I’m a Senior Researcher for Rethink Priorities and a Professor of Philosophy at Texas State University. I work on a wide range of theoretical and applied issues related to animal welfare. You can reach me here.
Bob Fischer
Karma: 5,355
I know the OP may not read this comment. I made it on his Substack post and I’m sharing it here in case it’s of interest to others on the Forum.
Thanks for your post, Rob. Meghan Barrett and I have a detailed reply to Eisemann et al. 1984 in the Quarterly Review of Biology. You can see it here:
Short version, very little in that paper has stood the test of time and the particular passage you quote has many problems. I’d encourage you to reconsider including it!
Hi Rob. A few more thoughts. I grant you that the evidence for sentience in PWS is thin and I certainly don’t want to suggest otherwise. Still, I’m not sold on some of the bases for skepticism that you’ve flagged.
First, you suggest that the miniaturized hemiellipsoid bodies in PWS provide evidence against integrative processing of the sort relevant to pain. That inference is plausible in some contexts, but not all. Small structures can be highly functional, and in small animals “miniaturization” is often just what efficiency looks like rather than what loss looks like. After all, selective pressures have done that in many other species and current breeding practices are, effectively, intense selective pressures: they might push toward compact integrative circuitry rather than cause the loss of integration. In addition, morphology is a weak proxy for function. For instance, people used to think that the mushroom bodies in insects were solely for processing olfactory information, largely because the olfactory afferents were obvious and the visual inputs weren’t. We now know that visual information also reaches the mushroom bodies and that these regions integrate multiple sensory modalities and support cross-modal learning. Even in very well-studied taxa, we’ve repeatedly underestimated what regions are doing. (Moreover, insofar as we know what the hemiellipsoid bodies in crustaceans are for, the comparative neuroanatomy suggests that variation in their size and complexity may track the amount of olfactory input, which has nothing to do with sentience. See the end of this paper for details.) Finally, even if the hemiellipsoid bodies are small relative to our expectations, other integrative regions—like the medulla terminalis and structures plausibly homologous to the insect central complex (central body + protocerebral bridge)—aren’t obviously reduced in the same way. So, we should be wary of placing much weight on this neuroanatomical point.
Second, I worry that the post puts too much weight on that one negative behavioral result—namely, the failure to observe directed grooming/rubbing in PWS under extreme pH stimuli. First, null results in animal behavior are notoriously common. Second, there are some additional reasons for caution in this particular case. For example, the authors used extremely high concentrations of NaOH and HCl for two species because lower concentrations produced no response in preliminary trials; they also explicitly note that such pH levels may be ecologically irrelevant, which raises the possibility that the stimulus wasn’t being processed in the expected way at all. On top of that, the behavioral coding differed from the studies they were attempting to replicate: for example, they didn’t count antennae contact with the tank wall as grooming, whereas earlier work did. More broadly, directed grooming is only one possible self-protective response. In many taxa, pain manifests as withdrawal, guarding, reduced feeding, or altered exploration. Third, animals are sensitive to many kinds of damage. Even if we don’t see responses to pH, it would be shocking if there weren’t any sensitivity to mechanical injury, for instance. Taken together, this makes it hard to treat the null grooming result as a strong strike against pain-like processing in PWS.
Third, you’re skeptical that nociception plus pharmacological modulation should shift our priors much. But granting that anesthetics can suppress movement in a fairly general way, pharmacological modulation is still one of the classic tools used to distinguish mere reflexive responsiveness from centrally mediated aversive processing in animals. So I think there’s some risk of people reading the post as saying “nociceptors, therefore basically nothing,” when a more accurate characterization is that we have at least some of the kinds of markers that pain researchers take seriously in other contexts.
Finally, while you’re right to caution us with respect to extrapolating across taxa, that’s abandoning one of the only tools available in our current, data-starved situation: phylogenetic inference. Decapods probably share a lot of conserved neural and behavioral machinery, and many pain-relevant capacities—learning, motivational tradeoffs, and injury-related protective behavior—appear across the group. Given that, the default evolutionary expectation isn’t obviously that pain-like processing appears in crabs and lobsters but is absent in shrimp unless we have strong reasons to think a major functional transition or loss event occurred. And I’m not sure we do. So, yes, taxonomic differences matter, and suborder distinctions may well track meaningful neurobiological differences, but they don’t by themselves defeat inference from related taxa.
Again, none of this is to claim that the evidence for pain in PWS is already strong or that skepticism is unreasonable. Just trying to push back a little on how skeptical we should be.
Thanks for explaining!
Interesting, Vasco. I wouldn’t have guessed that this has much to do with hedonic capacity at all. Endotherms sacrifice energy efficiency for thermal independence; ectotherms sacrifice thermal independence for energy efficiency. But these traits don’t obviously have much to do with the cognitive capacities of the animals in question. Would you say more about your hunch?
Thank you for sharing this, Stien. I’m grateful for your candor.
Sorry, Vasco; we weren’t clear. The idea is to use the DCM as a blueprint for aggregating the data we collected in the MWP, not to produce new estimates of sentience. The focus would be on capacity for welfare.
Thanks, Jess. And great question. This is a little difficult to assess because of standard assumptions in the discipline. For instance, the lore is that the most humane way to both anesthetize and euthanize bugs is to throw them in the freezer, even though invertebrate veterinarians question this. As it happens, that’s also the most convenient thing to do. So, we don’t have a situation where there is agreement that some alternative would be better for the bugs, but people do the suboptimal thing regardless. Likewise, when people choose to do live dissections and other highly aversive procedures, they often say that they have to do it because a reviewer is going to insist on it (because that’s the way it’s been done before and so live dissection is critical to getting comparable data or whatever). So people don’t conceive of themselves as having options where they really can choose a more humane alternative.
In any case, you are right to suggest that the average entomologist is not willing to take on huge inconveniences to do non-aversive work. But I do think an increasing number of them, particularly the under-40 crowd, are willing to take on some inconvenience, as shown by their interest in humane endpoints, reducing bycatch, learning about better husbandry options, etc.
Strongly agree about “the evidential situation with respect to comparing the individual welfare per animal-year”! I’ve always taken the numbers from the MWP much less seriously than others. I see that work as one part of a large picture, depending heavily on other arguments.
And thank you for voting for Arthropoda!
Thanks, Nick. A few quick thoughts:
It’s reasonable to think there are important differences between at least some insects and some of the smaller organisms under discussion on the Forum, like nematodes. See, e.g., this new paper by Klein and Barron.
I don’t necessarily want to give extra weight to net harm, as Michael suggested. My primary concern is to avoid getting mugged. Some people think caring about insects already counts as getting mugged. I take that concern seriously, but don’t think it carries the day.
I’m generally skeptical of Forum-style EV maximization, which involves a lot of hastily-built models with outputs that are highly sensitive to speculative inputs. When I push back against EV maximization, I’m really pushing back against EV maximization as practiced around here, not as the in-principle correct account of decision-making under uncertainty. And when I say that I’m into doing good vs. doing good in expectation, that’s a way of insisting, “I am not going to let highly contentious debates in decision theory and normative ethics, which we will never settle and on which we will all change our minds a thousand times if we’re being intellectually honest, derail me from doing the good that’s in front of me.” You can disagree with me about whether the “good” in front of me is actually good. But as this post argues, I’m not as far from common sense as some might think.
FWIW, my general orientation to most of the debates about these kinds of theoretical issues is that they should nudge your thinking but not drive it. What should drive your thinking is just: “Suffering is bad. Do something about it.” So, yes, the numbers count. Yes, update your strategy based on the odds of making a difference. Yes, care about the counterfactual and, all else equal, put your efforts in the places that others ignore. But for most people in most circumstances, they should look at their opportunity set, choose the best thing they think they can sweat and bleed over for years, and then get to work. Don’t worry too much about whether you’ve chosen the optimal cause, whether you’re vulnerable to complex cluelessness, or whether one of your several stated reasons for action might lead to paralysis, because the consensus on all these issues will change 300 times over the course of a few years.
Thanks, Michael. I’m quite sympathetic to the idea of bracketing!
Lastly, this article is good. The possibility the they’re right is one of the things that makes me inclined to see insects as the limit case.
Thanks, all. Let me add something that may help clarify why we’re always at loggerheads. I’m not actually thinking about these questions in probabilistic terms at all. In my view, the evidential situation for most arthropods is so sparse that I don’t actually believe we’re in a position to assign meaningful probabilities of sentience—even extremely rough ones. We’re squarely in the domain of the precautionary, not the probabilistic. When the evidence is this patchy and the mechanisms this poorly understood, numerical probability assignments feel more like artifacts of modeling choices than reflections of the world. So, when I talk about “robustness,” I’m not covertly appealing to narrower or wider probability distributions; I’m saying that the entire framework of attaching numbers to these uncertainties feels inappropriate.
This is one of several reasons why focusing on well-studied insects makes sense to me. It’s not that I think BSF larvae are 10× or 100× more likely to be sentient than springtails. It’s that we have a type of evidence for some insects—convergent behavioral, physiological, and neuroanatomical findings—that simply doesn’t exist at all for mites, springtails, and nematodes. And without that evidential base, I’m wary of using a first-pass model to set priorities. Expected value becomes extremely fragile under those conditions, as the inputs aren’t grounded: they’re guesses stacked on guesses.
So the way I think about prioritization has less to do with estimated probabilities and more to do with where precautionary reasoning can actually get traction. Work on farmed and research arthropods produces immediate welfare improvements, helps develop welfare indicators, and builds the scientific ecosystem we’ll need if we ever hope to understand smaller arthropods. That’s a much more stable basis for action than trying to set priorities via BOTECs.
Anyway, we’ll just have to agree to disagree, as we just keep running up against the same issues over and over!
I’m sorry that I don’t have time to respond to all your questions, Vasco. The short version, though, is that I also want robustness in the case for sentience, so I’m much less inclined to make the kinds of extrapolations you’re suggesting here. I have the same view about our moral weight work: I put very little stock in any specific numbers, as I think that plausible moral weights will be defensible from several angles, each of which will suggest somewhat different estimates, with no obvious right way to aggregate them. (Again, there’s that skepticism about expected value!)
Caring about Bugs Isn’t Weird
Arthropoda is a 501(c)(3). As this thread indicates, Mal Graham and I run the organization. We keep a lean profile because many science funders keep a lean profile. I realize that it isn’t optimal for fundraising, but I think it’s normal enough from the perspective of our grantees. If you’d like to discuss further, happy to chat.
I agree with Mal about Arthropoda being a good bet for this work. RP would be good too. On the macro-level issue of priorities, I’ve gathered some of my thoughts here.
Finally, I’ll say publicly what I’ve said privately: thank you for supporting Arthropoda. It means a lot to me that you donated.
Thanks, Vasco! Abraham’s post covers many more farmed insects than BSF and mealworms. (For instance, the lower end of his farmed cochineal estimate is 4.6T deaths annually.) When you include those other species, I think the “rounding error” claim becomes more plausible. (Sorry not to be clear in the post: I probably gave the impression that I was only thinking of the standard “insects as food and feed” species.)
In the interest of clarity, I’ve updated the original post in response to @Hugh P’s helpful question.
Yes: if you’re right! But that’s an awfully big bet. As you might expect, it isn’t one I’m prepared to make. And I’m not sure it’s one you should be prepared to make either, as your credence in this view would need to be extremely high to justify it.
In any case, thank you for the detailed reply. I have a much better understanding of our disagreement because of it.
Thanks for your comment, Dennis. One worry here is that you might be holding work on animal minds to an impossible standard. Yes, no one has a way to detect qualia directly, but surely we can make some inferences across the species boundary. Minimally, it seems very plausible that Neanderthals were sentient—and they, of course, were not homo sapiens. What makes that so plausible? Well, all the usual evidence: behavioral similarities, neurophysiological similarities, the lack of a plausible evolutionary story about why sentience would only have emerged after Neanderthals, etc.
Admittedly, plausibility decreases as phylogenetic distance grows (though the rate of the change is up for debate). Still, our epistemic situation seems to differ in degree, not in kind, when we consider stags—and, I submit, stag beetles.
One way to appreciate the value of the evidence that you’re criticizing is to imagine it away. Suppose that Bateson and Bradshaw had not found “the measurable quantities denoted… by the words ‘stress’ and ‘agony’ (such as enzyme levels in the bloodstream).” Surely it would have been less reasonable to believe that stags suffer in those circumstances. But if it would have been less reasonable without that evidence, it can be more reasonable with it.
Rob has a thoughtful reply here.