The case for caring about shrimp

We all know that it’s better to save five people’s lives than to save only one. But in 1977, one philosopher dared to argue…maybe it isn’t?

“Should the Numbers Count?” by John Taurek is among the few modern philosophy papers that might fairly be described as infamous. When I was taught it as an undergrad, it was presented as something between a cautionary tale and a punching bag, a set of dubious arguments in favor of a conclusion so absurd that it’s astonishing a respected UCLA professor put his name to it. The most prominent reply, from famed Oxford moral philosopher Derek Parfit, was simply titled, “Innumerate Ethics.”

Taurek asks the reader to imagine a situation in which there is “a supply of some lifesaving drugs. Six people will all certainly die if they are not treated with the drug. But one of the six requires all of the drug if he is to survive. Each of the other five requires only one-fifth of the drug.” What should be done?

Most people, Taurek concedes, will conclude that dividing the drug supply five ways, and saving five lives, is better than giving it all to the sixth person (whom he names David). But to conclude this is to make a mistake, he says. Implicit in the idea that the numbers count, he argues, is a belief that you can sum up suffering and happiness between different people, so that the suffering of five people “adds up” to more than the suffering of one.

“Suffering is not additive in this way,” he insists. David dying is bad for David. One of David’s five rivals for the drugs dying is bad for that person. There is no such thing as “bad for the world” or “bad, full stop.”

“I am not to compare [David’s] loss, on the one hand, to the collective or total loss to these five, on the other, whatever exactly that is supposed to be,” he concludes. “Rather, I should compare what David stands to suffer or lose, if I do not prevent it, to what will be suffered or lost by any other person, if I do not prevent that.”

None of the five others will suffer more by dying than David would. Thus, Taurek claims, the drug’s owner should not reflexively save the five instead of David. She should, rather, flip a coin: heads the five live, tails David lives. That is the best way to show equal concern for each person.

When I first read Taurek, my reaction was: Is this guy fucking with me? Would he flip a coin not between one and five but between one and a million? A billion? Would a world leader be justified in allowing a nuclear strike to go forward, if doing so saved his best friend and no one else? What exactly is wrong with this man?

I, apparently, was not alone in this reaction. Parfit — who was legendarily even-tempered and courteous, especially for a philosopher — was made so furious by the argument that by the end he was reduced to lecturing Taurek the way one would a preschooler: “Why do we save the larger number? Because we do give equal weight to saving each. Each counts for one. That is why more count for more.”

And yet over the years I have encountered a few philosophers and philosophy-adjacent folks who are, if not totally on board with Taurek, at least Taurek-curious. They’re skeptical that the numbers count, the way I intuitively feel they must count.

I didn’t understand, really, where such a person could possibly be coming from. I didn’t understand, that is, until the shrimp.

Let’s say that, contra Taurek, the numbers do count. Here are a few numbers.

There are, as of this writing, roughly 8.1 billion human beings on Earth. Per the research group Faunalytics, humans worldwide killed about 310 million cows for meat in 2023; 480 million rabbits; 520 million turkeys; 540 million goats; a little under 700 million sheep; 790 million geese; and 1.5 billion pigs. We also killed 4.2 billion ducks and 78 billion chickens.

What that means is that we slaughter something like 3.5 billion mammals a year, and over 20 times as many birds.

But just as there’s a gap between mammals and birds, there is an ever bigger gap between birds and fish. No one knows with certainty how many fish humans kill each year. One recent paper estimated the number of “finfish,” as distinct from shellfish, killed on farms in 2019 at between 78 and 171 billion. Even the low-end number would equal the number of chickens killed every year, meaning the total number of fish deaths almost certainly swamps that of land animals. And that’s just farmed fish. Another paper by two of the researchers from the farmed fish paper puts the number of wild-caught finfish at an average of 1.1 to 2.2 trillion per year.

What of shellfish, though? The research group Rethink Priorities estimated recently that roughly 440 billion shrimp are killed on farms annually. What’s more, they expect this number to balloon to over 760 billion by 2033, based on projections from the Organization for Economic Cooperation and Development and the Food and Agricultural Organization. Over eight shrimp will die for every chicken slaughtered that year, they forecast.

So those are the numbers. Now, suppose you care about animals’ welfare, or at least think humans have some kind of duties to the animals we raise in farms or take from the wild to feed ourselves. Suppose further that you think shrimp count even a little bit — not as much as a human, of course, or a cow, or even a trout, but they still count in some way as animals capable of feeling pain and worthy of some consideration.

If the numbers count, then surely it follows that the most pressing matter in the world of animal rights is the plight of the shrimp.

It is all well and good for me to furiously insist in a philosophy seminar that John Taurek is a madman and of course the numbers should count. But if that is so — should these numbers count? Does the seemingly basic conclusion of wanting to save five humans ahead of one commit me to a kind of totalizing shrimp fanaticism? How far down this road am I willing to walk?

Andrés Jiménez Zorrilla has walked down this road at least a little ways. In 2020, he left a career in private equity to cofound Shrimp Welfare Project, which is exactly what it sounds like. (Like the band Pixies, Shrimp Welfare Project eschews the definite article.)

When I asked Jiménez Zorilla about his switch, some of his explanations feel like what someone would say when leaving a finance job to work at a soup kitchen, or quitting a lucrative plastic surgery practice to help civilians injured in war. He just wanted to do some good. “My wife works with refugees, and I started to compare what she was doing and what I was doing,” he told me. “At some point I decided I should be doing something beyond just making someone else richer.”

He came across Charity Entrepreneurship, an effective altruism-aligned group that helps people create new charitable organizations targeting neglected problems. The group matched him up with a cofounder, Aaron Boddy, and gave them a menu of serious problems not currently attracting much charitable attention. One of the items on the menu was “shrimp welfare.” It’s a classic effective altruist idea: A cause that is important (440 billion shrimp a year!), neglected (no one else was working on shrimp welfare), and tractable (precisely because no one was working on it, there were likely easy ways to improve shrimps’ lives that no one had tried yet, even if those ways weren’t immediately obvious).

At first, Jiménez Zorilla recalled, “I thought, ‘the effective animal advocacy folks have really lost their minds.’” Then he kept reading. And he saw the massive number of shrimp being farmed every year. And the evidence that shrimp are sentient: that they are, at the very least, able to feel pain, able to suffer. And the fact that literally no one on planet Earth seemed to be working on this issue.

Before he knew it, he was founding a shrimp welfare group, to his knowledge still the only group singularly dedicated to the animals in the world.

Talking to Jiménez Zorilla is very effective at bringing one’s views of the shrimp dilemma down from the philosophy seminar to the ice slurry. That, the ice slurry, is one of the major methods through which shrimp are killed, or less killed than transported while slowly dying. Upon reaching maturity, some of the shrimp farms that Shrimp Welfare Project examined in Vietnam, one of the world’s leading shrimp producers, transport them alive to processing plants. Others put them on the ice slurry, which is meant to both kill them and keep their corpses preserved for transport.

In theory, the low temperature stuns them before they die, enabling a less painful demise. We know that subjecting shrimp to cold ice or water reduces their activity. But it’s not clear that this means they are stunned, and no longer experiencing pain, as opposed to paralyzed: feeling pain but unable to move. It is entirely possible that the shrimp in the slurry are frozen to death slowly, feeling the whole thing.

What of the shrimp transported alive? Many of them were crushed to death, the report’s authors, Trinh Lien-Huong and Nguyen Tran, conclude: “In many cases, the … containers contained a very small amount of water or ice slurry, with animals crowding inside. Shrimps and prawns suffered from asphyxia and weight crushing in these situations.”

When Jiménez Zorilla and Boddy did their own investigation at Indian shrimp farms, they found that 95 percent of the farmers they interviewed believed their animal wards could feel pain. “In most cases farmers would go on to explain how they try to relieve the suffering of shrimps using medicines or improving water or feed quality,” they write. One farmer told them that “when shrimp were stressed, he would attempt to improve water quality ‘to make the shrimp feel free.’”

While there is less scientific evidence on the mental state of shrimp than that of other decapods, like crabs, a 2021 review by London School of Economics researchers commissioned by the UK government found evidence backing these farmers up. The authors reported high confidence that penaeid shrimps, the most commonly commercially farmed variety, have “nociceptors,” or neurons that can respond to external stimuli that can hurt the shrimp. They also reviewed some studies showing that shrimp respond to painkillers by becoming calmer, and grooming the hurt area on their bodies less — more signs of the capacity to feel pain.

The evidence is thin, the review concluded — but mostly because it’s been barely researched. Whenever crustaceans have been closely examined by scientists, strong evidence of sentience is invariably found. The authors, including eminent sentience researcher Jonathan Birch, conclude that the UK should treat all decapods, including shrimp, as sentient animals.

Shrimp Welfare Project’s remedies are quite modest. The group offers shrimp farms in the countries where it works (largely in South Asia, Southeast Asia, and Latin America, the centers of the world shrimp industry) free electric stunning machines. The group has concluded that electric slaughter is likely more humane than being crushed or paralyzed in the ice slurry.

To get the stunning machines, though, farms have to commit to other welfare measures, like minimum sizes for shrimp ponds. Their shrimp must have enough room to move around, to burrow and rest; their water must be kept clean and free of noxious chemicals. They must procure shrimp from hatcheries that do not practice “eyestalk ablation,” a common procedure where unanesthetized mother shrimp have their eyes cut off, because farms have found that this increases egg-laying behaviors.

At the end of the day, Shrimp Welfare Project is a tiny nonprofit with about 10 full-time staff working to help hundreds of billions of animals die less painful deaths. They’re not trying to bring down the entire multibillion-dollar shrimp industry. They’re not even asking people to stop eating shrimp. Who could be mad about that?

Many, many people, it turns out, can be mad about that.

Since Shrimp Welfare Project emerged, it has become not just a focus of controversy among animal activists but a preferred cudgel for beating up on effective altruism in general. It’s not like such cudgels are, exactly, rare. EAs are (I say as one) very weird people. You may remember that one EA did a world-historic fraud a couple of years ago. Many of those remaining in the movement after that debacle have shifted to focusing on preventing AI-related catastrophe in a way that makes people seeking AI-related utopia furious, and frustrates people focused more on traditional tech issues like copyright or algorithmic bias.

But even with such rich material to work with, the EA movement’s support for Shrimp Welfare Project has generated particular ire. Representative reactions include:

When the pseudonymous blogger Flo of the substack Moral Law Within wrote a piece defending the importance of shrimp welfare, her reward was some 1.4 million views of her X post and a barrage of hate from enemies of the shrimp.

Flo isn’t an animal ethicist at all, she told me: “I’d be so bored if I did that. It’s like studying the ethics of punching random people in the face. Just don’t do it!” The point of her post was to explore the idea of scope sensitivity. This is, roughly put, the ethical concept that it’s sometimes a good idea to count stuff. Sometimes, as when reading John Taurek, the importance of counting seems obvious: It matters whether it’s one person or five people who are at risk of dying.

But especially when numbers reach into the millions and billions and trillions, our ability to intelligently compare starts to erode. Perhaps my favorite example is that of plastic straws. Over the past decade, huge amounts of global effort have gone into moving away from single-use plastic straws. My favorite coffee shop in Washington, DC, adopted a kind of cardboard straw that has all the tensile strength of a chocolate eclair. But even the most alarming estimates of plastic straw usage suggested that straws account for, at most, 0.03 percent of the plastic waste dumped in the world’s oceans every year. By contrast, fishing nets make up 46 percent of the waste. Focusing on plastic straws instead of fishing nets? That’s scope insensitivity.

The casual dismissal of shrimp welfare struck Flo as a similar kind of scope insensitivity. The number of shrimp killed every year is about four times greater than the number of humans who have ever lived in human history. If you think shrimp matter at all, even if you think shrimp matter only 1 percent or even 0.1 percent as much as people — these numbers should alarm you. The scope matters.

Ronny Chieng, the Daily Show correspondent, once did a segment on Shrimp Welfare Project, inspired by the furious Substack debate over it. He had penetrating questions for Jiménez Zorrilla (“Is this a sex thing?”) but the most trenchant he saved for an animal activist criticizing the quantitative approach of the group and of effective altruism in general: “Please don’t be offended by this. Are you just saying this because you’re bad at math?”

There are two layers to the negative reaction to shrimp welfare. One is, well, it’s shrimp. They’re tiny: it’s in the name. They look like gross ocean bugs. “Someone once asked me for cute shrimp pictures, and I sent them a few,” Jiménez Zorrilla once told an interviewer, “and they responded, ‘Well, clearly, you’ve been at this for way too long.’” There is an instinctive revulsion at the idea that an animal so minuscule and so evolutionarily far from humans could arouse our sympathies.

But the other layer is, I think, more fundamental: It’s not about shrimp but about counting. Shrimp Welfare Project serves as a kind of reductio ad absurdum of the view that John Taurek was wrong and that the numbers ought to count. Sure, at first that worldview just means that you save five humans rather than one. But once you get on that train, the last stop is the view that, to quote the title of Flo’s infamous post, “Yes, you should save 10^100 shrimp instead of one human.”

If one is told that they can choose between two logically consistent worldviews, and one of them means believing that saving five people isn’t better than saving one, and the other commits them to believing that the lives of shrimp are a matter of vast cosmic importance — I don’t know what most people would choose. I haven’t done the polling. But I’d guess they’d take the Taurek pill before they took the shrimp pill.

As I start to spiral out about this, about the choice between what seems a kind of moral nihilism where five lives don’t count for more than just one, and a kind of shrimp fanaticism that obliges me to consign myself and my wife and child to monkish poverty so we may serve the crustaceans who need us, I remember a man who takes neither of these positions: Andrés Jiménez Zorrilla.

“My cofounder and I tried to do this trolley exercise,” comparing the value of shrimps versus other animals or humans, “and dropped it five minutes in because it’s irrelevant,” he told me. “The point isn’t whether shrimp are more important than humans. The question is whether this thing is important enough for some people to spend some time on. The answer to us was incontrovertibly yes.”

The public’s reaction to press coverage like his Daily Show appearance, he recalled, was “overwhelmingly positive.” People didn’t fulminate about the evils of prioritizing shrimp lives over those of humans. They asked how they could know if the shrimp they’re buying is ethically raised and slaughtered. They asked for information about shrimp consciousness and pain awareness. They donated money.

Shrimp Welfare Project hasn’t exactly taken over the shrimp industry. But it’s making progress. “We have partners we’ve now given machines affecting on the order of 4 billion shrimps a year, or 1 percent of the total global volume,” he told me, visibly proud.

Jiménez Zorrilla expresses a worldview that I’ve come to appreciate. Call it “shrimp centrism.” The numbers matter. Taurek is wrong. But we are humans. We are often wrong. Our information is often imprecise. And certainly no one has enough information to conclude that shrimp welfare is the most important thing on earth.

What we probably do have enough information to conclude is that shrimp matter. At least a bit. And maybe it’s good that 4 billion of them a year get to die less painful deaths.

This story is part of a series supported by Animal Charity Evaluators, which received a grant from EarthShare.