In my entry “Problems With the Argument From Marginal Cases and Using Sentience as a Basis for Rights,” I attempted to debunk the argument from marginal cases, the keystone argument that holds up obligatory veganism and the notion that sentience is the basis for rights.

I’m getting tired of summarizing the argument from marginal cases, so in case you’re unfamiliar with it, here is Jack Norris and Ginny Messina’s take on it from Vegan For Life:

A human rights ethic suggest that no human—not just intelligent humans, but also babies, infants, and those who are mentally challenged—should be abused and used by others for whatever purpose they like. This raises the question about whether rights should be extended to animals. The idea that if we grant rights to humans of lesser intelligence or ability, we should also grant rights to animals is sometimes referred to as the argument from marginal cases. If intelligence and capability are not criteria for the possession of rights, why would animals—who have the capacity to feel fear and pain—be excluded from moral consideration? Some philosophers may reject the argument from marginal cases, but we have never known any of them to provide a compelling reason for doing so. (234 - 235)

Jeez, okay, I’ll try to do better this time.

First, for nostalgia’s sake, let’s look at the points I made in that earlier entry:

1) The argument from marginal cases seems to rely on a situation that has nothing to do with animals: the existence of humans with poor cognitive functioning. If it became possible to detect and correct the genetic combination that gave rise to that condition, and the means for doing so became widespread enough, it would theoretically be possible to end the existence of severely cognitively impaired humans. And then, since babies don’t belong in the argument from marginal cases, there would be no argument from marginal cases and it would become okay to eat meat again. Even though technology isn’t quite there yet, the fact that it is theoretically possible makes the argument from marginal cases look absurd — why should something that is entirely to do with humans affect what we can do with other animals?

After an email exchange with animal rights professor Jean Kazez, I accepted that this point wasn’t that strong. Abolishing human marginal cases would not in itself render the argument from marginal cases moot, since you could still use imaginary cognitively impaired humans as a thought experiment. Humans in a world with no cognitively impaired humans who want to say that rights are based on cognitive ability would still have to say, “And if there were humans at the cognitive level of other animals, we would treat them the same as we treat other animals at that cognitive level.” It would be much easier to say that in a world where there aren’t cognitively handicapped humans to call your bluff, but future people – being as compassionate and enlightened as everyone suspects imaginary future will be – would likely feel uncomfortable saying this nevertheless.

2) Vegans insist that there is no need for humans to eat animal products in order to thrive. But let’s say that we did. Surely that’s not an impossible fantasy, especially when you consider people who have various conditions limiting the kinds of plants they can eat. The argument from marginal cases, if accepted by humans who had a nutritional requirement for animal products, would either force them to sacrifice their health and maybe their lives in order to avoid speciesism, or it would require them to eat every existing species of animal, including cognitively impaired humans, in order to be logically consistent.

This is an interesting glitch, but doesn’t actually take us anywhere. It doesn’t debunk the argument so much as show that it is purely academic and potentially masochistic.

3) The argument from marginal cases requires us to “treat like cases alike,” yet babies, the intellectually impaired and the senile currently have more rights than vegans are willing to give animals; this means we either have to give human marginal cases fewer rights than most people (including vegans) are comfortable giving them, or we have to give animals more rights than even vegans are comfortable giving them. For instance, most vegans are okay with spaying and neutering companion animals, even though we would not do this to severely cognitively impaired humans. Also, if parents were unable to take care of a cognitively impaired human, society would feel a responsibility to step in and take care of that person. Would vegans say that society has a responsibility to take care of all animals who cannot take care of themselves? If so, are we obligated to take care of all companion animals without a home? Do we need to take care of all injured wild animals? If not, how is this not speciesism?

I believe that at the very least, the companion animal aspect of this point holds up. The argument from marginal cases is very clear that we are not to see a significant moral difference between cognitively impaired humans and other animals. While discussing the argument from marginal cases in Introduction to Animal Rights, Gary L. Francione writes:

There is no characteristic that serves to distinguish humans from all other animals. Whatever attribute we may think makes all humans ‘special’ and thereby different from other animals is shared by some group of nonhumans. Whatever ‘defect’ we may think makes animals inferior to us is shared by some group of us. In the end, the only difference between them and us is species, and species alone is not a morally relevant criterion for excluding animals from the moral community any more than is race a justification for human slavery or sex a justification for making women the property of their husbands. The use of species to justify the property status of animals is speciesism just as the use of race or sex to justify the property status of humans is racism or sexism. If we want animal interests to have moral significance, then we have to treat like cases like, and we cannot treat animals in ways in which we would not be willing to treat any human. (xxix)

Whatever you think of Francione, this is not an unusual interpretation of the argument from marginal cases. So if “we cannot treat animals in ways in which we would not be willing to treat any human,” why are most vegans okay with the widely accepted practice of spaying and neutering companion animals who have an interest in having sex, even though we wouldn’t do the same thing to cognitively impaired humans?

The vegan responses I’ve heard to this are that other animals are different from us because they are more easily prone to over-population, and that it’s good for the animals’ health to have their sex organs removed. The first of those two objections introduces a morally relevant difference into the mix: a particular breeding capacity means you have fewer rights and can be sterilized against your interests. Sentience, then, wouldn’t be the only issue when considering interests.

The second objection re-frames spaying and neutering as in the interest of companion animals. If we accept that, then the question becomes whether there is a moral difference between euthanizing companion animals and euthanizing severely cognitively impaired humans. If not, the argument from marginal cases wins this point. Quick poll: vegans, do you think parents euthanizing their adult cognitively impaired offspring is morally equivalent to pet owners taking their animals to a shelter to be euthanized? 

My point about our needing to foster wild animals to avoid speciesism if society would take care of orphaned cognitively disabled humans is on shakier ground, as one vegan commenter pointed out recently, but the reason for this puts the argument from marginal cases in even greater jeopardy, which is what the rest of this entry is about.

4) The argument from marginal cases arrives at a false conclusion; as Jean Kazez argues, sentience is not actually what we have on our minds when we don’t kill and eat the intellectually impaired.

I still agree with this one, fortunately, because that’s also what this entry is about.

I do have to give the argument from marginal cases some credit: it is correct that cognitive ability is not the only factor that determines the distribution of rights. However, that doesn’t mean that sentience is the right explanation. Here are some reasons why I think sentience does not work as a coherent unifying explanation for the distribution of rights/consideration of interests:

* Vegans often use “sentience” as shorthand for “an ability to experience pain.” But if pain is what’s at issue, there shouldn’t be anything in itself wrong with killing an animal, as long as you do it painlessly. It doesn’t seem like a disservice to take a being out of the world simply because it was able to suffer torture and agony. If the morally relevant aspect of sentience is pain experience, we need to devise better ways to raise and kill animals painlessly, or breed animals who don’t experience pain. However, vegans generally do not concede that this would make meat eating okay.

* Sometimes vegans say that it’s not just pain that counts in sentience – pleasure counts too. When we kill animals, even if painlessly, that wrongs them by depriving them of future pleasure. But if potential for future pleasure is the basis for rights, that means that we need to give rights to zygotes at conception, a policy that tends to displease vegans. It also smacks into most of the other problems that pain-centered sentience slams against, such as all the other problems I’m about to list.

* Some people, such as animal rights philosopher Tom Regan, say that sentience is just one trait in a litany of others such as “perception, memory, desire, belief, self-consciousness, intention, a sense of the future” that determine who deserves basic rights. Vegan blogger Tim Gier says Gary Francione uses sentience to encompass all those traits, even if he doesn’t admit it.

Clearly this doesn’t vindicate the sentience explanation, because it rejects sentience as sufficient for interest consideration by itself. But this “subject-of-a-life” standard is so stringent that it excludes many animals from rights, and could even exclude severely mentally impaired humans from rights/interest consideration. Certainly it would exclude unconscious humans from rights. It also runs into a lot of the same problems as the sentience explanation.

* If sentience or “subject-of-a-life” status grants a basic right to life, why are we allowed to kill someone in self-defense? Just because they are threatening our lives does not mean that they have lost their capacity for pain or pleasure, or that they aren’t subjectively experiencing life.

* Similarly, most animal rights advocates say it is okay to kill animals when immediate human survival is at stake. Since animals do not lose their sentience just because you are about to starve, sentience being the basis for a right to life cannot explain why this seems okay.

* According to the book Heart and Blood, as quoted by Melissa McEwen:

A few years back, a government agency promoting the American agrarian ideal shipped baby chickens and piglets to Koyukon Indian villagers- people who have been hunters, trappers, and fishers all their lives. Some folks took to the notion, built pens, raised healthy pigs and successful flocks, and eventually found eggs under their hens. That’s when things started going awry. After watching the chickens grow, many couldn’t bring themselves to eat the eggs, and it was even worse to think of dining on the birds or pigs. ‘People felt like they’d be eating their own children,’ a Koyukon woman told me. ‘A lot of them said, from now on they would only eat wild game they got by hunting. It felt a lot better that way.’

Farm animals and wild animals are equally sentient. Why would lifelong hunters suddenly feel bad about killing animals raised for food?

* Why are vegans accepting of agriculture and civilization when human expansion steals land from wild animals and kills them? Killing inhabitants for land grabs would be seen as a blatant rights violations if done to indigenous humans; if sentience explains considerations of interests such as life, why don’t we have to avoid killing these sentient wild animals?

“Intent” – saying that it’s okay to kill animals if killing them is not the end goal – does not work because the reasons that make intent an important concept for humans (getting revenge on societal menaces and attempting to prevent future offenses) mean nothing to animals killed by agriculture and civilization. Plus, vegans don’t let vegetarians get away with the intent argument when lacto-ovo vegetarians say that they don’t intend for laying hens and milk cows to die for meat. Also, few vegans would accept good intent as an excuse for killing indigenous humans for agricultural land; how can this be explained aside from speciesism?

Intuitively, something does feel different about killing a wild animal with a wheat thresher than killing a goat by knocking him out and then slitting his throat, even if both can be done with the same amount of pain. But what makes it feel different if both are sentient and intent doesn’t have any practical application for animals?

* In a burning house scenario, most people would choose to save their baby over their dog, and their friend over a stranger. Francione writes in Introduction to Animal Rights:

[M]ost of us would save our own child even if the other being in the burning house were someone else’s child, or Mother Teresa, or some other human whom we valued. Indeed, if we are willing to be honest about it, most of us would choose to save our own child over a dozen other people’s children.

If we protect life because of sentience, shouldn’t these decisions be a complete toss-up? Is anything else besides selfishness being considered in such hypothetical decisions?

* It is socially acceptable among the pro-choice to say that you would abort a fetus if you knew that it was going to be severely cognitively impaired, but it is far less socially acceptable to say that you would abort a fetus if you knew it was female, going to be gay, or racially mixed. It does seem that many people are at least a little more prone to discriminate against the severely mentally impaired than other human groups. Clearly many people do appear to think less of them, or think that their lives are less worth living. Nevertheless, we do extend rights to them once they’re here. Why the mixed feelings if sentience is all we need to consider?

* In The Great Starvation Experiment: Ancel Keys and the Men Who Starved for Science, Todd Tucker describes the hierarchy of life value that was exposed when the Nazis attempted to starve Leningrad into submission. First to go were zoo animals, followed by household pets, then wallpaper paste that was made of potatoes, then boiled leather, then corpses of humans who had already died, then children, and then people finally resorted to murdering adults or eating their own body parts. What informed this hierarchy if everyone – aside from the leather and potato paste – was equally sentient?

* Vegans say that the argument from marginal cases proves that we are speciesist, because it shows that the only difference between all humans and all other animals is species category. But if we met intelligent aliens who could communicate with us as humans can communicate with each other, would we capture them and raise them for food and clothing? Unless we saw them as a threat, I don’t think it’s so sure that we would. Doesn’t that suggest that cognitive ability and communication is a more decisive factor for rights than species? On the other hand, if we don’t raise cognitively impaired humans for food and clothing, we also can’t say that cognitive level is the only consideration here. What could explain this apparent contradiction?

* Most meat eaters believe that animals are sentient. Why is there so much agreement about sentience, but so little agreement about rights? If sentience is the reason we give basic rights, why do we give rights to the severely mentally impaired but not to other animals? Vegans see this as a problem in our own logical consistency, but given the incongruities with the sentience explanation for rights I already mentioned, the real mistake appears to be with the argument itself, and hanging onto sentience as the only criterion for rights despite all the exceptions it fails to explain. Clearly there’s something wrong with the argument from marginal cases because the conclusion it arrives at – that sentience alone provides the justification for a basic right to life – doesn’t mesh with reality.

But if it’s not sentience, what is it?

Selfishness and prejudice are other possibilities—we respect the rights of others only so that they will respect ours, and refuse to give rights to groups we don’t like—and those are often significant, but like cognitive ability and sentience, they leave too much unexplained to be the main explanations. A purely selfish humanity with a fetish for smarts wouldn’t have much patience for humans with severe cognitive disabilities, nor would it ever have concern for other animals.

It might be that there is not a unifying explanation for how rights play out the way they do, and that it turns out to be emotive and arbitrary. But that’s no fun. So if I were going to come up with an alternative to sentience to explain the guiding concepts behind how we distribute rights, I would say it’s a mixture of responsibility (both altruistic and reciprocal) and attachment.

Sorry, I know, it would be catchier if I’d boiled it down to a single word, but that’s what I’ve got.

I started thinking about this because of an objection that vegan commenter Rob W. made to my third point against the argument from marginal cases. Rob took exception to my suggestion that if we were to follow the argument from marginal cases literally, human society would need to take care of all animals that cannot take care of themselves – domesticated animals as well as wild animals – because it would be speciesist to say that humans had an obligation to take care of orphaned mentally impaired humans but not other animals who were similarly unable to fend for themselves. To paraphrase Rob, he said I was overlooking the question of responsibility, and that we’re not responsible for aiding helpless wild animals in the same way that we are for aiding humans who are unable to take care of themselves.

In other words, it’s our job to take care of our children, no matter their cognitive abilities. It’s not our job to rescue wild animals.

If that’s true, and it certainly sounds reasonable, it explains a lot.

If sentience and cognitive ability are out and it’s a sense of responsibility that explains why parents and institutions take care of severely mentally impaired humans when we aren’t obligated to take care of wild animals, we now have a non-speciesist explanation for why we can give severely mentally impaired people rights that we don’t give to animals. It’s not that the cognitively impaired are H. sapiens, it’s that we brought them into this world—that they are of us—and so we feel like we need to protect them.

If this makes sense, it’s not wrong or inconsistent to feel more responsible for severely mentally impaired humans than we feel for domesticated animals. Which means that it can be a legitimate position to raise animals for food, even though we wouldn’t do the same to severely mentally impaired humans.

If you deny the responsibility explanation, then you’re back to human society being obligated to take care of all animals, wild and domestic, who cannot fend for themselves. As a practical matter, human society will be unable to do this to any serious extent, but that obligation will still exist and will need to be pursued as much as possible.

Even if you’re willing to accept human responsibility for wild animals, what about all those other scenarios where sentience struggles? Since responsibility does a fine job of explaining why we take care of the severely mentally impaired without giving rights to animals, let’s see how it fares elsewhere.

The sentience explanation and Tom Regan’s “inherent value” theory can’t make sense of the hierarchy of life value that becomes so stark in extreme cases like mass starvation. Decisions between who to eat in Leningrad or who to save in burning houses should be arbitrary if all that matters is sentience, but they’re not.

Responsibility/attachment explains why.

We feel most responsible for and attached to those closest to us. The more we care about someone, or feel responsible for them, the more value their safety and lives have for us. As beings become further removed from our inner circle, our responsibility for them and attachment to them weakens. Usually this puts husbands, wives, children, parents and tribe members at the top of the life value hierarchy, though of course this can change depending on how much you like all of them.

And a child raised by wolves will feel more loyalty and sense of duty to that pack of wolves – and probably all wolves everywhere – than they will to human society.

The very top responsibility for most people is themselves; because we are the ones with the most control over own lives, are the only ones who directly experience them and are most affected when something goes wrong in them, we usually feel most responsible for ourselves and most attached to our own continued existences. But that’s not inevitable either, as plenty of people are willing to sacrifice themselves to save a larger number of people, the president, their romantic partner, their best friend or their child.

This is not to say that sentience, species and cognitive ability are never factors. But those are mostly significant as traits that can play on our sense of responsibility and attachment.

In Introduction to Animal Rights, Francione relays a number of burning house scenarios that seem believable enough. According to Francione, we would most likely save our friend over a stranger, our own child over multiple other children, and we might even save a dog over a human if we knew that the human were a mass murderer. If cognitive ability and speciesism were the basis of concern, these are not the patterns that would arise, and if sentience were the basis of concern, there wouldn’t be a pattern at all, except that we would want to save the dog over the Chia Pet.

The sentience focus can’t explain why you would kill a bear to save your child, but wouldn’t kill a bear to save a squirrel. Responsibility and attachment can. When your child is in danger, you are driven to action, not because the child has a capacity for pain and pleasure, but because you brought her into the world and owe it to her to do all you can to defend her. You of course have selfish motives too. Your child is an emotional investment that you want to see pay off; you want all your love and effort to amount to something, so you want to see this child grow up and have a good life.

Your sense of responsibility might be of a more reciprocal nature with friends. They’ve done favors for you in the past, and you would expect them to rush straight for you in a burning house, so you better have their backs. On top of that, you like hanging out with them, and who knows how well you’d get along with that random dude who is already pretty badly burned anyway.

So why do you (hopefully) save a random human over your own dog? In general we tend to be more loyal to other humans than animals because you would want them to do the same for you, and because you know there are other people wanting their loved ones to survive and you’d be letting them down by rescuing the dog instead. We can feel responsible to people we haven’t met yet and may never meet. If you love your dog enough to forget all that and save the dog instead, keep in mind, you will be shunned for the rest of your life. (Although there might be some crazy dog people who would be really into that.)

Do you have a responsibility to save a drowning human, even if you didn’t push her into the water? Yes, you are responsible to that person and the people who love her, at least if you and your loved ones would expect her to do the same for you.

Even people we don’t like benefit from the reciprocity aspect of responsibility; if they don’t hit us, we won’t hit them (if we’re smart).

As for the murderer begging you to save him over the far more adorable golden retriever, “for the sake of humanity,” who cares that he is human, sentient and has higher cognitive abilities than the dog? You have no responsibility for the human life of someone who doesn’t respect the lives of other humans. (Which is also why no one seems to have a moral problem with violent self-defense against sentient, cognitively capable attackers. Stop taking responsibility for others and they will stop taking responsibility for you.)

Even the starving can be discerning, if given options. With the choice between someone’s dog and a wild animal who offers the same quality of sustenance, most starving westerners would pick the wild animal, even though both are sentient, not human and have around the same level of cognition. There are a number of potential factors. Maybe it feels wrong to exploit the trust of an animal who grew up around humans, maybe it’s cultural repulsion to eating dogs, or maybe concern for the dog owner. Also important is that wild animals are the furthest removed from most of our senses of responsibility and attachment because they have no part in our society, we have little to nothing to do with their being in this world, and we know them the least well.

Though vegans don’t like to admit it, even they cannot believe that wild animals have rights, because then civilization and agriculture become impossible. When wild animals threaten our vegetables, or they inhabit land that we want to develop, they are in the way of our responsibility to ourselves. And when human-introduced invasive species threaten the ecosystem, killing them just to get rid of them could feel like a duty.  

Vegans should be happy. This responsibility explanation gets rid of the anti-vegan objection that if vegans really cared about suffering reduction, they would be more worried about wild animal suffering than factory farms.

It makes sense that the zoo animals went before the companion animals in Leningrad. They’re exotic and neat looking and all, but they’re strangers. Companion animals live with us, they reduce our stress and in return we feel responsible to care for their lives and safety. Defenders of the sentience explanation for rights don’t have a very good answer for why even vegans are okay with spaying and neutering sentient pets — or why PETA euthanizes them — as removing their sex organs or killing them would seem to infringe on their interests. If responsibility is the standard, however, there is no contradiction; sentience be damned, it is our responsibility to not let these animals we have domesticated breed at unsustainable rates.

Farm animals are not from us and they don’t live in our homes and provide us with friendship, but humans breed them and raise them. For this reason, farm animals raised for food fall somewhere between wild animals and companion animals on the responsibility radar of most people. This explains the intuition that vegans and even many omnivores have that killing a wild animal for agriculture and civilization is different than raising and killing an animal for food.

Farm animals hover in a gray zone which causes a lot of people to disagree about whether it is okay to kill them or not, like those hunters who had no problem killing sentient animals when they were in the wild, but balked at killing their family-like domesticated animals. It’s not that it’s necessarily worse to be the farm animal who is raised for food and killed in a slaughterhouse than it is to be the wild animal who is hunted. What’s different is our own feelings about it; there are more responsibility cues with the domesticated animal, which makes many of us conflicted.

Being the ones who actually interact with the animals, farmers have the greatest potential to get attached to individual animals (though vegans are better at getting attached to farm animals in the abstract). I’ve never been a farmer, nor have I talked to many of them in depth about farming. My guess, though, is that — at least with the good farmers — they feel responsible for raising the animals well, but they don’t feel responsible for keeping them alive as long as possible because the animals’ early deaths serve what they see as a greater responsibility: providing food and a living for themselves and their families, and providing sustenance to other humans.

Okay, so maybe responsibility and attachment explain why the hierarchy of life value exists, but can’t we shove it under the surface, only to be revealed in desperate situations? We feel more responsibility and attachment to domesticated farm animals than we do to agricultural plants and the wild animals who die for them, so shouldn’t we eat only the plants if we can?

Well, yes, if the responsibility for and attachment to farm animals you feel is enough to outweigh the responsibility you feel to yourself to be happy, to be healthy, enjoy convenience or taste, or whatever it is that appeals to you about animal products. If these are trivial advantages to you, or you don’t experience any advantages to animal products at all, concern for farm animals might possibly win out. If you feel miserable as a vegan, or you just don’t care about other animals at all, responsibility to yourself will probably win out. Both approaches are in line with the reasons we give rights. Neither is “wrong.”

One of the revelations I had as I was quitting veganism was that for me, veganism was a sacrifice. I hadn’t ever thought of it that way and excitedly announced this insight to one of my roommates at the time. She was an ex-vegan too, and her response was something along the lines of “duh.” But it was news to me, and once I realized that being vegan was abdicating responsibility to my own well-being and happiness, I felt much better about quitting it.

There is no contradiction between liking animals and eating them if rights come from responsibility and you feel a greater responsibility for yourself and have found that you cannot thrive or be happy without animal products. The same values that make us concerned about animals can make us more concerned about ourselves.

Many vegans want to say this murky issue is crystal clear, that animals have rights whether we like it or not, even if using them makes our lives significantly better. But the only clear-cut way to give rights is through reciprocal agreements, which animals cannot consciously enter with us. For this reason, animal rights will always be controversial. We can still feel responsible for animals, and we can feel attached to them, but there’s no objective destination for those feelings.

In response to a comment I wrote saying that responsibility rather than sentience could be the explanation for why we protect the severely mentally impaired, Rob W. wrote:

Do you think there’s only one basis for moral consideration? What would it be? I like responsibility because it’s relational, it’s social. It’s not about either the capacities of the human or the capacities of the animal, but defining the proper relationship between them.

The reason he and I can agree on the basis of rights and not agree on what the conclusion of that should be is that responsibility and attachment are more subjective than sentience. Emotion is just as important as rationality. That doesn’t make all of this random, but it does make it impossible to design a single code of morality that everyone can support.

It’s likely that we will never agree about how we should co-exist with other animals.