Er, it does seem like there might conceivably be reasons not to “soak every meter of Earth with DDT”[1], even if you think that exterminating all insects would be a good thing.
Not convinced of this actually. If, for example, ant suffering is anywhere near as bad as the honey post claims bee suffering to be, then the Amazon Rainforest is significantly worse than Christian Hell. A policy of “end it all as fast as possible yes immediately yes even if there are casualties yes even if the casualties are all of humanity yes even if we could avoid those casualties by increasing the odds of failure even by a percentage point, the stakes are too high and everything else is a rounding error in expectation” seems pretty straightforwardly correct on these numbers. (Of course, this is a reason to believe that these numbers are wrong! So do that instead.)
Well, for one thing, disbelieving the numbers, as such, is not the only way to end up disagreeing with the conclusion. But let’s set that aside.
Suppose you actually decide that we should “soak every meter of Earth with DDT” (or something close to it). Let’s consider some practical obstacles:
How? I mean, literally, how? What’s the plan for doing this? Who’s going to do it, with what equipment, with DDT produced by whom, where? What’re the logistics of this operation going to be like? Is this even remotely feasible, in any practical sense?
Who’s going to pay for it?
How are you going to convince people to let you do it? How are you going to convince all the people of all the countries in the world to let you do it? What’s your plan for when people try to stop you?
What about life in the oceans? What’s your plan for those animals? (Do you have a plan for them? There are a whole lot of them!)
I could keep going, but I think you get the idea.
In contrast, “try to convince people to stop eating honey” is immediately and directly actionable: just start writing blog posts about it, and that will (one may hope, anyhow) straightforwardly lead to less honey being eaten.
So, first: The logistical details of reducing wild impact biomass are mooted by the fact that I meant it as a reductio, not a proposal. I have no strong reason to think that spraying insecticide would be a better strategy than gene drives or sterile insect technique or deforestation, or that DDT is the most effective insecticide.
To put rough numbers on it: honeybees are about 4e-7 by count or 7e-4 by biomass of all insects (estimate by o3). There is no such extreme skew for mammals and birds (o3). While domesticated honeybees have some bad things happen to them, they don’t seem orders of magnitude worse than what happens to wild insects.
Caring highly about insect suffering, in a way that scales linearly with population, does not match my values but does not seem philosophically incoherent. But because of the wild/domestic population skew, avoiding honey for this reason does seem philosophically incoherent.
I think I was being pithy instead of clear, I apologize. I’m trying to emphasize that your criticism of this argument is substantially wrong—this is indeed not feasible, but that ( (a) in your enumeration) is the smallest of these criticisms and the only one that applies, and this error does not damage the implied critique of the original argument as I read it. This would actually accomplish a decent portion of the goals that are implied by this valuation of insect pain and the “horrible consequences” would not actually be horrible in this account.
This argument against eating honey relies on a sort of social trust—you probably don’t accept that bee suffering is 15% as important as time-equivalent human suffering, but somebody else is saying that it is, and the result of that belief is really catastrophic. Rhetorically this makes sense and is completely valid—to paraphrase and generalize, “here’s a method of reasoning that you probably accept, here’s a conclusion it can reach, here’s the amount of uncertainty you can have before that conclusion is no longer reached, that somebody else who thinks like you and cares about the things you do believes these numbers to be correct is a good reason to trust this argument at least a small amount, so don’t eat honey.”
The implicit criticism here, as I read it, is something like “sure, but these numbers are very obviously batshit. There’s a silent addendum if you keep thinking this way, ‘don’t eat honey’ is the bottom line but the postscript is ‘also this is only a hedge until we can, by whatever means available, completely eliminate non-reflective or semi-reflective life in the universe even at the cost of all reflective life’, and if this conclusion actually does follow, then the social trust that compels me towards the original conclusion disappears.”
That is, there’s a straightforward counterargument to the original post, which is that the argument is completely numerical and the numbers make no sense. (You could also, as you note, take issue with the structure of the argument, but outside of nitpicks that I feel pretty confident are fixable, I do not—a calculation more-or-less of this form could actually compel me to seemingly bizarre conclusions if I bought the numbers). The obvious response is that the conclusion is very strong and if you think it’s at all plausible that you’re wrong you should still stop eating honey to be sure. If you assign a normal amount of social credence to the post’s author, which you probably do if they clearly buy your moral and epistemic framework, you should probably give their argument enough credence to accept its conclusion even if you don’t think it’s right. But this extension shows that you should probably not extend that much social credence.
Not convinced of this actually. If, for example, ant suffering is anywhere near as bad as the honey post claims bee suffering to be, then the Amazon Rainforest is significantly worse than Christian Hell. A policy of “end it all as fast as possible yes immediately yes even if there are casualties yes even if the casualties are all of humanity yes even if we could avoid those casualties by increasing the odds of failure even by a percentage point, the stakes are too high and everything else is a rounding error in expectation” seems pretty straightforwardly correct on these numbers. (Of course, this is a reason to believe that these numbers are wrong! So do that instead.)
Well, for one thing, disbelieving the numbers, as such, is not the only way to end up disagreeing with the conclusion. But let’s set that aside.
Suppose you actually decide that we should “soak every meter of Earth with DDT” (or something close to it). Let’s consider some practical obstacles:
How? I mean, literally, how? What’s the plan for doing this? Who’s going to do it, with what equipment, with DDT produced by whom, where? What’re the logistics of this operation going to be like? Is this even remotely feasible, in any practical sense?
Who’s going to pay for it?
How are you going to convince people to let you do it? How are you going to convince all the people of all the countries in the world to let you do it? What’s your plan for when people try to stop you?
What about life in the oceans? What’s your plan for those animals? (Do you have a plan for them? There are a whole lot of them!)
I could keep going, but I think you get the idea.
In contrast, “try to convince people to stop eating honey” is immediately and directly actionable: just start writing blog posts about it, and that will (one may hope, anyhow) straightforwardly lead to less honey being eaten.
So, first: The logistical details of reducing wild impact biomass are mooted by the fact that I meant it as a reductio, not a proposal. I have no strong reason to think that spraying insecticide would be a better strategy than gene drives or sterile insect technique or deforestation, or that DDT is the most effective insecticide.
To put rough numbers on it: honeybees are about 4e-7 by count or 7e-4 by biomass of all insects (estimate by o3). There is no such extreme skew for mammals and birds (o3). While domesticated honeybees have some bad things happen to them, they don’t seem orders of magnitude worse than what happens to wild insects.
Caring highly about insect suffering, in a way that scales linearly with population, does not match my values but does not seem philosophically incoherent. But because of the wild/domestic population skew, avoiding honey for this reason does seem philosophically incoherent.
I think I was being pithy instead of clear, I apologize. I’m trying to emphasize that your criticism of this argument is substantially wrong—this is indeed not feasible, but that ( (a) in your enumeration) is the smallest of these criticisms and the only one that applies, and this error does not damage the implied critique of the original argument as I read it. This would actually accomplish a decent portion of the goals that are implied by this valuation of insect pain and the “horrible consequences” would not actually be horrible in this account.
This argument against eating honey relies on a sort of social trust—you probably don’t accept that bee suffering is 15% as important as time-equivalent human suffering, but somebody else is saying that it is, and the result of that belief is really catastrophic. Rhetorically this makes sense and is completely valid—to paraphrase and generalize, “here’s a method of reasoning that you probably accept, here’s a conclusion it can reach, here’s the amount of uncertainty you can have before that conclusion is no longer reached, that somebody else who thinks like you and cares about the things you do believes these numbers to be correct is a good reason to trust this argument at least a small amount, so don’t eat honey.”
The implicit criticism here, as I read it, is something like “sure, but these numbers are very obviously batshit. There’s a silent addendum if you keep thinking this way, ‘don’t eat honey’ is the bottom line but the postscript is ‘also this is only a hedge until we can, by whatever means available, completely eliminate non-reflective or semi-reflective life in the universe even at the cost of all reflective life’, and if this conclusion actually does follow, then the social trust that compels me towards the original conclusion disappears.”
That is, there’s a straightforward counterargument to the original post, which is that the argument is completely numerical and the numbers make no sense. (You could also, as you note, take issue with the structure of the argument, but outside of nitpicks that I feel pretty confident are fixable, I do not—a calculation more-or-less of this form could actually compel me to seemingly bizarre conclusions if I bought the numbers). The obvious response is that the conclusion is very strong and if you think it’s at all plausible that you’re wrong you should still stop eating honey to be sure. If you assign a normal amount of social credence to the post’s author, which you probably do if they clearly buy your moral and epistemic framework, you should probably give their argument enough credence to accept its conclusion even if you don’t think it’s right. But this extension shows that you should probably not extend that much social credence.