In the long term, we will hopefully invent forms of delicious meat like cultured meat that do not involve sentient animal suffering… When that happens, pro-natalism might make more sense.
As Kevin Kuruc argues, progress happens from people (or productive person-years), not from the bare passage of time. So we should expect there’s some number of productive person-years required to solve this problem. So there simply is no meat-eater problem. As a first-pass model: removing person-years from the present doesn’t reduce the number of animals harmed before a solution is found; it just makes the solution arrive later.
I imagine a longer analysis would include factors like: 1. If intense AI happens in 10 to 50 years, that could do inventing afterwards. 2. I expect that a very narrow slice of the population will be responsible for scientific innovations here, if humans do it. Maybe instead of considering the policies [increase the population everywhere] or [decrease the population everywhere], we could consider more nuanced policies. Related, if one wanted to help with animal welfare, I’d expect that [pro-natalism] would be an incredibly ineffective way of doing so, for the benefit of eventual scientific progress on animals.
I think no one here is trying to use pronatalism to improve animal welfare. The crux for me is more whether pronatalism is net-negative, neutral, or net-positive, and its marginal impact on animal welfare seems to matter in that case. But the total impact of animal suffering dwarfs whatever positive or negative impact pronatalism might have.
As a first-pass model: removing person-years from the present doesn’t reduce the number of animals harmed before a solution is found; it just makes the solution arrive later.
I doubt that is a good way to model this (for farmed animals). Consider the extremes:
If we reduce the human population size to 0, we reduce the amount of suffering of farmed animals to zero, since there will be no more farmed animals
If we increase the human population to the Malthusian limit, we increase the amount of suffering of farmed animals in the short and probably medium terms, and may or may not decrease farmed animal suffering in the longer term. One reason to think we would increase the amount of suffering by adding many more people is that, historically, farmed animal suffering and human population have likely been closely correlated. At any rate, the amount of farmed animal suffering in this scenario is likely nonzero.
So as a first approximation, we should just assume the amount of suffering in factory farms increases monotonically with the human population, since we can be fairly confident in these three data points (no suffering with no humans; lots of suffering with 8B humans; maybe more, maybe less suffering at the Malthusian limit). Of course that would be an oversimplified model. But it is a starting point, and to get from that starting point to “adding people on the margin reduces or doesn’t affect expected farmed animal suffering” needs a better argument.
Basically, yes, assume that meat eating increases with the size of human population. But the scientific effort towards ending the need to meat eat also increases with the size of the human population, assuming marginal extra people are as equally likely to go into researching the problem as the average person. Under a simple model the two exactly balance out, as you can see in the spreadsheet.
I just think real life breaks the simple model in ways I have described below, in a way that preserves a meat-eater problem.
Yeah, but as you point out below, that simple model makes some unrealistic assumptions (e.g., that a solution will definitely be found that fully eliminates farmed animal suffering, and that a person starts contributing, in expectation, to solving meat eating at age 0). So it still seems to me that a better argument is needed to shift the prior.
right—in that simple model, each extra marginal average person decreases the time taken to invent cultured meat at the same rate as they contribute to the problem, and there’s an exact identity between those rates. But there are complicating factors that I think work against assuring us there’s no meat-eater problem:
An extra person starts eating animals from a very young age, but won’t start contributing to the meat-eater problem until they’re intellectually developed enough to make a contribution (21 yers to graduate undergraduate, 25-30 to get a PhD).
There’s a delay between when they invent a solution and when meat eating can actually be phased out, though perhaps that’s implicitly built into the model by the previous point
I do concede that the problem is mitigated somewhat because if we expect cultured meat to take over within the lifetime of a new person, then their harm (and impact) is scaled down proportionately, but the intrinsic hedonic value of their existence isn’t similarly scaled down.
But it doesn’t sound as simple as just “there’s no meat-eater problem”.
IFF cultured meat is a technological inevitability. I don’t think this is necessarily true, we could also improve the efficiency of factory farming and create a very bad future for animals.
Or if any other kind of progress (including moral progress, some of which will come from future people) will eventually abolish factory-farming. I’d be utterly shocked if factory-farming is still a thing 1000+ years from now. But sure, it is a possibility, so you could discount the value of new lives by some modest amount to reflect this risk. I just don’t think that will yield the result that marginal population increases are net-negative for the world in expectation.
As Kevin Kuruc argues, progress happens from people (or productive person-years), not from the bare passage of time. So we should expect there’s some number of productive person-years required to solve this problem. So there simply is no meat-eater problem. As a first-pass model: removing person-years from the present doesn’t reduce the number of animals harmed before a solution is found; it just makes the solution arrive later.
I imagine a longer analysis would include factors like:
1. If intense AI happens in 10 to 50 years, that could do inventing afterwards.
2. I expect that a very narrow slice of the population will be responsible for scientific innovations here, if humans do it. Maybe instead of considering the policies [increase the population everywhere] or [decrease the population everywhere], we could consider more nuanced policies. Related, if one wanted to help with animal welfare, I’d expect that [pro-natalism] would be an incredibly ineffective way of doing so, for the benefit of eventual scientific progress on animals.
I think no one here is trying to use pronatalism to improve animal welfare. The crux for me is more whether pronatalism is net-negative, neutral, or net-positive, and its marginal impact on animal welfare seems to matter in that case. But the total impact of animal suffering dwarfs whatever positive or negative impact pronatalism might have.
I doubt that is a good way to model this (for farmed animals). Consider the extremes:
If we reduce the human population size to 0, we reduce the amount of suffering of farmed animals to zero, since there will be no more farmed animals
If we increase the human population to the Malthusian limit, we increase the amount of suffering of farmed animals in the short and probably medium terms, and may or may not decrease farmed animal suffering in the longer term. One reason to think we would increase the amount of suffering by adding many more people is that, historically, farmed animal suffering and human population have likely been closely correlated. At any rate, the amount of farmed animal suffering in this scenario is likely nonzero.
So as a first approximation, we should just assume the amount of suffering in factory farms increases monotonically with the human population, since we can be fairly confident in these three data points (no suffering with no humans; lots of suffering with 8B humans; maybe more, maybe less suffering at the Malthusian limit). Of course that would be an oversimplified model. But it is a starting point, and to get from that starting point to “adding people on the margin reduces or doesn’t affect expected farmed animal suffering” needs a better argument.
I think Richard is right about the general case. It was a bit unintuitive to me until I ran the numbers in a spreadsheet, which you can see here:
https://docs.google.com/spreadsheets/d/1pRW3WinG1gzJM3RER2Q4Tl5kscJRESuG8qupHGN1Wnw/edit?usp=drivesdk
Basically, yes, assume that meat eating increases with the size of human population. But the scientific effort towards ending the need to meat eat also increases with the size of the human population, assuming marginal extra people are as equally likely to go into researching the problem as the average person. Under a simple model the two exactly balance out, as you can see in the spreadsheet.
I just think real life breaks the simple model in ways I have described below, in a way that preserves a meat-eater problem.
Yeah, but as you point out below, that simple model makes some unrealistic assumptions (e.g., that a solution will definitely be found that fully eliminates farmed animal suffering, and that a person starts contributing, in expectation, to solving meat eating at age 0). So it still seems to me that a better argument is needed to shift the prior.
Fair enough.
My central expectation is that value of one more human life created is roughly about even with the amount of nonhuman suffering that life would cause (based on here https://forum.effectivealtruism.org/posts/eomJTLnuhHAJ2KcjW/comparison-between-the-hedonic-utility-of-human-life-and#Poultry_living_time_per_capita). I’m also willing to assume cultured meat is not too long away. Then the childhood delay til contribution only makes a fractional difference and I tip very slightly back into the pro natalist camp, while still accepting that the meat eater problem is relevant.
right—in that simple model, each extra marginal average person decreases the time taken to invent cultured meat at the same rate as they contribute to the problem, and there’s an exact identity between those rates. But there are complicating factors that I think work against assuring us there’s no meat-eater problem:
An extra person starts eating animals from a very young age, but won’t start contributing to the meat-eater problem until they’re intellectually developed enough to make a contribution (21 yers to graduate undergraduate, 25-30 to get a PhD).
There’s a delay between when they invent a solution and when meat eating can actually be phased out, though perhaps that’s implicitly built into the model by the previous point
I do concede that the problem is mitigated somewhat because if we expect cultured meat to take over within the lifetime of a new person, then their harm (and impact) is scaled down proportionately, but the intrinsic hedonic value of their existence isn’t similarly scaled down.
But it doesn’t sound as simple as just “there’s no meat-eater problem”.
IFF cultured meat is a technological inevitability. I don’t think this is necessarily true, we could also improve the efficiency of factory farming and create a very bad future for animals.
Or if any other kind of progress (including moral progress, some of which will come from future people) will eventually abolish factory-farming. I’d be utterly shocked if factory-farming is still a thing 1000+ years from now. But sure, it is a possibility, so you could discount the value of new lives by some modest amount to reflect this risk. I just don’t think that will yield the result that marginal population increases are net-negative for the world in expectation.