maybe if we finally abolish factory farming some day, our view on animals might also change and make the second condition more likely - but I'm just speculating at this point, so I'll leave it at that.)
I couldn't have said it better. I also read the linked paper, and it really resonated with a lot of doubts I already had about the apparent consensus about net-negative insect lifes. So thank you both!
I think it is worth noting that, even given the immense scale of insect welfare/suffering, some longtermist projects (especially those trying to prevent X-risks) might still be extremely important - not because humanities continued existence is so important in and of itself, but because humanity might have an enormous instrumental value for wild animal suffering. After all, if evolution (or AI) does not give rise to another lifeform with our means of altering the planet (and maybe beyond) before the sun runs out, we are the only ones who could actively change the insect's lifes for the better. Of course, we might also make it worse (for example by spreading them to other planets if they do have net-negative lifes after all), so I guess this thought depends on the optimistic assumption that we either will actually care about (and for) them at some point or that our interests coincide with theirs (which seems a lot more plausible if they have net-positive lifes).
Thank you for the post!
I thought about some of these questions myself. Years ago even, while I was turning into a vegetarian.
"If we would not eat those animals, they wouldn't exist at all" is sometimes used as a justification for consuming animal products and if the animals in question do in fact have lifes worth living then it is quite a compelling argument. (At least for me. Substantial parts of the animal rights movement would probably disagree.) So I definitley understand why you write it might actually be morally permissible or even obligatory to eat animal products of animals living net-positive lifes.
Still, after some deliberation I came to the conclusion that this is not the case and that we should still try to end factory farming altogether, even in the variations where the animals have lifes worth living. I think it might be worthwile to share my thinking here and I'm also curious about your thoughts.
So first of all when considering our behaviour towards non-human animals (I will from now on write animals for short) I often find it useful to ask myself a simple question: What would I think if they were humans instead? I do this to get a better intuition, since our moral intuitions are mostly calibrated to work with humans instead of other animals. (Of course intuition often goes wrong, but while I don't put too much credibility on my intuitions I still find them useful as a first orientation.)
If I would find something acceptable even when done to humans then I normally agree with doing it to animals as well. When I would NOT allow this for humans, I see if I can find reasons why humans are so fundamentally different from other animals in regards to this question that different behaviour is warranted. For example, it is easy to see why a right to democratic participation and education it not something that should apply to animals the same way it applies to humans. Maybe the same can be said about some sort of important medical research, though I'm very uncertain in this area. And thinking about humans instead of animals makes it intuitively clear to me that wildlife population control shouldn't be done by just killing animals whenever we think there are too many of them. (Though admittedly we don't have perfect alternative solutions yet, so I'm interested where wild animal welfare might lead to eventually.)
So when I stumbled upon aforementioned justification, I applied my method here as well. You may skip this paragraph if you find it unsettling, as it illuminates how and why I came to find my arguments but not the arguments themselves.
With that said, suppose we had the power to bring into existence some humans whom we would keep in captivation and eventually kill for profit. (Since we are not normally interested in human flesh, you may assume it is done in order to harvest their organs.) Since humans may be bothered by captivation much more than some of the animals (though this is just my personal guess), let's also assume they just don't really know or care about them being held in some sort of (outside?) prison, or we don't hold them captive at all. We somehow make sure these humans live lifes that are essentially good. They are held warm, get enough healthy food, they can enjoy going outside, live in small communities without predators, and so on. We essentially grant them the lifes of hunter-gatherers (which is probably the "bar to cross" in regards to humans?) with some limitations and a lot of upsides. Or maybe even the amenities of modern life. In any case, we are nearly certain their lifes are net-positive. (We can even ask them, after all.) We just sometimes take a few of them, kill them and sell their organs - in order to avoid fear and loss, we may always kill a complete group painlessly in their sleep.
Intuitively we probably agree that doing so would be very much wrong. Still, assuming our intuition is correct that leaves the question why. Since these humans live net-positive lifes, it cannot really be out of concern for them, can it?
In the human case the intuition is much more obvious than in the case of animals. Maybe because the case of animals is different - there is ongoing debate if painlessly killing an animal which does not have plans for or even concepts of their own future is permissible. But personally I think killing animals is wrong pretty much for the same reasons killing humans is: we rob them of all the remaining time and happy experiences they might have had. In any case, this possible distinction can not explain my intuition in regards to humans. After all, I would still prefer being born as one of those humans from my thought experiment over not being born at all, yet I am am against creating humans-to-be-slaughtered. Why?
In the end I came up with four reasons to defend my intuition and in turn (mostly) reject the claim that farming happy animals which would otherwise not be born is permissible.
Slippery Slope: From a strictly utilitarian perspective, if killing someone is good or bad depends solely on the outcome. (However, if we give some credibility to moral theories based on rights or at least to moral uncertainty it becomes obvious that regularly killing sentient beings in industrial scale is something ... highly problematic.)
But even as a utilitarian I can agree that behaviour changes thinking which in turn changes behaviour again. (For this reason, in the human case the human rights or the principle of treating everyone the same, no matter how they came into existence, is too important to reject easily...) So if we kill (billions of) animals on a regular basis, this may change our general behaviour towards animals and how we see them in unfavourable directions, even if originally only done out of careful consideration. I am generally sceptical of slippery slope arguments, as they rest heavily on some assumption of human thinking and behaviour and how they change. Still - if I remember correctly - in this case we have some existing studies pointing out that people who just ate beef are less inclined to grant cows sentience, or something similar, so maybe the argument is warranted.
Abuse of power: This might be somewhat connected to 1, but in my opinion still is a different point. If we grant some farmers near absolute power over some animals and if their welfare interests are in opposition to the economic interests of those in power, then I have very little faith in human conscience. Of course, with strict regulations and controls this problem can be weakened substantially. Still, as long as (individual) humans have an interest in getting higher productivity out of their lifestock, abuse will happen - though maybe seldom enough to accept this as a necessary evil.
Still, even if objections 1. and 2. were taken care of in the human case (so neither abuse nor different behaviour towards the remainig humans would occur) my intuitions would remain unchanged. So it seems some deeper reasons have to be at play here. And I think the reasons are the following:
False Dichotomy: This, I think, is in some sense the strongest point. (It is also the one I mention whenever I hear the argument above that "they wouldn't exist otherwise" and I don't have time or opportunity for a detailed discussion.) The basic idea goes like this: If the animals have net-negative lifes we shouldn't bring them into existence at all. If they have net-positive lifes however, we shouldn't just end these lifes preemptivly. (Since if they lived longer, the additional "value" would be even bigger.)
Of course it is better to have a net-positive life then not to be born at all. But it is even better to have a net-positive and not to be killed after some (rather short) time. In fact there is nothing wrong with bringing into existence happy beings, while there is pretty much everything wrong with killing happy beings - no matter why they exist in the first place.
So if we really think it is good to have more happy animals, then the conclusion is not organic farming but to bring into existence even more of these happy animals and stop the farming altogether!
(One might argue that having an animal that lives their full life is just as good as having an animal that lives for one year and is afterwards killed and replaced by a new one. I would still prefer for animals to live their full lifes instead, if only for reason 2 and moral uncertainty - and again my intuition if I were to replace them with humans instead.)
Of course we cannot just bring into existence an unlimited amount of happy animals without creating a lot of problems at the same time, which is why we also need to consider the next point.
Animals are expensive: Animal products are notoriously bad for the environment. To feed animals we grow crops which often could just be eaten directly. I find it hard to believe that the most efficient possibility to spend our (us being humanity) limited resources is to raise a lot of animals with slightly net-positive lifes instead of caring for the already existing problems first. (Including wild animal suffering.) If anything creating more happy beings seems like something that might be done with surplus resources after all our immediate problems are resolved.
There are two conclusions to draw from these thoughts. First, it seems as if most production of animal products should still stop. This is mostly due to reasons 3 and 1, but even if killing the animals is not already immoral for those two reasons it still seems inefficient for reason 4. (However, while keeping factory farming is expensive for humanity, fighting it is sadly enough also expensive for EA. In some sense this could mean that the resources of EA might be better allocated to raising welfare instead of ending factory farming if the latter is too hard/expensive to achieve. Alternative Proteins still seem like a good bet, though, and most of EA's resources within the animal welfare pot already are directed at better welfare regulations anyways, right? In any case, if we adopt stricter welfare regulations this will not only lead to less animal "torturing" but also to increasing prices for animal products, so I suspect that using our resources to promote animal welfare regulations is actually a very good strategy to end factory farming in the long term. Which is also an answer to your last question.)
Second, some sorts of animal farming might actually be permissible: In some instances they do not cost resources but actually gain them - there are areas where farming crops is essentially impossible but animals can be held and eat the grass. Sometimes these animals are even necessary to preserve a natural habitat. And even if our farming methods do take up some resources, if they are not too costly and we also get something out of it, point 4 essentially vanishes and - assuming we have strict welfare regulations and really care for their welfare - my only remaining objection is that it is still not permissible to kill these animals, at least not before they have grown so old and sick that being killed might be in their interest. Of course we wouldn't want to eat meat from them afterwards, but I can see how we might get ethical leather, wool eggs this way.
Thank you so much for this article! (BTW: Are Gary David O’Brien and BrownHairedEevee the same person? If not, also thanks to the latter for sharing it.)
I was wondering myself for years why longtermists seemingly didn´t really account for (non-human) animals, since it seems quite obvious that in a lot (if not most) of the possible futures, humanity will be nothing but an insignificant minority of all the sentient beings. In fact, my personal motivation for caring about X-risks is mostly due to the instrumental value humanity might have in regards to non-human wellbeing.
Well, I still am wondering, since besides of Beckstead´s argument (which is not really about intrinsic values) you also didn´t mention any particuar arguments human-focused longtermists actually made in regards to this. So I´m still curious why this topic gets so little attention, aside of speciesist reasons (which I am hoping are less common among longtermists compared to the rest of society). Maybe part of this is because the argument for longtermism gets even harder if you also want to convince your audience to abolish speciesism at the same time?
In any case, I completely agree it is a mistake to neglect animals in longtermism.
Im less convinced that WAS is actually a net-negative state. (My guess, though I wouldn´t really call it an informed one, would be that in average an animal in the wild lives a moderate-positive life, despite all the horrors. This average probably depends mostly on insects and small animals in the ocean, however, and I´m really not an expert, so I cannot exculde the possibility that my guess is only wishful thinking.) While this discussion in general is probably better lead elsewhere, I would at least note that your "reasonable estimate" about 1-minute-suffering before any death seems like a good estimate about some of the suffering, but is not really a reason to believe it is also a good estimate about the net-value, since it focuses on one particularly bad minute in a far longer life and does not really tell us anything about the expected value of given life as a whole. (Which would probably be highly dependent on the species.)
If WAS were actually net-positive, it would change a lot in your conclusion about preventing the spread of animals outside of earth. (And sadly also about the current influence of humanity on the overall wellbeing.)
However, most of the arguments (and especially the conclusion that it is a mistake to ignore animas in longtermism) are valid regardless, since it is true in any case that humanity might heavily influence the duration, spread and quality of wildlife as well as the exploitation of farmed or engineered animals.
Thanks for citing the survey here, and thank you Joshua for your analysis.
Your post doesn´t seem strange to me at this place; at the very least I can´t find any harm in posting it here. (If someone is more interested in other discussions, they may read the first two lines and then skip it.) The only question would be if this is worth YOUR time, and I am confident you are able to judge this (and you apparently did and found it worth your time).
Since you already delved that deep into the material and since I don´t see myself doing the same, here a question to you (or whoever else feeling inclined to answer):
Were there a significant part of experts who thought that HLMI and/or FAOL are downright impossible (at least with anything resembling our current approaches)? I do hear/read doubts like these sometimes. If so, how were these experts included in the mean, since you can´t just include infinity with non-zero probability without the whole number going up to infinity? (If they even used a mean. "Aggregate Forecast" is not very clear; if they used the median ore something similar the second question can be ignored.)
What are your thoughts on pursuing a PhD compared to 3-4 more years of work experience? How much career capital is a formal PhD degree actually worth?
This answer might depend a lot on the context, so:
These are valid concerns and I wouldn't really argue against them. I at least partially share your scepticism about meaningful longterm impacts, after all. (Though if anything, then preventing our extinction in the next centuries seems like the most promising candidate for longterm change - but of course it is also questionable wether we can actually predictably decrease the X-risk probability within this timeframe, let alone on a larger scale.)
I guess I am just very pessimistic about our ability (and willingness) to do something meaningful about wild animal welfare (especially for insects) within the next few (hundred) years. Aside of in- or decreasing their numbers, which is not a consideration, as long as we don't even know (with some certainty) if their lifes are net-positive or net-negative. So it seems to me as if the only possibility for a substantial change with a predictable outcome in the area of wild animal suffering/welfare might happen only if we survive at least a few hundred more years and have 1. the time to investigate the most crucial questions and 2. a shift within political/ethical considerations which would allow to implement larger-scale programs solely for the purposes of wild animal welfare. I have no idea if the second condition will ever be met, though...
Nevertheless, given the massive uncertainty about the solvabilty of wild animal suffering and X-risk prevention, I still think that the fight for farmed animals (especially the more overlooked ones like fish and shrimps) might actually be one or maybe even the top priority right now. At least it might be the most impactful action where we actually know that we have a positive impact at all... (And who knows, maybe if we finally abolish factory farming some day, our view on animals might also change and make the second condition more likely - but I'm just speculating at this point, so I'll leave it at that.)
I couldn't have said it better. I also read the linked paper, and it really resonated with a lot of doubts I already had about the apparent consensus about net-negative insect lifes. So thank you both!
I think it is worth noting that, even given the immense scale of insect welfare/suffering, some longtermist projects (especially those trying to prevent X-risks) might still be extremely important - not because humanities continued existence is so important in and of itself, but because humanity might have an enormous instrumental value for wild animal suffering. After all, if evolution (or AI) does not give rise to another lifeform with our means of altering the planet (and maybe beyond) before the sun runs out, we are the only ones who could actively change the insect's lifes for the better. Of course, we might also make it worse (for example by spreading them to other planets if they do have net-negative lifes after all), so I guess this thought depends on the optimistic assumption that we either will actually care about (and for) them at some point or that our interests coincide with theirs (which seems a lot more plausible if they have net-positive lifes).
Anyways, thank you both again.
Thank you for the post! I thought about some of these questions myself. Years ago even, while I was turning into a vegetarian.
"If we would not eat those animals, they wouldn't exist at all" is sometimes used as a justification for consuming animal products and if the animals in question do in fact have lifes worth living then it is quite a compelling argument. (At least for me. Substantial parts of the animal rights movement would probably disagree.) So I definitley understand why you write it might actually be morally permissible or even obligatory to eat animal products of animals living net-positive lifes. Still, after some deliberation I came to the conclusion that this is not the case and that we should still try to end factory farming altogether, even in the variations where the animals have lifes worth living. I think it might be worthwile to share my thinking here and I'm also curious about your thoughts.
So first of all when considering our behaviour towards non-human animals (I will from now on write animals for short) I often find it useful to ask myself a simple question: What would I think if they were humans instead? I do this to get a better intuition, since our moral intuitions are mostly calibrated to work with humans instead of other animals. (Of course intuition often goes wrong, but while I don't put too much credibility on my intuitions I still find them useful as a first orientation.) If I would find something acceptable even when done to humans then I normally agree with doing it to animals as well. When I would NOT allow this for humans, I see if I can find reasons why humans are so fundamentally different from other animals in regards to this question that different behaviour is warranted. For example, it is easy to see why a right to democratic participation and education it not something that should apply to animals the same way it applies to humans. Maybe the same can be said about some sort of important medical research, though I'm very uncertain in this area. And thinking about humans instead of animals makes it intuitively clear to me that wildlife population control shouldn't be done by just killing animals whenever we think there are too many of them. (Though admittedly we don't have perfect alternative solutions yet, so I'm interested where wild animal welfare might lead to eventually.)
So when I stumbled upon aforementioned justification, I applied my method here as well. You may skip this paragraph if you find it unsettling, as it illuminates how and why I came to find my arguments but not the arguments themselves. With that said, suppose we had the power to bring into existence some humans whom we would keep in captivation and eventually kill for profit. (Since we are not normally interested in human flesh, you may assume it is done in order to harvest their organs.) Since humans may be bothered by captivation much more than some of the animals (though this is just my personal guess), let's also assume they just don't really know or care about them being held in some sort of (outside?) prison, or we don't hold them captive at all. We somehow make sure these humans live lifes that are essentially good. They are held warm, get enough healthy food, they can enjoy going outside, live in small communities without predators, and so on. We essentially grant them the lifes of hunter-gatherers (which is probably the "bar to cross" in regards to humans?) with some limitations and a lot of upsides. Or maybe even the amenities of modern life. In any case, we are nearly certain their lifes are net-positive. (We can even ask them, after all.) We just sometimes take a few of them, kill them and sell their organs - in order to avoid fear and loss, we may always kill a complete group painlessly in their sleep. Intuitively we probably agree that doing so would be very much wrong. Still, assuming our intuition is correct that leaves the question why. Since these humans live net-positive lifes, it cannot really be out of concern for them, can it? In the human case the intuition is much more obvious than in the case of animals. Maybe because the case of animals is different - there is ongoing debate if painlessly killing an animal which does not have plans for or even concepts of their own future is permissible. But personally I think killing animals is wrong pretty much for the same reasons killing humans is: we rob them of all the remaining time and happy experiences they might have had. In any case, this possible distinction can not explain my intuition in regards to humans. After all, I would still prefer being born as one of those humans from my thought experiment over not being born at all, yet I am am against creating humans-to-be-slaughtered. Why?
In the end I came up with four reasons to defend my intuition and in turn (mostly) reject the claim that farming happy animals which would otherwise not be born is permissible.
Still, even if objections 1. and 2. were taken care of in the human case (so neither abuse nor different behaviour towards the remainig humans would occur) my intuitions would remain unchanged. So it seems some deeper reasons have to be at play here. And I think the reasons are the following:
There are two conclusions to draw from these thoughts. First, it seems as if most production of animal products should still stop. This is mostly due to reasons 3 and 1, but even if killing the animals is not already immoral for those two reasons it still seems inefficient for reason 4. (However, while keeping factory farming is expensive for humanity, fighting it is sadly enough also expensive for EA. In some sense this could mean that the resources of EA might be better allocated to raising welfare instead of ending factory farming if the latter is too hard/expensive to achieve. Alternative Proteins still seem like a good bet, though, and most of EA's resources within the animal welfare pot already are directed at better welfare regulations anyways, right? In any case, if we adopt stricter welfare regulations this will not only lead to less animal "torturing" but also to increasing prices for animal products, so I suspect that using our resources to promote animal welfare regulations is actually a very good strategy to end factory farming in the long term. Which is also an answer to your last question.) Second, some sorts of animal farming might actually be permissible: In some instances they do not cost resources but actually gain them - there are areas where farming crops is essentially impossible but animals can be held and eat the grass. Sometimes these animals are even necessary to preserve a natural habitat. And even if our farming methods do take up some resources, if they are not too costly and we also get something out of it, point 4 essentially vanishes and - assuming we have strict welfare regulations and really care for their welfare - my only remaining objection is that it is still not permissible to kill these animals, at least not before they have grown so old and sick that being killed might be in their interest. Of course we wouldn't want to eat meat from them afterwards, but I can see how we might get ethical leather, wool eggs this way.
So, what do you think?
Thank you so much for this article! (BTW: Are Gary David O’Brien and BrownHairedEevee the same person? If not, also thanks to the latter for sharing it.)
I was wondering myself for years why longtermists seemingly didn´t really account for (non-human) animals, since it seems quite obvious that in a lot (if not most) of the possible futures, humanity will be nothing but an insignificant minority of all the sentient beings. In fact, my personal motivation for caring about X-risks is mostly due to the instrumental value humanity might have in regards to non-human wellbeing.
Well, I still am wondering, since besides of Beckstead´s argument (which is not really about intrinsic values) you also didn´t mention any particuar arguments human-focused longtermists actually made in regards to this. So I´m still curious why this topic gets so little attention, aside of speciesist reasons (which I am hoping are less common among longtermists compared to the rest of society). Maybe part of this is because the argument for longtermism gets even harder if you also want to convince your audience to abolish speciesism at the same time?
In any case, I completely agree it is a mistake to neglect animals in longtermism.
Im less convinced that WAS is actually a net-negative state. (My guess, though I wouldn´t really call it an informed one, would be that in average an animal in the wild lives a moderate-positive life, despite all the horrors. This average probably depends mostly on insects and small animals in the ocean, however, and I´m really not an expert, so I cannot exculde the possibility that my guess is only wishful thinking.) While this discussion in general is probably better lead elsewhere, I would at least note that your "reasonable estimate" about 1-minute-suffering before any death seems like a good estimate about some of the suffering, but is not really a reason to believe it is also a good estimate about the net-value, since it focuses on one particularly bad minute in a far longer life and does not really tell us anything about the expected value of given life as a whole. (Which would probably be highly dependent on the species.)
If WAS were actually net-positive, it would change a lot in your conclusion about preventing the spread of animals outside of earth. (And sadly also about the current influence of humanity on the overall wellbeing.)
However, most of the arguments (and especially the conclusion that it is a mistake to ignore animas in longtermism) are valid regardless, since it is true in any case that humanity might heavily influence the duration, spread and quality of wildlife as well as the exploitation of farmed or engineered animals.
Best regards,
Marc
Thanks for citing the survey here, and thank you Joshua for your analysis.
Your post doesn´t seem strange to me at this place; at the very least I can´t find any harm in posting it here. (If someone is more interested in other discussions, they may read the first two lines and then skip it.) The only question would be if this is worth YOUR time, and I am confident you are able to judge this (and you apparently did and found it worth your time).
Since you already delved that deep into the material and since I don´t see myself doing the same, here a question to you (or whoever else feeling inclined to answer):
Were there a significant part of experts who thought that HLMI and/or FAOL are downright impossible (at least with anything resembling our current approaches)? I do hear/read doubts like these sometimes. If so, how were these experts included in the mean, since you can´t just include infinity with non-zero probability without the whole number going up to infinity? (If they even used a mean. "Aggregate Forecast" is not very clear; if they used the median ore something similar the second question can be ignored.)