Jeff because he doesn’t seem to have provided any justification (from what I’ve seen) for the claim that animals don’t have relevant experiences that make them moral patients. He simply asserts this as his view. It’s not even an argument, let alone a strong one.
I agree I haven’t given an argument on this. At various times people have asked what my view is (ex: we’re taking here about something prompted by my completing a survey prompt) and I’ve given that.
Explaining why I have this view would be a big investment in time: I have a bundle of intuitions and thoughts that put me here, but converting that into a cleanly argued blog post would be a lot more work than I would normally do for fun and I don’t expect this to be fun.
This is especially the case because If I did a good job at this I might end up primarily known for being an anti-animal advocate, and since I think my views on animals are much less important than many of my other views, I wouldn’t see this as at all a good thing. I also expect that, again, conditional on doing a good job of this, I would need to spend a lot of time as a representative of this position: responding to the best counter arguments, evaluating new information as it comes up, people wanting me to participate in debates, animal advocates thinking that changing my mind is really very important for making progress toward their goals. These are similarly not where I want to put my time and energy, either for altruistic reasons personal enjoyment.
The normal thing to do would be to stop here: I’ve said what my view is, and explained why I’ve never put the effort into a careful case for that position. But I’m more committed to transparency than I am to the above, so I’m going to take about 10 minutes (I have 14 minutes before my kids wake up) to very quickly sketch the main things going into my view. Please read this keeping in mind that it is something I am sharing to be helpful, and I’m not claiming it’s fully argued.
The key question for me is whether, in a given system, there’s anyone inside to experience anything.
I think extremely small collections of neurons (ex: nematodes) can receive pain, in the sense of updating on inputs to generate less of some output. But I draw a distinction between pain and suffering, where the latter requires experience. And I think it’s very unlikely nematodes experience anything.
I don’t think this basic pleasure or pain matters, and if you can’t make something extremely morally good by maximizing the number of happy neurons per cubic centimeter.
I’m pretty sure that most adult humans do experience things, because I do and I can talk to other humans about this.
I think it is pretty unlikely that very young children, in their first few months, have this kind of inner experience.
I don’t find most things that people give as examples for animal consciousness to be very convincing, because you can often make quite a simple system that displays these features.
While some of my views above could imply that some humans are more valuable come up morally than others, I think it would be extremely destructive to act that way. Lots and lots of bad history there. I treat all people as morally equal.
The arguments for extending this to people as a class don’t seem to me to justify extending this to all creatures as a class.
I also think there are things that matter beyond experienced joy and suffering (preference satisfaction, etc), and I’m even less convinced that animals have these.
Eliezer’s view is reasonably close to mine, in places where I’ve seen him argue it.
(I’m not going to be engaging with object level arguments on this issue—I’m not trying to become an anti-animal advocate.)
This is especially the case because If I did a good job at this I might end up primarily known for being an anti-animal advocate
I’d be interested to know how likely you think it is that you could do a “good job”. You say you have a “bundle of intuitions and thoughts” which doesn’t seem like much to me.
I’m also very surprised you put yourself at the far end of the spectrum in favor of global health > animal welfare based on a “bundle of intuitions and thoughts” on what is ultimately a very difficult and important question.[1] In your original comment you say “This isn’t as deeply a considered view as I’d like”. Were you saying you haven’t considered deeply enough or that the general community hasn’t?
And thanks for the sketch of your reasoning but ultimately I don’t think it’s very helpful without some justification for claims like the following:
I think it is pretty unlikely that very young children, in their first few months, have this kind of inner experience.
I also put myself at the fair end of the spectrum in the other direction so I feel I should say something about that. I think arguments for animal sentience/moral patienthood are pretty strong (e.g. see here for a summary) and I would not say I’m relying on intuition. I’m not of course sure that animals are moral patients, but even if you put a small probability on it, the vast numbers of animals being treated poorly can justifiably lead to a strong view that resources for animal welfare are better in expectation than resources for global health. Ultimately for this argument not to work based on believing animals aren’t moral patients, I think you probably need to be veryconfident of this to counteract the vast numbers of animals that can be helped.
I’d be interested to know how likely you think it is that you could do a “good job”.
I do think I could do a good job, yes. While I’ve been thinking about these problems off and on for over a decade I’ve never dedicated actual serious time here, and in the past when I’ve put that kind of time into work I’ve been proud of what I’ve been able to do.
You say you have a “bundle of intuitions and thoughts” which doesn’t seem like much to me.
What I meant by that is that I don’t have my overall views organized into a form optimized for explaining to others. I’m not asking other people to assume that because I’ve inscrutably come to this conclusion I’m correct or that they should defer to me in any way. But I’d also be dishonest if I didn’t accurately report my views.
In your original comment you say “This isn’t as deeply a considered view as I’d like”. Were you saying you haven’t considered deeply enough or that the general community hasn’t?
Primarily the former. While if someone in the general community had put a lot of time into looking at this question from a perspective similar to my own and I felt like their work addressed my questions that would certainly help, given that no one has and I’m instead forming my own view I would prefer to have put more work into that view.
To clarify, when I asked if you could do a good job I meant can you put together a convincing argument that might give some people like me pause for thought (maybe this is indeed how you understood me).
If you think you can, I would strongly encourage you to do so. As per another comment of mine, tens of millions of dollars goes towards animal welfare within EA each year. If this money is effectively getting burned it is very useful for the community to know. Also, there is no convincing argument that animals are not moral patients on this forum (or indeed anywhere else) that I am aware of, so your view is exceedingly neglected. I think you could really do a whole lot of good if you do have a great argument up your sleeve.
Your argument that you would effectively be forced into becoming an anti-animal advocate if you convincingly wrote up your views—sorry I don’t really buy it. For example, I don’t think Luke Muehlhauser has been forced into becoming a pro-animal advocate, in the way you hypothesise that you would, after writing his piece. This just seems like too convenient an excuse, sorry.
Of course you’re not under any obligation to write anything (well...perhaps some would argue you are, but I’ll concede you’re not). But if I thought I had a great argument up my sleeve, mostly ignored by the community, which, if true, would mean we were effectively burning tens of millions of dollars a year, I know I’d write it up.
Ah, thank you for clarifying! That is a much stronger sense of “doing a good job” than I was going for. I was trying to point at something like, successfully writing up my views in a way that felt like a solid contribution to the discourse. Explaining what I thought, why I thought it, and why I didn’t find the standard counter arguments convincing. I think this would probably take me about two months of full-time work, so a pretty substantial opportunity cost.
I think I could do this well enough to become the main person people pointed at when they wanted to give an example of a “don’t value animals” EA (which would probably be negative for my other work), but even major success here would probably only result in convincing <5% of animal-focused EAs to change what they were working on. And much less than that for money, since most of the EA money is from OP, which funds animal work as part of an explicit process of worldview diversification.
Your argument that you would effectively be forced into becoming an anti-animal advocate if you convincingly wrote up your views—sorry I don’t really buy it.
I would be primarily known as an anti-animal advocate if I wrote something like this, even if I didn’t want to be.
On whether I would need to put my time into continuing to defend the position, I agree that I strictly wouldn’t have to, but I think that given my temperament and interaction style I wouldn’t actually be able to avoid this. So I need to think of this as if I am allocating a larger amount of time than what it would take to write up the argument.
Your argument that you would effectively be forced into becoming an anti-animal advocate if you convincingly wrote up your views—sorry I don’t really buy it.
OK so he says he would primarily be “known” as an anti-animal advocate not “become” one.
This is especially the case because If I did a good job at this I might end up primarily known for being an anti-animal advocate
But he then also says the following (bold emphasis mine):
I also expect that, again, conditional on doing a good job of this, I would need to spend a lot of time as a representative of this position: responding to the best counter arguments, evaluating new information as it comes up, people wanting me to participate in debates, animal advocates thinking that changing my mind is really very important for making progress toward their goals. These are similarly not where I want to put my time and energy, either for altruistic reasons personal enjoyment.
I’m struggling to see how what I said isn’t accurate. Maybe Jeff should have said “I would feel compelled to” rather than “I would need to”.
To my eyes “be known as an anti-animal advocate” is a much lower bar than “be an anti-animal advocate.”
For example I think some people will (still!) consider me an “anti-climate change advocate” (or “anti-anti-climate change advocate?”) due to a fairly short post I wrote 5+ years ago. I would, from their perspective, take actions consistent with that view (eg I’d be willing to defend my position if challenged, describe ways in which I’ve updated, etc). Moreover, it is not implausible that from their perspective, this is the most important thing I do (since they don’t interact with me at other times, and/or they might think my other actions are useless in either direction).
However, by my lights (and I expect by the lights of e.g. the median EA Forum reader) this would be a bad characterization. I don’t view arguing against climate change interventions as an important aspect of my life, nor do I believe my views on the matter as particularly outside of academic consensus.
Hence the distinction between “known as” vs “become.”
It’s the only part of my comment that argues Jeff was effectively saying he would have to “be” an animal advocate, which is exactly what you’re arguing against.
So I guess my best reply is just to point you back to that...
I guess I still don’t think of “I would need to spend a lot of time as a representative of this position” as being an anti-animal advocate. I spend a lot of time disagreeing with people on many different issues and yet I’d consider myself an advocate for only a tiny minority of them.
Put another way, I view the time spent as just one of the costs of being known as an anti-animal advocate, rather than being one.
Rats and pigs seem to be able to discriminate anxiety from its absence generalizably across causes with a learned behaviour, like pressing a lever when they would apparently feel anxious.[1] In other words, it seems like they can be taught to tell us what they’re feeling in ways unnatural and non-instinctive to them. To me, the difference between this and human language is mostly just a matter of degree, i.e. we form more associations and form them more easily, and we do recursion.
Graziano (2020, pdf), an illusionist and the inventor of Attention Schema Theory, also takes endogenous/top-down/voluntary attention control to be evidence of having a model (schema) of one’s own attention.[2] Then, according to Nieder (2022), there is good evidence for the voluntary/top-down control of attention (and working memory) at least across mammals and birds, and some suggestive evidence for it in some fish.
And I would expect these to happen in fairly preserved neural structures across mammals, at least, including humans.
I also discuss desires and preferences in other animals more here and here.
Carey and Fry (1995) showed that pigs generalize the discrimination between non-anxiety states and drug-induced anxiety to non-anxiety and anxiety in general, in this case by pressing one lever repeatedly with anxiety, and alternating between two levers without anxiety (the levers gave food rewards, but only if they pressed them according to the condition). Many more such experiments were performed on rats, as discussed in Sánchez-Suárez, 2016, summarized in Table 2 on pages 63 and 64 and discussed further across chapter 3.
Rats could discriminate between the injection of the anxiety-inducing drug PTZ and saline injection, including at subconvulsive doses. Various experiments with rats and PTZ have effectively ruled out convulsions as the discriminant, further supporting that it’s the anxiety itself that they’re discriminating, because they could discriminate PTZ from control without generalizing between PTZ and non-anxiogenic drugs, and with the discrimination blocked by anxiolytics and not nonanxiolytic anticonvulsants.
Rats further generalized between various pairs of anxiety(-like) states, like those induced by PTZ, drug withdrawal, predator exposure, ethanol hangover, “jet lag”, defeat by a rival male, high doses of stimulants like bemegride and cocaine, and movement restraint.
But could such results merely reflect a “blindsight-like” guessing: a mere discrimination response that need not reflect underlying awareness? After all, as we have seen for S.P.U.D. subjects, decerebrated pigeons can use colored lights as DSs (128), and humans can use subliminal visual stimuli as DSs [e.g., (121)]. We think several refinements could reduce this risk.
I would expect that checking which brain systems are involved and what their typical functions are could provide further evidence. The case for other mammals would be strongest, given more preserved functions across them, including humans.
Any creature that can endogenously direct attention must have some kind of attention schema, and good control of attention has been demonstrated in a range of animals including mammals and birds (e.g., Desimone & Duncan, 1995; Knudsen, 2018; Moore & Zirnsak, 2017). My guess is that most mammals and birds have some version of an attention schema that serves an essentially similar function, and contains some of the same information, as ours does. Just as other animals must have a body schema or be condemned to a flailing uncontrolled body, they must have an attention schema or be condemned to an attention system that is purely at the mercy of every new sparkling, bottom-up pull on attention. To control attention endogenously implies an effective controller, which implies a control model.
Thanks for taking the time to expose your view clearly here, and explaining why you do not spend a lot of time on the topic (which I respect).
If I understand correctly, the difference in consideration you make between humans and animals seems to boil down to “I can talk to humans, and they can tell me that they have an inner experience, while animals cannot (same for small children)”.
While nobody disputes that, I find it weird that your conclusion is not “I’m very uncertain about other systems”, but “other systems that cannot tell me directly about their inner experience (very small children, animals) probably don’t have any relevant inner experience”. I’m not sure how you got to that conclusion. At the very least, this would justify extreme uncertainty.
Personally, I think that the fact that animals display a lot of behaviour similar to humans in similar situations should be a significant update toward thinking they have some kind of experience. For instance, a pig is screaming and trying to escape when it is castrated, just as humans would do (we have to observe behaviours).
We can probably build robots that can do the same thing, but that just means we’re good at mimicking other life forms (for instance, we can also build LLMs which tell us they are conscious, and we don’t use that to think humans are not sentient).
If I understand correctly, the difference in consideration you make between humans and animals seems to boil down to “I can talk to humans, and they can tell me that they have an inner experience, while animals cannot (same for small children)”.
I don’t think this is what Jeff believes, though I guess his literal words are consistent with this interpretation.
I agree I haven’t given an argument on this. At various times people have asked what my view is (ex: we’re taking here about something prompted by my completing a survey prompt) and I’ve given that.
Explaining why I have this view would be a big investment in time: I have a bundle of intuitions and thoughts that put me here, but converting that into a cleanly argued blog post would be a lot more work than I would normally do for fun and I don’t expect this to be fun.
This is especially the case because If I did a good job at this I might end up primarily known for being an anti-animal advocate, and since I think my views on animals are much less important than many of my other views, I wouldn’t see this as at all a good thing. I also expect that, again, conditional on doing a good job of this, I would need to spend a lot of time as a representative of this position: responding to the best counter arguments, evaluating new information as it comes up, people wanting me to participate in debates, animal advocates thinking that changing my mind is really very important for making progress toward their goals. These are similarly not where I want to put my time and energy, either for altruistic reasons personal enjoyment.
The normal thing to do would be to stop here: I’ve said what my view is, and explained why I’ve never put the effort into a careful case for that position. But I’m more committed to transparency than I am to the above, so I’m going to take about 10 minutes (I have 14 minutes before my kids wake up) to very quickly sketch the main things going into my view. Please read this keeping in mind that it is something I am sharing to be helpful, and I’m not claiming it’s fully argued.
The key question for me is whether, in a given system, there’s anyone inside to experience anything.
I think extremely small collections of neurons (ex: nematodes) can receive pain, in the sense of updating on inputs to generate less of some output. But I draw a distinction between pain and suffering, where the latter requires experience. And I think it’s very unlikely nematodes experience anything.
I don’t think this basic pleasure or pain matters, and if you can’t make something extremely morally good by maximizing the number of happy neurons per cubic centimeter.
I’m pretty sure that most adult humans do experience things, because I do and I can talk to other humans about this.
I think it is pretty unlikely that very young children, in their first few months, have this kind of inner experience.
I don’t find most things that people give as examples for animal consciousness to be very convincing, because you can often make quite a simple system that displays these features.
While some of my views above could imply that some humans are more valuable come up morally than others, I think it would be extremely destructive to act that way. Lots and lots of bad history there. I treat all people as morally equal.
The arguments for extending this to people as a class don’t seem to me to justify extending this to all creatures as a class.
I also think there are things that matter beyond experienced joy and suffering (preference satisfaction, etc), and I’m even less convinced that animals have these.
Eliezer’s view is reasonably close to mine, in places where I’ve seen him argue it.
(I’m not going to be engaging with object level arguments on this issue—I’m not trying to become an anti-animal advocate.)
Thanks for your response.
I’d be interested to know how likely you think it is that you could do a “good job”. You say you have a “bundle of intuitions and thoughts” which doesn’t seem like much to me.
I’m also very surprised you put yourself at the far end of the spectrum in favor of global health > animal welfare based on a “bundle of intuitions and thoughts” on what is ultimately a very difficult and important question.[1] In your original comment you say “This isn’t as deeply a considered view as I’d like”. Were you saying you haven’t considered deeply enough or that the general community hasn’t?
And thanks for the sketch of your reasoning but ultimately I don’t think it’s very helpful without some justification for claims like the following:
I also put myself at the fair end of the spectrum in the other direction so I feel I should say something about that. I think arguments for animal sentience/moral patienthood are pretty strong (e.g. see here for a summary) and I would not say I’m relying on intuition. I’m not of course sure that animals are moral patients, but even if you put a small probability on it, the vast numbers of animals being treated poorly can justifiably lead to a strong view that resources for animal welfare are better in expectation than resources for global health. Ultimately for this argument not to work based on believing animals aren’t moral patients, I think you probably need to be very confident of this to counteract the vast numbers of animals that can be helped.
I do think I could do a good job, yes. While I’ve been thinking about these problems off and on for over a decade I’ve never dedicated actual serious time here, and in the past when I’ve put that kind of time into work I’ve been proud of what I’ve been able to do.
What I meant by that is that I don’t have my overall views organized into a form optimized for explaining to others. I’m not asking other people to assume that because I’ve inscrutably come to this conclusion I’m correct or that they should defer to me in any way. But I’d also be dishonest if I didn’t accurately report my views.
Primarily the former. While if someone in the general community had put a lot of time into looking at this question from a perspective similar to my own and I felt like their work addressed my questions that would certainly help, given that no one has and I’m instead forming my own view I would prefer to have put more work into that view.
To clarify, when I asked if you could do a good job I meant can you put together a convincing argument that might give some people like me pause for thought (maybe this is indeed how you understood me).
If you think you can, I would strongly encourage you to do so. As per another comment of mine, tens of millions of dollars goes towards animal welfare within EA each year. If this money is effectively getting burned it is very useful for the community to know. Also, there is no convincing argument that animals are not moral patients on this forum (or indeed anywhere else) that I am aware of, so your view is exceedingly neglected. I think you could really do a whole lot of good if you do have a great argument up your sleeve.
Your argument that you would effectively be forced into becoming an anti-animal advocate if you convincingly wrote up your views—sorry I don’t really buy it. For example, I don’t think Luke Muehlhauser has been forced into becoming a pro-animal advocate, in the way you hypothesise that you would, after writing his piece. This just seems like too convenient an excuse, sorry.
Of course you’re not under any obligation to write anything (well...perhaps some would argue you are, but I’ll concede you’re not). But if I thought I had a great argument up my sleeve, mostly ignored by the community, which, if true, would mean we were effectively burning tens of millions of dollars a year, I know I’d write it up.
Ah, thank you for clarifying! That is a much stronger sense of “doing a good job” than I was going for. I was trying to point at something like, successfully writing up my views in a way that felt like a solid contribution to the discourse. Explaining what I thought, why I thought it, and why I didn’t find the standard counter arguments convincing. I think this would probably take me about two months of full-time work, so a pretty substantial opportunity cost.
I think I could do this well enough to become the main person people pointed at when they wanted to give an example of a “don’t value animals” EA (which would probably be negative for my other work), but even major success here would probably only result in convincing <5% of animal-focused EAs to change what they were working on. And much less than that for money, since most of the EA money is from OP, which funds animal work as part of an explicit process of worldview diversification.
I would be primarily known as an anti-animal advocate if I wrote something like this, even if I didn’t want to be.
On whether I would need to put my time into continuing to defend the position, I agree that I strictly wouldn’t have to, but I think that given my temperament and interaction style I wouldn’t actually be able to avoid this. So I need to think of this as if I am allocating a larger amount of time than what it would take to write up the argument.
I don’t think this is what Jeff said.
OK so he says he would primarily be “known” as an anti-animal advocate not “become” one.
But he then also says the following (bold emphasis mine):
I’m struggling to see how what I said isn’t accurate. Maybe Jeff should have said “I would feel compelled to” rather than “I would need to”.
To my eyes “be known as an anti-animal advocate” is a much lower bar than “be an anti-animal advocate.”
For example I think some people will (still!) consider me an “anti-climate change advocate” (or “anti-anti-climate change advocate?”) due to a fairly short post I wrote 5+ years ago. I would, from their perspective, take actions consistent with that view (eg I’d be willing to defend my position if challenged, describe ways in which I’ve updated, etc). Moreover, it is not implausible that from their perspective, this is the most important thing I do (since they don’t interact with me at other times, and/or they might think my other actions are useless in either direction).
However, by my lights (and I expect by the lights of e.g. the median EA Forum reader) this would be a bad characterization. I don’t view arguing against climate change interventions as an important aspect of my life, nor do I believe my views on the matter as particularly outside of academic consensus.
Hence the distinction between “known as” vs “become.”
You seem to have ignored the bit I made in bold in my previous comment
I don’t think there is or ought to be an expectation to respond to every subpart of a comment in a reply
It’s the only part of my comment that argues Jeff was effectively saying he would have to “be” an animal advocate, which is exactly what you’re arguing against.
So I guess my best reply is just to point you back to that...
Oh well, was nice chatting.
I guess I still don’t think of “I would need to spend a lot of time as a representative of this position” as being an anti-animal advocate. I spend a lot of time disagreeing with people on many different issues and yet I’d consider myself an advocate for only a tiny minority of them.
Put another way, I view the time spent as just one of the costs of being known as an anti-animal advocate, rather than being one.
What do yo think of the following evidence?
Rats and pigs seem to be able to discriminate anxiety from its absence generalizably across causes with a learned behaviour, like pressing a lever when they would apparently feel anxious.[1] In other words, it seems like they can be taught to tell us what they’re feeling in ways unnatural and non-instinctive to them. To me, the difference between this and human language is mostly just a matter of degree, i.e. we form more associations and form them more easily, and we do recursion.
Graziano (2020, pdf), an illusionist and the inventor of Attention Schema Theory, also takes endogenous/top-down/voluntary attention control to be evidence of having a model (schema) of one’s own attention.[2] Then, according to Nieder (2022), there is good evidence for the voluntary/top-down control of attention (and working memory) at least across mammals and birds, and some suggestive evidence for it in some fish.
And I would expect these to happen in fairly preserved neural structures across mammals, at least, including humans.
I also discuss desires and preferences in other animals more here and here.
Carey and Fry (1995) showed that pigs generalize the discrimination between non-anxiety states and drug-induced anxiety to non-anxiety and anxiety in general, in this case by pressing one lever repeatedly with anxiety, and alternating between two levers without anxiety (the levers gave food rewards, but only if they pressed them according to the condition). Many more such experiments were performed on rats, as discussed in Sánchez-Suárez, 2016, summarized in Table 2 on pages 63 and 64 and discussed further across chapter 3.
Rats could discriminate between the injection of the anxiety-inducing drug PTZ and saline injection, including at subconvulsive doses. Various experiments with rats and PTZ have effectively ruled out convulsions as the discriminant, further supporting that it’s the anxiety itself that they’re discriminating, because they could discriminate PTZ from control without generalizing between PTZ and non-anxiogenic drugs, and with the discrimination blocked by anxiolytics and not nonanxiolytic anticonvulsants.
Rats further generalized between various pairs of anxiety(-like) states, like those induced by PTZ, drug withdrawal, predator exposure, ethanol hangover, “jet lag”, defeat by a rival male, high doses of stimulants like bemegride and cocaine, and movement restraint.
However, Mason and Lavery (2022) caution:
I would expect that checking which brain systems are involved and what their typical functions are could provide further evidence. The case for other mammals would be strongest, given more preserved functions across them, including humans.
Graziano (2020, pdf):
Thanks for taking the time to expose your view clearly here, and explaining why you do not spend a lot of time on the topic (which I respect).
If I understand correctly, the difference in consideration you make between humans and animals seems to boil down to “I can talk to humans, and they can tell me that they have an inner experience, while animals cannot (same for small children)”.
While nobody disputes that, I find it weird that your conclusion is not “I’m very uncertain about other systems”, but “other systems that cannot tell me directly about their inner experience (very small children, animals) probably don’t have any relevant inner experience”. I’m not sure how you got to that conclusion. At the very least, this would justify extreme uncertainty.
Personally, I think that the fact that animals display a lot of behaviour similar to humans in similar situations should be a significant update toward thinking they have some kind of experience. For instance, a pig is screaming and trying to escape when it is castrated, just as humans would do (we have to observe behaviours).
We can probably build robots that can do the same thing, but that just means we’re good at mimicking other life forms (for instance, we can also build LLMs which tell us they are conscious, and we don’t use that to think humans are not sentient).
I don’t think this is what Jeff believes, though I guess his literal words are consistent with this interpretation.