Peter Godfrey-Smith is an important voice on this. Given that major media find the “pardon” of turkeys just good, silly fun, it’s tough sledding to get people to believe that shrimp welfare is a thing.
I’d appreciate a restack of my recent post about Singer’s Consider the Turkey. I’m new to Substack but a long term editor and writer now tackling philosophic topics as a fresh project.
21,000 shrimps? It's the best time of the year! Thanks for this article. Excited to see what others will cook up (hopefully not shrimp) for shrimpact season.
Have you spoken with Dr Avi about shrimp welfare before? He has an axe to grind with the SWP and the EA community more broadly https://youtu.be/0MHshBjRTX4?si=72zRBNljWtL4zHlx&t=26 In essence, the SWP may cause more suffering for shrimp (i.e. they probably feel electric shocks worse than asphyxiation), but more importantly they are probably not sentient to any degree we should care about at all. I think it'd be a great conversation between you two, and I know you've had good interactions in the past.
We've spoken a bit. I don't get what his argument is about asphyxiation because the unpleasantness of that wouldn't be mediated by nociceptors. I similarly think his arguments against shrimp being conscious aren't any good and rely on neuron count as a proxy when there isn't good reason to think it is a good proxy, especially for simple organisms.
I mean his other ones are even more dramatic like synapse count and will have even bigger problems.
It would be mediated by other brain regions responsible for unpleasantness not just physical pain. When humans feel sad for example that’s not caused by nociceptors
As far as I understand, it's not even synapse count. Most of the inference is made from the presence of certain structures e.g. cerebral cortex, gyrencephaly. So a correlate adjusted for things like these (e.g. cortical neuron count). Not that these are perfect, but I don't see why they're any more problematic than the RP welfare ranges for e.g.
What part of the shrimp's brain would mediate non-pain related asphyxiation unpleasantness? Also, this is kind of strange considering how much emphasis you place on the pain part of the asphyxiation process.
Side point: when you speak about "If A is 1/x sentient as B, then killing X number of A is as bad as killing 1 B" what are your thoughts on the dust-speck debate you had with him previously, where (I believe) you conceded it wasn't unreasonable to have thresholds for value under which any amount of repeat instances of suffering would not be able to surmount? I think this actually compounds the issues in the previous points as well.
Let's distinguish between pain and suffering. If you suffocate, that is very unpleasant, but it's not painful. There's no physical hurting in the same way as when you stub your toe. I think suffocation causes shrimp to suffer but it's not obvious that it's painful, and thus wouldn't be mediated through nociceptors.
It's alright to say that the cerebral cortex leads to more expected pain. But what is unreasonable is to be extremely confident that it leads to hundreds of thousands or millions of times more expected pain. We shouldn't be massively certain that creatures with simple brains feel barely any pain, and that's enough to get the EV argument off the ground.
No, I think it is unreasonable to have value thresholds. That will imply infinite things slightly below the threshold are less bad than one thing above the threshold. I don't really remember what I said when we argued about this many years ago.
But even if you have thresholds, there's no extremely strong reason to think shrimp are below the threshold. There are some systems that are pretty advanced even in simple creatures (E.g. they can see pretty well). There is no extremely strong argument that pain isn't like that, and even a moderate amount of uncertainty makes the case for saving thousands of shrimp per dollar very robust.
“Let's distinguish between pain and suffering. If you suffocate, that is very unpleasant, but it's not painful. There's no physical hurting in the same way as when you stub your toe. I think suffocation causes shrimp to suffer but it's not obvious that it's painful, and thus wouldn't be mediated through nociceptors.”
It’s not painful? You described it as "painful" 26 times in your original article and "agonizing" 5 times. This is a major walk-back!
Now I'm a bit confused as to why you're invoking pain on your side of the argument in your response here. If asphyxiation isn't painful, then the only source of pain now is the shocking, so the precautionary principle being invoked now counts against using it. Though I will disagree that asphyxiation is not painful, there are ample data on nociception in these situations, just not the type that shrimp are capable of.
Edit: In addition, on revisiting the RP welfare range estimates, I don't see any non-pain types of suffering that would be evoked by the methods of killing shrimp in currently in use—so citing their estimates here doesn't appear to make a whole lot of sense iff their estimates are not capturing the capacity for any relevant experiences. Another interesting question is how quickly any alleged "unpleasantness" would even last during hypoxia. I can't imagine much of their ganglion last very long.
To the side point: I don't think it's absurd to have hard thresholds, for e.g. infinite dust-specks are still way less bad than 1 torture, but we don't even need to have a threshold. It could just be a different relationship than purely linear and directly proportional to value like your original claim. I'm not sure the opportunity cost is met in such a case. I just think it was a very sweeping claim about aggregate suffering that I don't think is very defensible on its own.
Peter Godfrey-Smith is an important voice on this. Given that major media find the “pardon” of turkeys just good, silly fun, it’s tough sledding to get people to believe that shrimp welfare is a thing.
I’d appreciate a restack of my recent post about Singer’s Consider the Turkey. I’m new to Substack but a long term editor and writer now tackling philosophic topics as a fresh project.
Just donated $70
Thanks for the post, I was looking for any EA / Animal Welfare donation match opportunities for Thanksgiving.
I donated $20 for now, will add some more soon. Happy shrimp welfare week! May their suffering soon end.
Awesome
21,000 shrimps? It's the best time of the year! Thanks for this article. Excited to see what others will cook up (hopefully not shrimp) for shrimpact season.
Have you spoken with Dr Avi about shrimp welfare before? He has an axe to grind with the SWP and the EA community more broadly https://youtu.be/0MHshBjRTX4?si=72zRBNljWtL4zHlx&t=26 In essence, the SWP may cause more suffering for shrimp (i.e. they probably feel electric shocks worse than asphyxiation), but more importantly they are probably not sentient to any degree we should care about at all. I think it'd be a great conversation between you two, and I know you've had good interactions in the past.
We've spoken a bit. I don't get what his argument is about asphyxiation because the unpleasantness of that wouldn't be mediated by nociceptors. I similarly think his arguments against shrimp being conscious aren't any good and rely on neuron count as a proxy when there isn't good reason to think it is a good proxy, especially for simple organisms.
They don't rely on neuron count as a proxy, he's actually argued against that: https://www.youtube.com/watch?v=2WPGcE8l1zA
What would the unpleasantness of asphyxiation be mediated by then?
I mean his other ones are even more dramatic like synapse count and will have even bigger problems.
It would be mediated by other brain regions responsible for unpleasantness not just physical pain. When humans feel sad for example that’s not caused by nociceptors
As far as I understand, it's not even synapse count. Most of the inference is made from the presence of certain structures e.g. cerebral cortex, gyrencephaly. So a correlate adjusted for things like these (e.g. cortical neuron count). Not that these are perfect, but I don't see why they're any more problematic than the RP welfare ranges for e.g.
What part of the shrimp's brain would mediate non-pain related asphyxiation unpleasantness? Also, this is kind of strange considering how much emphasis you place on the pain part of the asphyxiation process.
Side point: when you speak about "If A is 1/x sentient as B, then killing X number of A is as bad as killing 1 B" what are your thoughts on the dust-speck debate you had with him previously, where (I believe) you conceded it wasn't unreasonable to have thresholds for value under which any amount of repeat instances of suffering would not be able to surmount? I think this actually compounds the issues in the previous points as well.
Let's distinguish between pain and suffering. If you suffocate, that is very unpleasant, but it's not painful. There's no physical hurting in the same way as when you stub your toe. I think suffocation causes shrimp to suffer but it's not obvious that it's painful, and thus wouldn't be mediated through nociceptors.
It's alright to say that the cerebral cortex leads to more expected pain. But what is unreasonable is to be extremely confident that it leads to hundreds of thousands or millions of times more expected pain. We shouldn't be massively certain that creatures with simple brains feel barely any pain, and that's enough to get the EV argument off the ground.
No, I think it is unreasonable to have value thresholds. That will imply infinite things slightly below the threshold are less bad than one thing above the threshold. I don't really remember what I said when we argued about this many years ago.
But even if you have thresholds, there's no extremely strong reason to think shrimp are below the threshold. There are some systems that are pretty advanced even in simple creatures (E.g. they can see pretty well). There is no extremely strong argument that pain isn't like that, and even a moderate amount of uncertainty makes the case for saving thousands of shrimp per dollar very robust.
“Let's distinguish between pain and suffering. If you suffocate, that is very unpleasant, but it's not painful. There's no physical hurting in the same way as when you stub your toe. I think suffocation causes shrimp to suffer but it's not obvious that it's painful, and thus wouldn't be mediated through nociceptors.”
It’s not painful? You described it as "painful" 26 times in your original article and "agonizing" 5 times. This is a major walk-back!
Now I'm a bit confused as to why you're invoking pain on your side of the argument in your response here. If asphyxiation isn't painful, then the only source of pain now is the shocking, so the precautionary principle being invoked now counts against using it. Though I will disagree that asphyxiation is not painful, there are ample data on nociception in these situations, just not the type that shrimp are capable of.
Edit: In addition, on revisiting the RP welfare range estimates, I don't see any non-pain types of suffering that would be evoked by the methods of killing shrimp in currently in use—so citing their estimates here doesn't appear to make a whole lot of sense iff their estimates are not capturing the capacity for any relevant experiences. Another interesting question is how quickly any alleged "unpleasantness" would even last during hypoxia. I can't imagine much of their ganglion last very long.
To the side point: I don't think it's absurd to have hard thresholds, for e.g. infinite dust-specks are still way less bad than 1 torture, but we don't even need to have a threshold. It could just be a different relationship than purely linear and directly proportional to value like your original claim. I'm not sure the opportunity cost is met in such a case. I just think it was a very sweeping claim about aggregate suffering that I don't think is very defensible on its own.
Any chance you can give a complimentary subscription with a donation?
Yes! I give out subscriptions if people give at least 30 a month
When does the week end officially?
Liked and shared immediately.