Lyman Stone Continues Being Dumb About Shrimp

Lying Lyman!

Introduction

Well, Lyman Stone is at it again. You might think he’d have let up after writing his embarrassing critique of effective altruism where, among other things, he used aggregate data about how much people donate in the Bay Area as a proxy for how much effective altruists donate. This ignores that effective altruists are such a tiny portion of the Bay Area and Boston that even if they all gave 100% of their wealth away it would have no detectable effect on aggregate Bay Area and Boston donation data. Most embarrassingly, Stone ignored direct survey data on the question. This would be like trying to determine how much a random Californian named Fred gives to charity by looking at how much Californians give on average, when Fred’s donation history is publicly available on the internet.

He also argued that EA isn’t effective at solving malaria based on the fact that malaria hasn’t gone down much since 2015 but it was going down before. To call this—looking at bare correlations between a tiny movement and a global disease—statistically sloppy would be an insult to statistical sloppiness. When you control for population size, it becomes clear that it wasn’t 2015 when progress on malaria reversed but instead 2019—with COVID being the obvious culprit. This would be like blaming Fred for malaria increasing after 2015, when he started donating around 2011, and 2019 was when malaria really started increasing.

Anyway, Lyman has continued taking ill-thought-out and confused swings at EA, with his unique combination of statistical errors, elementary confusion about morality, and curious obsession with graphs, leading to ridiculous attempts to derive sweeping ethical judgments from survey data. This time, Lyman has written a few posts about shrimp welfare, which include the bizarre claim that animal welfare proponents are “con men,” and shrimp welfare proponents are “quite dubious persons.” Lyman adds to the growing trend of people confidently declaring that shrimp welfare is dumb, who then produce some of the worst arguments in the history of the human species.

Before I respond to Lyman’s dreck, let me outline the basic case for shrimp welfare. This is not a new case—I’ve made it in various places and responded to all the main objections. The core idea: suffering is bad because of how it feels. When people with severe mental disabilities and babies suffer, this is bad, which shows you don’t have to be smart for your suffering to be morally significant. It’s bad when anyone suffers, and this is because suffering feels bad.

There’s good evidence that shrimp and other decapods suffer. They respond to anesthetic, make tradeoffs between pain and reward, get addicted to drugs, display anxiety, and so on. Even if we’re not certain that they suffer, it’s likely enough that we should take their interests at least somewhat seriously. The most detailed report on the subject’s mean estimate was that they suffer 19% as intensely as we do.

Hundreds of billions of shrimp live in hellish conditions for their entire lives before they’re painfully killed. They’re overcrowded, ravaged by diseases, and have their eyes crushed. About half of them die before coming to slaughter. Generally, shrimp are killed by being thrown on ice where they freeze and suffocate to death. If one holds that inflicting extreme agony on shrimp is even slightly morally serious, then the fact that every year we inflict months long horrendous agony on a population four times greater than the number of humans that have ever lived is morally serious, as is the slaughter of trillions of wild-caught fish.

Fortunately, the shrimp welfare project provides stunners which stun around 15,000 shrimp per dollar the shrimp welfare project has been given. Just as it would be valuable to spend a penny to stun 150 lobsters before they’re slowly boiled to death, it’s worth spending a dollar to stun 15,000 shrimp before a painful slaughter. If we go by the earlier estimate that shrimp suffer 19% as intensely as we do, then giving a dollar to the shrimp welfare project averts about as much agony as anesthetizing roughly 3,000 people prior to slaughter. Very valuable! Thus the argument in a nutshell for giving to the shrimp welfare project is:

1. If a dollar given to an organization has a sizeable probability of averting an enormous amount of suffering and averts an enormous amount of expected suffering, it’s very good to donate to the organization.

2. A dollar given to the shrimp welfare project has a sizeable probability of averting an enormous amount of suffering and averts an enormous amount of expected suffering.

3. So it’s very good to give to the shrimp welfare project.

Doesn’t seem so crazy! It doesn’t assume utilitarianism or any particularly contentious ethical view. It just invokes the utterly commonsensical notion that extreme suffering is bad. It only diverges from common sense because most people ignore this commitment when it inconveniently implies the wrongness of common behaviors. As Matt Yglesias, noted champion of commonsense writes:

But while shrimp are obviously not adorable or as relatable as some mammals, if you saw a kid with a tank of water and some shrimp in it and he was torturing them, you’d find that to be objectionable behavior. The idea that the shrimp’s interests count for something is not that strange.

Those who think human welfare matters but shrimp welfare doesn’t matter at all owe us an explanation of what about shrimp makes their interests utterly irrelevant? Unfortunately, no plausible account has been given that doesn’t imply horrendous conclusions—the fact they’re not human isn’t enough, as Lord Of The Rings elves aren’t human either, but it would be wrong to slowly torture them to death. I’ve rebutted the common proposals here, if you’re interested.

Survey data?

Lyman doesn’t spend much time discussing this—the core argument for caring about shrimp welfare. The title of his first article is “Nobody Cares About Shrimp,” with the characteristically moderate subtitle “You can kill millions of them however you want and it's fine.” Really a champion of commonsense, holding that chucking live shrimps into the blender is totally morally permissible!

Now, careful readers might notice a bit of a gap between the claim of the title and the claim of the subtitle. The title says that people don’t care about shrimp. The subtitle says that shrimp don’t matter. But these are not the same claim, any more than claiming that 1830s Americans didn’t care much about black people’s interests is the same as claiming that slavery is fine. People can be wrong, and proponents of shrimp welfare have written at great length about why we should expect people to be wrong about subject—shrimp aren’t cute or photogenic and don’t engender strong emotional reactions, so it’s no surprise that most people discount their interests.

Bizarrely, Lyman Stone spends almost the entire article showing from survey data that people, including utilitarians, don’t care much about shrimp. This was a surprise to absolutely no one. No one was operating previously under the delusion that concern about shrimp welfare was common; the claim, rather, was always that shrimp welfare was important even if widely ignored. The fact that Lyman thinks that survey data can come close to settling a complex ethical question shows the overwhelming naiveté of his whole approach—it would be rather like trying to figure out the economic impact of tariffs by surveys.

How bad is pain?

Stone does have a few rushed paragraphs at the end explaining why he thinks shrimp don’t matter that mostly involve rehashing old and failed objections that have been rebutted at length. He first claims that “physical pain is overrated as a source of moral weights, and that more complex psychological states should be given much greater weight.” But even if one thinks that there’s a common tendency to overweight physical pain, the view that underrates it enough to ignore shrimp welfare is insane.

As described before, by our best estimates, a dollar given to the shrimp welfare project prevents about as much agony as anesthetizing 3,000 humans before they’re slowly froze and suffocated to death—so that they don’t have to experience the agony of a slow torture. Let’s be generous to Lyman and assume this is off by a factor of 100. If this is right, a dollar given to the shrimp welfare project prevents about as much agony as preventing 30 humans from experiencing the agony of being slowly tortured to death. Even if one thinks that pain is overrated as a source of moral significance, spending around 3 cents to allow a human a painless death rather than an excruciatingly painful one is obviously extremely valuable.

How simple are shrimp brains?

Lyman’s next argument is that the primitiveness of shrimp’s cognitive apparatuses make them morally unimportant. He starts by claiming that shrimp don’t have brains, which, characteristically of Lyman’s claims, is false

Lyman is right that shrimp have very few neurons. But he ignores the mountain of evidence that neuron count has virtually no correlation with intensity of experiences like pain, which is why no serious neuroscientists use number of active neurons as a proxy for intensity of experience. In many cases, there’s an inverse correlation between the number of neurons that are active and the intensity of a being’s experiences. Elephants have more neurons than humans and in some cases humans have lost significant numbers of neurons without their experiences becoming much less intense. What matters more than the number of neurons is the functions that the neurons perform—but in the case of shrimp, the neurons still perform functions similar to those involved in human pain. It’s thus rather likely that shrimp can feel pain. Given the mysteriousness of consciousness, we shouldn’t be extremely confident that shrimp have experience of limited intensity.

Death to shrimp and Lyman’s credibility

Lyman’s next article is titled “Death to shrimp,” in which he responds to some of his critics, including my good friend Amos Wollen and Theodore Yohalem Shouse. This article is perhaps even more embarrassing and shows that Lyman cannot be trusted to accurately provide the facts. In the article, tries to argue that shrimp aren’t conscious by citing a report that he says claims that. Lyman writes:

Here’s one example, identifying seven indicators of pain sensitivity/sentience. “VH” means “very high likelihood” a species meets that criteria. “VL” means “very low.” You get it.

As you can see, whereas some crabs have high or very high likelihoods for 5 criteria, shrimp have high likelihoods of just 1 or 2.

If a student has a test with 8 questions, and they get just 2 correct, they failed. Shrimp fail the test of sentience, evidence of their suffering is utterly uncompelling.

But he ignores the part of the report that says “Importantly, low/very low confidence implies only that the scientific evidence one way or the other is weak, not that the animal fails or is likely to fail the criterion.” In fact, the authors explicitly state that the reason for the low confidence designation is simply that shrimp haven’t been studied very much. When they have been studied, there’s never been any evidence of them failing a criteria:

True crabs have attracted sustained scientific attention, hence the strong evidence, whereas (for example) penaeid shrimps have barely been studied, resulting in a much weaker case. Nonetheless, there are no cases of either very high or high confidence that a taxon fails a criterion.

This was pointed out to Lyman in the comments. Someone called Forest wrote:

It seems you cropped out the part of the study that says "Importantly, low/very low confidence implies only that the scientific evidence one way or theother is weak, not that the animal fails or is likely to fail the criterion".

And then you went on to insist that shrimps fail the criterion and thus must not have sentience.

Frankly this seems to be intellectually dishonest.

In response to this, Lyman wrote:

No, I saw it. And I laughed, because it’s a dishonest framing. There is no evidence for the claim, so no reason to believe it’s true. It fails the test, their hemming and hawing aside.

Well perhaps there’s “no evidence” if you don’t find induction or comparisons with other analogous species to be evidence and you ignore the universal results of the studies that have been done. But of course this would be ludicrous. If no study has been done on whether shrimp meet some criteria, you shouldn’t declare this to be strong evidence that they fail the criteria. This is either dishonest or braindead—hard to tell which.

Lyman also ignores the abundant evidence for shrimp consciousness. If we go by studies on other decapods, each of which are quite similar to shrimp, there’s abundant evidence for them possessing many plausible correlates of consciousness, as the chart below shows:

Lyman then repeats his argument about shrimp having few neurons, once again ignoring the abundance of evidence that neuron count is a bad proxy for moral significance. Perhaps the shrimp aren’t the only ones with few neurons…

His next claim was that, because shrimp have poor memories, helping out humans is more important, as their suffering has long-lasting negative effects. Now it’s true that trauma makes suffering worse. But extreme suffering is bad even if it causes no trauma. As described before, if you could spend a dollar to anesthetize 3,000 babies before slaughter, this would be a very good use of a dollar. Lyman claims that suffering you remember is far more significant by relying on the deeply crazy claim that the long-lasting effects of suffering are, per day, 10% as bad as the suffering itself. This is crazy—if you were tortured by and then escaped from a serial killer, the lasting trauma over the course of ten random days would be obviously much less bad than one day of horrendous torture at the hands of the serial killer.

A few miscellaneous claims

Next Lyman argues that opposition to pain is not a human universal. Some cultures undergo rituals where they get tattoos because it’s painful. But caring about shrimp welfare doesn’t require thinking that pain is always bad—perhaps some kinds of pain are good. Rather, it only requires thinking that extreme horrendous agony that serves no further purpose is by default morally serious. No one thinks, for instance, that we shouldn’t anesthetize people before surgery on the grounds that opposition to pain isn’t a human universal. Pain being good is generally the exception not the rule, and when it is good, it’s because it brings about something else that’s valuable.

The latter bit of Lyman’s article is even more bizarre. He presents survey data showing that people concerned about shrimp welfare are slightly more likely to be anti-natalists. But obviously demographic data about whether some view correlates with some other obscure view doesn’t tell you whether the first view is right. This argument is bad enough that I couldn’t make it up.

Lyman rejects the claim that we’re biased by the small size of shrimp on the grounds that “you should value big creatures more than small ones.” But some fact can bias us even if it correlates with greater moral worth. Maybe better looking people are smarter, but being good looking might also bias us to think someone is smarter. If insects were larger, we’d be less reluctant to kill them for trivial reasons.

Lyman next has a very confused section titled “Scope Insensitivity Cuts Both Ways.” Humans display a phenomenon called scope insensitivity—because we have no intuitive sense of large numbers, our intuitions systematically distort the value of large amounts of stuff. This bias is so extreme that humans will pay almost exactly the same amount to help 2,000, 20,000, or 200,000 birds. This can help explain why we underestimate the seriousness of the industrial scale slaughter of trillions of shrimp. Lyman writes:

But people also don’t handle small numbers well! If you ask people “How much would you say a single shrimp life is worth?” they will give you a number which, if you scale it by 20 trillion, will be insane, because people can’t comprehend 20 trillion! There’s literally no way to solve scope insensitivity: if you use peoples’ low-scale tradeoffs, you cannot extrapolate to larger scales, because elasticity calculations in all economic tradeoffs are infamously scaling-variant!

This is badly confused. You can actually just multiply the value of a shrimp by 20 trillion to find the value of 20 trillion shrimp. This will, of course, turn up the result that 20 trillion shrimp matter a lot, but that’s correct. Unlike goods going into demand curves, the welfare of others doesn’t have declining marginal utility.

Lyman then talks about some of the objections to relying on survey data to address ethical conclusions. In Amos’s article criticizing Lyman, he points out that if you surveyed most people on the repugnant conclusion—whether a world with 10^100 just barely happy people would be better than one with 10 billion people with amazing lives—they’d find it repugnant. But this doesn’t settle the question of whether it’s right; you have to consider the arguments on both sides. Lyman notes that a survey hasn’t been done on whether people accept the repugnant conclusion—this is right, but it’s obvious how it would go if it was done.

In fact, one can’t simply rely on the methodology of always adhering to a widespread consensus because there’s often a widespread consensus in contradictory propositions. Most people would, for instance, hold that a world with infinity galaxies each filled with 10 billion happy people and one sad person would be better than if the galaxies each had 10 billion sad people and one happy person. But most people would also hold that simply moving people around without affecting their welfare can’t make a world arbitrarily better. Unfortunately, these judgments are in conflict. I haven’t surveyed people on this question, but it’s obvious how most people would answer.

Lyman addresses one objection to his view which is that it would imply that slavery, something believed to be permissible for almost all of history, was fine. He says that in the antebellum south lots of people opposed slavery. But this ignores the earlier history; as of the year 1600, there were almost no people who thought slavery was impermissible in principle.

Fraud?

Lastly, Lyman says:

Shrimp Welfarists Must Give Their Number Or Be Seen As Frauds

How many shrimp would you trade for a human? It’s that simple. If you want to caveat species of shrimp or traits of a human you can. But you cannot seriously be out there arguing for diverting charitable funds from effective human-related causes and then not be on record on how many shrimps are worth a person’s life to you, because you are trading off against human lives!

This is a rather strange claim. Suppose a person suggests that people should give their money to give clean water to poor children in Malawi. Should they have to give the exact number of people whose being given clean water is as valuable as saving one person’s life. If they don’t do this, are they a fraud?

But fine, I’m happy to answer. My view is that a shrimp’s life is not very valuable—they’re short and don’t have much that’s good. Likely they have negative lives, so I’d probably save one human over infinite shrimp because it’s likely good when shrimp die. Though, of course, I’d give it a lot of thought before making the decision.

But the shrimp welfare project doesn’t save shrimp from death but instead from agony. If I was given the choice between preventing a human from experiencing some painful experience or preventing some number of shrimp from experiencing that painful experience, I’d be indifferent at about six shrimp, assuming it would leave no lasting trauma. If the experience would leave the humans with lasting trauma, then it would be some bigger number—it’s probably worse, for instance, to sexually abuse one human than even hundreds of shrimp.

There’s a strong argument that the exchange rate between human and shrimp welfare can’t be infinite. Humans evolved from beings very much like shrimp. If we assume each generation has a non-infinite exchange rate with the one before it, then this entails that the exchange rate between human welfare and shrimp welfare is not infinite.

The last article

Stone’s final article is perhaps the most bizarre. In it, Lyman explicitly calls people concerned about shrimp welfare “frauds” based on almost comically sloppy reasoning.

The first part of Lyman’s article is about how, if we take shrimp welfare seriously, then we should support things that are bad for humans but good for shrimp. Shrimp welfare then eclipses human welfare in importance—it would be worth rolling back most improvements to human welfare to improve the well-being of shrimp. Worse, he claims this would mean that because of humanities deleterious impact on wild animals, the world would be getting worse, not better.

Now, one could, of course, dispute his numbers. I dispute that humans have had a negative impact on wild animal welfare—I think it has likely been positive. But I’m very dubious of these kinds of arguments. It wouldn’t be at all surprising if we lived in a very morally weird world, where weird-sounding stuff matters more than publicly salient issues.

We should expect morally weird judgments if the world is morally weird. Almost anything that’s morally serious could be the worst thing in the world if there was enough of it. If every single person broke several bones per week, it wouldn’t be counterintuitive to think that this was the worst thing in the world. But by the same logic, if the world has almost unimaginable quantities of shrimp death, we shouldn’t be shocked if that’s the worse thing in the world.

Let’s be super conservative and say that a shrimp being farmed for its whole life, tortured, and killed is one ten-thousandth as bad a person breaking a bone. Your odds of breaking a bone per day are about one in twelve-thousand—so this would mean that the odds of you breaking a bone today is about as bad as if a shrimp was farmed for its entire lives and then killed. This would mean that you should farm for months in hellish conditions (which include their eyes getting stabbed out) and then painfully slaughter a shrimp if doing so would guarantee you wouldn’t break a bone today. But even by these ridiculously conservative estimates, global shrimp farming for a year is about as bad as 44 million people breaking their bones every year. If there was some phenomenon that resulted in 44 million people breaking their bones every year, that would be one of the worst things in the world—even by insanely conservative estimates shrimp farming is equally serious.

If we go by more reasonable—though still insanely conservative—estimates, according to which a shrimp being farmed for its whole life and then killed is as bad as a person having a one in a thousand chance of breaking a bone, then shrimp farming is as bad as 440 million people breaking bones every year. This estimate is still so conservative that it implies that you should farm a shrimp in torturous conditions for several months, and then freeze and suffocate it to death, if doing so would guarantee that you wouldn’t break a bone in the next ten days.

If we go by our best estimates—that shrimp suffer about 19% as intensely as humans—and assume (to make the math convenient) that 19% of the suffering of shrimp comes from their slaughter, then shrimp farming is as morally serious as a disease that affected 440 billion people per year and caused them to endure an experience as painful as being slowly suffocated to death. This would mean that humans would have to experience the pain of slow suffocation about once every 7 days!

The reason that shrimp farming is worse than we’d expect is that there’s more of it than we expect. If you take almost any bad thing and then produce 440 billion instances of it, that will almost inevitably end up being one of the worst things in the world. This disease would obviously be much more significant than most of the world’s other problems.

Lyman’s a pro-natalist. I am too! Now, it turns out that on one of the leading theories of physics, random innocuous actions cause the universe to split huge numbers of times—producing exact copies of the universe. If this is true and pro-natalism is right, then the primary determinant of the significance of our actions is their impact on universe splitting.

Is this a reason to reject pro-natalism? No! It just means that the world is very weird, and so the right ethical view might turn out to have strange implications regarding what we should do in our deeply weird world.

Lyman’s also a Christian who believes in eternal hell. Given that salvation lasts forever, it would be probably worth destroying industrial civilization if it would cause a few extra people to be saved. On Christianity, the near-exclusive determinant of how good a world is is how many people are saved. It might therefore be that various seemingly horrific events end up being for the best if they cause more people to be saved.

We shouldn’t reject a view outright because it implies super weird practices conclusions—like that shrimp farming might be on the order of badness of all human suffering. Especially given that:

1. Given scope neglect and shrimp looking weird, we’re likely to systematically underestimate the badness of shrimp farming.

2. If you were acting behind the veil of ignorance, equally likely to be born into the world as any conscious being, you’d regard shrimp suffering as very significant. You’d be much more likely to be born as a farmed shrimp that gets tortured for its entire life than a person.

3. If you lived the life of every creature on Earth, you’d spend far more time as a shrimp than a human. You’d begin to care quite a lot about shrimp welfare. What our priorities would be if we experienced everyone’s perspective is a much better guide to what matters that what we actually care about given that we only experience our own.

4. Shrimp farming likely causes much more suffering every few years than has existed in all of human history even by rather conservative estimates. A population four times greater than the number of humans who have ever lived is farmed and slaughtered each year. Let’s be really conservative, ignore all wild caught shrimp, and suppose that a shrimp painfully dying experiences .1% the pain of a human painfully dying and that half of the suffering of farmed shrimp comes at their death. Well, even by these extremely conservative estimates, shrimp farming causes about as much suffering as the painful deaths of about 880 million people per year. In light of this, thinking that shrimp are the only issue isn’t that crazy.

Imagine we discovered that bacteria, rather than being non-sentient automata, were in fact complex, morally significant creatures, who could feel joy, and love. They read books and possessed roughly the cognitive sophistication of humans.

Such a world would be insanely ethically inconvenient. It would imply that the main determinant of the ethics of our actions is how they affect the roughly 10^30 bacteria on earth. It would be worth grinding civilization to a halt to slightly reduce the scale of mass bacteria slaughter.

In response to this, a person in this world could argue that bacteria don’t matter. They could call proponents of bacteria rights conmen, and argue that taking bacteria seriously has wildly revisionary ethical implications, that our treatment of microbes matters more than our treatment of other humans.

That person is Lyman Stone.

Yes, taking shrimp seriously leads to some weird because we routinely kill trillions of them for trivial reasons. But that’s not a bug (pun intended). We shouldn’t be surprised if the true moral view produces surprising results regarding the real world.

The last section of his article is titled Animal Welfarists Are Con-Men. In it, Lyman writes:

If animal welfarists believed what they say, they would not just be vegans. They would be uniformly ardent anti-natalists, and also advocates of forced sterilization.

I don’t think this is right. Humans are the only hope of wild animals in the long term. The development of civilization has the potential to reduce longterm wild animal welfare. Even in the short term, it’s very unclear how we’ve affected wild animals overall—by reducing their population sizes, we might be reducing suffering a lot. In addition, I think the far future could be very good and outweigh present suffering.

If it turned out that taking animal welfare seriously implied anti-natalism, then I don’t think that would be such a defect. In the world where bacteria were conscious and intelligent, it might turn out that we were morally required not to have kids because humans mass-kill bacteria. I’m skeptical that the real world was like that, but if it was, that would just be one other instance of us living in a morally weird world.

Do we have other evidence animal welfarists are con men? Yes, we do. Organizations like PETA largely lie about how they treat animals and what they do with funds.

I haven’t investigated the PETA issue, and I don’t trust Lyman to accurately present the facts, but obviously the fact that there’s one bad animal rights organization tells us little about whether animal rights activists are scammers in general.

So this is not shaping up to be a big con. Where’s the big reveal?

Well, it’s more of a subtle reveal: only two companies make shrimp stunners, a British and a Norwegian company. One of the companies has investment stakes from green-specific investors, the other doesn’t list investors. The Shrimp Welfare Project is UK-based. The Shrimp Welfare Project is funded mostly by individual donations and grants from animal welfarist organizations. What isn’t clear to me is if the companies themselves making the shrimp stunners might be donors to the animal welfarist causes, effectively making the SWP a lobbying arm aiming to achieve regulatory capture of global aquaculture for monopolistic European equipment suppliers.

What a ridiculous claim. Lyman calls the shrimp welfare people conmen because…they live in the same country as the group producing stunners. Thus, he thinks maybe they’re bought off by Big Humane Stunning based on literally no evidence. In fact, you can literally look online; most of their funding comes from Open Philanthropy.

By the way, totally unrelated, Lyman Stone is funded by David Duke. My evidence for this: he and David Duke are in the same country. Totally ironclad logic.

Stone also accuses me of grifting—only writing about shrimp welfare to get engagement. His evidence for this is that my animal welfare posts seem to get more engagement than other posts. I can assure you: if my aim was engagement, I would write about politics, the source of most of my most popular articles, not shrimp welfare. If I cared about engagement and money, I wouldn’t offer free subscriptions to people who donate to the shrimp welfare project.

Lyman is right that taking shrimp welfare seriously has revisionary implications. But we shouldn’t be shocked by ethical theories having revisionary implications. The world is often morally inconvenient, implying that the stuff we care about doesn’t matter much in the grand scheme of things. Taking seriously extremely obvious ethical principles about the badness of extreme suffering makes the case for caring about shrimp obvious and straightforward.

Conclusion

Well, that’s the end. As should, I hope be obvious at this point, the standard objections to caring about shrimp welfare aren’t any good. The only one with any force is about the revisionary ethical judgments that taking shrimp welfare seriously forces us to adopt, but as the last section argued, even that isn’t a substantial argument. The world might just turn out to be morally weird. The reason shrimp farming is morally serious is that a population 55 times the human population is tortured and killed every year. Almost all bad things would inevitably dominate our ethical calculations if there were 440 billion instances of them.

Despite the overwhelmingly compelling logical case for caring about shrimp welfare, few people do. Why? The answer is straightforward: shrimp welfare seems weird and shrimp are not photogenic. Because people are mostly indifferent to ethical appeals that fail to resonate with them, hundreds of billions of shrimp are painfully consigned to a horrendous fate for the sake of trivial human pleasure. Though shrimp welfare easily wins in the marketplace of ideas, you just can’t get people to care.