You seem to be assuming that decreasing factory-farming is beneficial, but I think it is harmful if the lives of soil animals have more suffering than happiness. I estimate factory-farming increases the animal-years of wild soil animals much more than it decreases the animal-years of farmed animals. For example, I estimate decreasing the consumption of chicken meat by 0.1 kg results in a reduction of 2.87 chicken-days, but in an increase of 6.16 M soil-animal-years due to decreasing the amount of chicken’s feed, and therefore causing some crops to be replaced by biomes with more soil animals. The increase in the animal-years of soil animals is 783 M times the decrease in the animal-years of chickens. This suggests decreasing the consumption of chicken only increases animal welfare if the suffering of chickens is more than 783 M times as intense (as bad per animal-year) as that of soil animals. Nematodes, the most simple soil animal, have 302 neurons, and chickens have 221 M, 732 k (= 221*10^6/302) times as many as nematodes. So the suffering of chickens is 732 k times as intense as that of soil animals if the intensity of suffering is proportional to the number of neurons. If so, decreasing the consumption of chicken would cause 1.07 k (= 783*10^6/(732*10^3)) times as much suffering to soil animals as it averts the suffering of chickens, thus hugely increasing net suffering.
Moreover, I suspect the above underestimates the suffering caused to soil animals relative to the suffering of chickens averted. The expected welfare ranges presented in Table 8.6 of Bob Fischer’s book about comparing animal welfare across species are very well explained by “number of neurons”^0.188 (R^2 of 78.6 %). This suggests the suffering of chickens is only 12.7 (= (732*10^3)^0.188) times as intense as that of soil animals. If so, decreasing the consumption of chicken would cause 61.7 M (= 783*10^6/12.7) times as much suffering to soil animals as it averts the suffering of chickens. I personally prefer an exponent of the number of neurons of 0.5, which suggests the suffering of chickens is 856 (= (732*10^3)^0.5) times as intense as that of soil animals. However, this still results in decreasing the consumption of chicken causing 915 k (= 783*10^6/856) times as much suffering to soil animals as it averts the suffering of chickens.
My crucial assumption above that the lives of soil animals have more suffering than happiness is not unusual. Charity Entrepreneurship estimated in 2018 that the lives of wild insects have more suffering than happiness. Moreover, I suspect many people in the animal welfare space think that bringing more lives into existence could be bad even if they have more happiness than suffering. However, I think having more lives with more happiness than suffering is robustly good all else equal, and I can easily see the lives of soil animals having more happiness than suffering (although this is not my best guess). So it would be great to have more research on this question.
If my assumption holds, decreasing the animal-years of soil animals is beneficial, and therefore increasing agricultural land is beneficial too, as it has fewer soil animals than alternative biomes. In this case, replacing plant-based foods with animal-based foods tends to be beneficial due to tendentially increasing agricultural land. However, if soil animals have lives with more happiness than suffering, my conclusion would be reversed, such that replacing animal-based foods with plant-based foods would tend to be beneficial.
Quickly → I sympathize with these arguments, but I see the above podcast as practically a different topic. Could be a good separate blog post on its own.
Down voted. This podcast is about ending factory farming and I don’t think it would make sense to bring in those arguments about smallee creatures—especially on a podcast where messages get to be clear and simple.
Below is the comment I left on the original post.
Thanks for the podcast and post, Lewis!
You seem to be assuming that decreasing factory-farming is beneficial, but I think it is harmful if the lives of soil animals have more suffering than happiness. I estimate factory-farming increases the animal-years of wild soil animals much more than it decreases the animal-years of farmed animals. For example, I estimate decreasing the consumption of chicken meat by 0.1 kg results in a reduction of 2.87 chicken-days, but in an increase of 6.16 M soil-animal-years due to decreasing the amount of chicken’s feed, and therefore causing some crops to be replaced by biomes with more soil animals. The increase in the animal-years of soil animals is 783 M times the decrease in the animal-years of chickens. This suggests decreasing the consumption of chicken only increases animal welfare if the suffering of chickens is more than 783 M times as intense (as bad per animal-year) as that of soil animals. Nematodes, the most simple soil animal, have 302 neurons, and chickens have 221 M, 732 k (= 221*10^6/302) times as many as nematodes. So the suffering of chickens is 732 k times as intense as that of soil animals if the intensity of suffering is proportional to the number of neurons. If so, decreasing the consumption of chicken would cause 1.07 k (= 783*10^6/(732*10^3)) times as much suffering to soil animals as it averts the suffering of chickens, thus hugely increasing net suffering.
Moreover, I suspect the above underestimates the suffering caused to soil animals relative to the suffering of chickens averted. The expected welfare ranges presented in Table 8.6 of Bob Fischer’s book about comparing animal welfare across species are very well explained by “number of neurons”^0.188 (R^2 of 78.6 %). This suggests the suffering of chickens is only 12.7 (= (732*10^3)^0.188) times as intense as that of soil animals. If so, decreasing the consumption of chicken would cause 61.7 M (= 783*10^6/12.7) times as much suffering to soil animals as it averts the suffering of chickens. I personally prefer an exponent of the number of neurons of 0.5, which suggests the suffering of chickens is 856 (= (732*10^3)^0.5) times as intense as that of soil animals. However, this still results in decreasing the consumption of chicken causing 915 k (= 783*10^6/856) times as much suffering to soil animals as it averts the suffering of chickens.
My crucial assumption above that the lives of soil animals have more suffering than happiness is not unusual. Charity Entrepreneurship estimated in 2018 that the lives of wild insects have more suffering than happiness. Moreover, I suspect many people in the animal welfare space think that bringing more lives into existence could be bad even if they have more happiness than suffering. However, I think having more lives with more happiness than suffering is robustly good all else equal, and I can easily see the lives of soil animals having more happiness than suffering (although this is not my best guess). So it would be great to have more research on this question.
If my assumption holds, decreasing the animal-years of soil animals is beneficial, and therefore increasing agricultural land is beneficial too, as it has fewer soil animals than alternative biomes. In this case, replacing plant-based foods with animal-based foods tends to be beneficial due to tendentially increasing agricultural land. However, if soil animals have lives with more happiness than suffering, my conclusion would be reversed, such that replacing animal-based foods with plant-based foods would tend to be beneficial.
Quickly → I sympathize with these arguments, but I see the above podcast as practically a different topic. Could be a good separate blog post on its own.
Thanks, Ozzie. You may be interested in my post More animal farming increases animal welfare if soil animals have negative lives?.
Down voted. This podcast is about ending factory farming and I don’t think it would make sense to bring in those arguments about smallee creatures—especially on a podcast where messages get to be clear and simple.