For animal welfare… eat meat!
Updated: Apr 9, 2021
A refutation of the utilitarian claim that we should all be vegetarian
According to animal rights activists, we should give equal consideration to animal and human happiness, and therefore we should be vegan to reduce animal suffering. However, when we look at the philosophical grounds behind this reasoning, we discover that the logical conclusion might not be so clear, and might even be the opposite. This article can be seen as a refutation of the thesis of Animal Liberation by Peter Singer, even acknowledging its premises of both utilitarianism and equal consideration for animals.
In this article, we are going to think about the question of whether eating meat is morally a good or bad thing, and for that, we are going to think within the moral philosophy called utilitarianism. Don’t worry if you don’t know what it is, it’s a very simple idea, as we can summarize it as: “In a given situation, the moral choice is the one which causes the greatest happiness”. The name “utilitarianism” is actually quite counterintuitive, as “utility” here means people’s happiness (it has nothing to do with social utility or something being “useful” in the common sense).
If you have read my previous articles about philosophy, you might remember that I am quite skeptical about utilitarianism. As humans in modern society evolve in an environment where laws are predominant, I think it is more useful to think about how to build just laws, and then moral behavior is a lot about following these just laws. However, animals can’t be expected to negotiate laws with humans so these “advanced” philosophies cannot be applied. On the issue of animals, the fact that utilitarianism is simplistic becomes an advantage, as the notions of happiness and suffering still makes sense when applied to animals, and therefore this philosophy can be applied directly. Note that for very advanced animals like great apes, it might also be insufficient, but here I am interested in the species we raise in farms for food.
The strong part of the argument
Usually, animal rights advocates spend a lot of time justifying that animal happiness is as worthy of consideration as human happiness. This argument was already put forward by the founder of utilitarianism, Jeremy Bentham, as early as 1790. But the most famous modern argument is given in the book Animal Liberation by Peter Singer, which is even cited as its philosophical reference by PETA, a famous American radical animal rights association promoting veganism and a complete abolition of animal use in human society.
The basic reasoning is: animals are capable of suffering and happiness, therefore their suffering and happiness matter just like human suffering and happiness do. Denying this is just plain racism against animals (the specific term is “speciesism”) which has no scientific basis since humans have evolved continuously from animals.
Don’t get their position wrong. It is sometimes caricatured as saying that the life of a human does not matter more than the life of a chicken for instance. Peter Singer is not saying something that extreme. He acknowledges that different animals are capable of suffering pain and enjoying pleasure to different degrees depending on the species. So the happiness of more “advanced” animals matters more, but this is a difference in degree, not in nature. So for instance the suffering of 1 cat might matter less than the suffering of 1 human, but the suffering of say 1,000 will matter more.
After demonstrating this proposition, animal rights activists think that the hardest part is done. After that, they point that the happiness of a human eating meat is lower than the suffering of the animal in question in the farm, and therefore, eating meat is wrong.
People who want to defend eating meat usually attack the first part of the reasoning, that is they try to argue that there is a valid reason for discounting animal suffering as always secondary to human happiness, so that the pleasure of eating meat outweighs the suffering of animals. I think this approach is deeply wrong, as it attacks the part of the argument which is actually quite strong. The weak part is the second one, as I will try to argue in the following sections.
The question we are trying to answer
Let’s get to the real-world issue. The most important aspect of animal happiness is clearly meat production, as the number of animals raised for meat massively outweighs animals raised for other purposes. Animals raised in farms represent around 70 billion animals per year, and the vast majority of those are pigs, cows and chickens. So clearly, the biggest moral issue about animals is what happens in farms.
Now let’s consider an ethical choice on buying meat. Let’s say you want to buy some chicken, and you have 3 choices:
“Regular” chicken raised with “factory farming” techniques
“Free range” chicken raised in conditions that provide good animal welfare
Not buying the chicken at all
Comparing #1 and #2 is easy, because the outcome of your choice has no impact on the total number of chickens raised. It only changes the life of one chicken from a miserable one to a good one, so basically you are making one chicken happier. Of course the chickens in the supermarket are already dead so nothing will change for them, but since production adapts to demand, each extra “free range” chicken bought instead of a “factory farm” one will result in one future chicken raised in good conditions instead of bad ones. You might object that production is not fine-tuned for every single chicken, but it does not change the validity of this argument because if people buy 1 million “free range” chickens per year instead of “factory farm” chickens, the industry will adapt to produce 1 million more “free range” chickens and 1 million less “factory farm” chickens, so in the end, on average, every “free range” chicken bought does correspond to one “free range” chicken bred.
So there is a clear case for buying meat from farms with concern for animal welfare. But what about the choice between #1 vs #3 or #2 vs #3? It’s more complicated, because the outcome is not a chicken being happy vs being less happy, it’s about a chicken being born vs not being born at all. So is it better or worse? As we are going to see in the next section, it raises a difficult question: whether it is average or total happiness that matters.
Total or average happiness?
Utilitarianism tells us to maximize the happiness of humans and animals. But is the total or average happiness? The only difference is that in the “average” version we divide by the number of people to get the average happiness of someone on Earth, while in the “total” variant we simply compute the total, so the more people the better.
If the decision you are making does not influence the number of humans and animals on the planet, this question is irrelevant. But when we are talking about having more children or buying meat, it does matter. Let’s consider total utilitarianism first, which is the original version of utilitarianism created by Jeremy Bentham in 1790 (also called “classical” utilitarianism for this reason).
The problem with the “total” variant it is that it leads to the “mere addition paradox” which is that we should make as many children as possible, even if it leads to extreme poverty and starvation, as long as the lives of these new people are not miserable to the point where they would prefer to commit suicide. This sounds counter-intuitive, we would probably rather be less numerous but more happy. It looks especially strange to advocate this version of utilitarianism in the modern era, as we are reaching limits on some world resources.
So okay, the solution seems obvious, what matters is the average happiness, isn’t it? It is what a lot of utilitarians have started to advocate since John Stuart Mill in 1860. One intuitive argument in favor of this view is that it maximizes the chances of a child who comes to this world to be happy (mathematically it maximizes the expected happiness of someone being born).
But the “average” variant has its issues too. Should you give birth to a child? With the “average happiness” logic, it depends whether your child would be happier than the average living being on Earth. If this is the case, it is a good action to have a child, otherwise it’s a bad action. So if you live in a poor country and your child would be less happy than the average human, then it is immoral to give birth to one!
Imagine that intelligent aliens exist, and one day they discover the Earth, and compute, using their advanced technology, that the average happiness of a human is lower than that of an average alien. Average utilitarianism would then conclude that the most moral action for them is to exterminate all humanity to increase the average happiness in the universe!
So it seems that average utilitarianism leads to strange consequences too, and we might think that in fact the “total” variant is more adapted to the question of adding people or animals to the world.
Note also that in both “average” and “total” variants, there is no way to distinguish between a morally good but not required act, and a morally required act. Utilitarianism can simply measure how good or bad something is, but does not say anything between optional and required actions. This is a general limitation with utilitarianism (but is solved by Rawlsian ethics on the other hand).
The conclusion I think is that both “average” and “total” models have their issues, so we should be careful when thinking about notions which involve a changing population size. This is why I will consider meat-eating in the next section with both models, so you can see the consequences depending on your favorite one.
Let’s go back to animals
So now, how does all this apply to the question of whether or not to eat meat?
With total utilitarianism
Suppose we consider total utilitarianism, then whether adding more animals is good or bad depends on whether their happiness along all their life is globally positive or negative (neither a human nor an animal can be happy 100% of the time, so what matters is the net total along their life).
What does this mean in practice? Are farm animals below or above that level? Well, think about it, if you could magically euthanize all animals in farms (without any pain), would you do it? I don’t think so. Most animals don’t have lives that are so bad that it would be better that they were dead. Of course, it depends on farming practices, and it might be the case that in some farms, and for some animals, their life is so miserable that they might better be dead. So, to summarize, even though animals could live happier lives, most farm animals' lives are probably worth living anyway compared to not being lived at all. If this is the case, then it is a better action to buy meat than to be vegetarian. On the other hand, it does not invalidate the fact that buying meat from welfare-friendly farms is of course much better, as it adds even more happiness to the world.
Depending on whether you put regular farming practises above or below the “animals would prefer to die” level, you get either:
“Optimistic” about farms: “Buying free range chicken” better than “Buying regular chicken” better than “Not buying chicken”
“Pessimistic” about farms: “Buying free range chicken” better than “Not buying chicken” better than “Buying regular chicken”.
This result is summarized in the diagram below.
In both cases, the striking conclusion is that, with total utilitarianism, it is always better to buy “free range” chicken than being vegetarian.
With average utilitarianism
Now let’s consider the utilitarianism variant which tells us to maximize average happiness. The intuitive idea behind it is to maximize the chances of being happy for a sentient being coming to the world.
When adding an animal to the world, are you doing a good or bad action? Well, it depends on whether the added animal is happier than the average animal level of happiness on Earth. And here the “average level” is not only for animals of the same species, nor even all farm animals. No, I am talking about all animals, including wild animals, and including humans! So basically we should compute the average happiness of all humans, wild and farm animals, and get the result. Let’s call X the “average happiness” of all sentient beings on Earth.
Let’s go back to our question whether or not to buy chicken? Well, if you believe the chickens will have an average happiness level higher than X during their life, then yes, otherwise no. Good luck estimating that! Even funnier, it changes the morality of giving birth to new humans too! Imagine that in fact, animals are on average happier than most humans. Then, it becomes morally wrong to give birth to new humans as they are less happy than animals!
Let’s try to work around this issue
If, like me, you are not a strong believer in utilitarianism, you might say “let’s just put humans out of the calculation”. Indeed, if a smarter philosophy is needed to deal with human society (like Rawls’ one for instance), we can restrict utilitarianism to thinking about animal issues only. In this case the question might be more workable as “Are farm chicken more happy than the average happiness of an animal?”. But still, it’s pretty impossible to get an intuition of where the “average happiness of all animals” lies, as it includes many different species.
So let’s try a bit harder to save average-based utilitarianism. One way would be to reason on a per-species basis. It makes much more sense and seems much more doable in practice. The answer to the question “Was this chicken in the supermarket happier than the average chicken?” seems now much more accessible to human knowledge. At least we can probably get an intuition about its answer as we can compare the happiness of animals of the same species (it’s more feasible to know in practice if one chicken is happier than another than to say if a chicken is happier than a cow for instance).
In that case, there are two possibilities:
If the animal does not exist in the wild at all, the relevant question is then: Is this animal happier than the average happiness of this animal in farms.
If the animal does exist in the wild, we should wonder what the average happiness is among all these animals, both domestic and wild.
In the case of chicken (and most farm animals in fact), almost all of them are domestic chickens, so basically the question becomes: Is this farm chicken happier than the average? For instance, if we assume that there are only two types of farming techniques, “factory farming” and “free range”, then the average is somewhere between the happiness of chickens in these two, closer to “factory farming” since it is the most widely used. Therefore, buying “factory” chickens pushes the average down to the “factory farming” level while buying “free range” chicken pushes it up. Not buying chicken at all is neutral in that case, as it does not change the average.
To summarize, in this theory we can evaluate our choice with chickens like this:
Buying “factory farm” chicken => Bad action, as it reduces chicken average happiness.
Buying “free range” chicken => Good action, as it increases the average happiness.
Not buying the chicken at all => Neutral action, the average is unchanged.
This conclusion is perhaps even more puzzling than the previous one, as it makes buying meat a good or a bad action depending on what the average happiness currently is. A first strange thing with this vision is that some chicken-buyers are necessarily doing a good deed and others a bad one, as not everyone can be above average by definition. Buying “free range” chicken would also become a bad action if it became the norm and some “super free” chicken were available.
In the “total utilitarianism” model, eating meat is better than being vegetarian if the animals raised for this meat have an happiness level above the level where we would think euthanasia is a service to them.
In the “average utilitarianism” model, eating meat is better than being vegetarian only if the meat you buy comes from farms with above average animal welfare.
I think the most important take-away message is that philosophy is messy when it comes to questions that involve a variation in the population size. As both total and average utilitarianism lead to questionable conclusions, even when applied to humans only, we should be careful about conclusions we reach with those. All of this shows that there isn’t such a strong case for veganism, unlike what pro-animal-rights philosophers pretend.
However, don’t get me wrong. I am not saying that living conditions in farms don’t matter. That’s quite the opposite. The case for improving animal welfare in farms is strong because it does not depend on the dubious choice of model for handling varying population size. Finally, remember that we have explored only the question of reducing meat consumption for moral reasons about animals, but a good case could be made for reducing meat consumption in order to reduce global warming impact on humans.
Appendix: Singer's response to this argument
To be honest, I want to mention that Peter Singer tried to respond to this kind of objections in the second edition of Animal Liberation.
“Noting that if we were all vegetarians there would be far fewer pigs, cattle, chickens, and sheep, a few meat-eaters have claimed that they are actually doing the animals they eat a favor, since but for their desire to eat meat, those animals would never have come into existence at all! [...]
It certainly does not justify eating meat from factory-produced animals, for they suffer lives of boredom and deprivation, unable to satisfy their basic needs to turn around, groom, stretch, exercise, or take part in the social interactions normal for their species. To bring them into existence for a life of that kind is no benefit to them, but rather a great harm. At the most, the argument from the benefit of bringing a being into existence could justify continuing to eat free-range animals (of a species incapable of having desires for the future), who have a pleasant existence in a social group suited to their behavioral needs, and are then killed quickly and without pain.” 
So it seems Peter Singer reasons here according to the “total utilitarianism” model and assumes that animals living in most farms are in what I haved called the “pessimistic case”. This last hypothesis is questionable I think. Singer has actually changed a few times of philosophy on the core philosophical questions behind this issue .
For more information about the current state of the philosophical debate on this issue, you can read review article  (link below).
 Singer, Peter. Animal Liberation (p. 276). Random House. Kindle Edition.
 Delon, Nicolas. (2016). The Replaceability Argument in the Ethics of Animal Husbandry. 10.1007/978-94-007-6167-4_512-1.