Okay, okay, calm down. It’s a bit of a clickbait title, I’ll admit. To be clear, this is what I’m NOT arguing:
Societies as a whole should eat meat
And here is the slightly more modest thing for which I am going to argue:
Given certain plausible assumptions, in a utilitarian ethical framework, a non-negligible portion of individuals ought to eat meat.
I’ll get right into it and give the argument in broad strokes first and then we’ll examine each of the premises and consider possible responses.
No individual, through their own dietary choices, can impact the scale of farming at all; that is to say, the exact same animals will be subjected to the exact same suffering whether or not a person P personally chooses to eat meat. In other words, the utility function Ua (for utility of farm animals) is completely independent of whether P eats meat or not.
Given a person P who enjoys eating meat, the utility function UP (utility of P) is greater if P chooses to eat meat.
The utility function U[~P & ~a] for all the other sentient beings who aren’t either P or farm animals is completely independent of whether P eats meat or not.
Ergo, Uw(M)>Uw(~M), where Uw is the total utility function of the world and the proposition M stands for “P eats meat”.
There it is. That’s the argument. Note that, while the conclusion of the argument is that P should eat meat, utilitarianism will simultaneously demand that P advance the cause of ending farming on a societal level; as we noted above, it’s obviously the case that Uw(~M’)>Uw(M’), where Uw is the total utility function of the world and M’ stands for the proposition “eating farmed meat is a widespread practice,” or something to that effect.
Let’s consider the argument more in-depth.
(1) is probably the most controversial premise. It’s easy to imagine that one’s individual choices can impact the net demand for farmed meat and therefore impact the net supply. After all, if enough people just stopped eating meat, the factory farms wouldn’t do the irrational thing of continuing to raise and slaughter as many animals as they do. Right?
I think we can consider (1) as being analogous to the case of voting, actually. In some cases, voting is rational. If you’re one of three voters in a given election, for example, not voting is probably irrational, if you have any preference as to the outcome at all. However, if the cost (in time, energy or whatever else) of voting is sufficiently high and the probability that your individual vote will sway the outcome of the election is sufficiently low, it may be (and often is) the case that voting is irrational in the same way that buying a lottery ticket is irrational; the expected value of voting will be non-positive.
Similarly, if you were one of three omnivores in the world, I’d concede that your individual action, your choice to abstain from eating meat, would save lives. In that case, it’s clear to me that utilitarianism demands veganism. Even in the real world, if you not eating meat could save a single farm animal’s life, then I think utilitarianism demands you not eat meat.
But it’s not clear to me that that’s the case. I don’t think the big factory farms’ financial analysts collect data so fine-grained that they can isolate the signal of one person choosing to go veg from the statistical noise and adjust the number of animals they slaughter accordingly1. In fact, it occurs to me that there may be enough slack on the demand side of the equation that, even if Alice chooses not to eat meat, someone else who would like to eat more meat than they currently do would fill in the gap she’s left. Demand for meat is essentially inexhaustible.
Of course, we can also consider some probability function Pr on the proposition Q, where Q stands for the proposition “If P doesn’t eat meat, then at least one fewer animal will be slaughtered,” i.e., we can talk about the value Pr(Q). Pr(Q) obviously shouldn’t be zero, and if it’s sufficiently high, it may be that Pr(Q) will outweigh whatever gustatory pleasure P gets in the utilitarian calculation. But it’s not clear to me that Pr(Q) is sufficiently high to do so. At the very least, even if I haven’t convinced utilitarians to outright change their dietary habits, the utilitarian calculus doesn’t seem so clear anymore as on a prima facie analysis. I think utilitarians must at least take this argument seriously. But this—probabilistic grounds—is, I think, where there’s the most wiggle room for a utilitarian to object.
You could object to premise (2) on empirical, nutritional grounds, by arguing that eating meat is sufficiently bad for a person’s health to outweigh any gustatory pleasure they get from it. Fine; I agree that this is plausible in the case of red meat. Let’s then limit ourselves to just our hypothetical person P eating fish. Among other health benefits, eating fish lowers risk of heart attack and stroke (whereas frequently eating red meat raises the risk of both). Fish also provide nutrients that a veg diet alone would require supplements for. It’s not at all clear to me that a pescetarian diet is less healthy than a veg diet.
I don’t see any serious way to object to (3). There are probably some people who would be genuinely upset to hear that you’re eating meat (or fish), and their utility functions would decrease if they knew about it. But there are probably just as many people who would be genuinely pleased to hear it, so I think the utility calculation is a wash.
That aside, we can easily solve this objection by just saying that P should eat meat in private. This also pre-empts objections about how eating meat might impact P’s ability to effectively advocate for the end of meat on a societal scale because he’d appear hypocritical. As long as you’re confident you won’t be discovered, just do it in private2!
Now for part two:
The libertarian case against eating meat
I side with Peter Singer in rejecting speciesism. I think that if the NAP applies to humans, it ought to apply to all sentient creatures; ergo, I think you can make just as good a libertarian case for society-wide veganism as you can a utilitarian case. Why should the NAP apply to animals? Well, we can make pretty much the same arguments utilitarians do: if our criteria for who the NAP applies to is to be reasonably non-arbitrary, i.e., not just “the NAP applies to humans and only humans” by fiat, then there will always be edge cases where some human is missing the conditions we list, and I think it’s fairly uncontroversial to claim that, insofar as we take the NAP seriously, we do want it to apply to all humans.
“But,” we might object, “buying and eating meat once it’s dead isn’t itself a NAP violation, so can’t we make roughly the same argument, that no individual libertarian has a duty to avoid eating meat?”
Well, it’s possible. But it’s plausible to me that trafficking in NAP violations is itself a NAP violation. Since I (and possibly you) reject speciesism, let’s think of a direct human analogue to our farming case: suppose a serial killer is going around murdering people, and after the fact, you buy one of the corpses off him. Is that a NAP violation on your part? In the case of a one-off purchase, it’s not clear to me that it is, but if you engage in a prolonged pattern of buying corpses from the serial killer, we could make the case that you’ve established a tacit understanding between killer and buyer that you will buy the corpses; it’s hard to see, in NAP terms, how this differs from you just hiring a hitman outright.
What if, like with meat, demand for the hitman’s services is effectively inexhaustible? I’ve touched upon the idea, in a previous post, of some sort of “Aggregate NAP Theory" of libertarianism, by which we choose between two scenarios by picking the one with fewer or less bad NAP violations. It’s not clear that such a theory should care at all about who in particular is violating the NAP; in other words, in cases of inexhaustible demand, it shouldn’t matter on this theory whether you are the one buying the hitman’s services or someone else is, as long as the aggregate number/severity of NAP violations remains invariant. And, yes, if we buy into the Aggregate NAP Theory, then I think it’s going to be hard for us to refute an analogous version of the utilitarian argument for eating meat from the first section3. If we buy into a more traditional NAP theory, we may yet be safe from such arguments. Or you can just accept speciesism, but I think that’s kind of silly.
If any such analysts are reading this post, I’d welcome your input.
Utilitarians are probably tired of hearing this, since this is how I answer every utilitarian argument about social signalling. Sam Harris argues that if a utilitarian doctor killed one healthy patient to harvest his organs and save five, it’d create an atmosphere of fear where nobody would go to the doctor. My brother in Christ, I can just change the thought experiment to be such that the healthy patient is a homeless person nobody knows, and you dispose of his body somewhere no one will find it! Just. Do it. In private.
Although one noteworthy difference is that, whereas the conclusion of the utilitarian argument is that one ought to eat meat insofar as doing so maximizes the utility function, the conclusion of the libertarian argument will be something slightly more modest, that is, the permissibility of eating meat.
I think (1) is unlikely and that’s the weak point. Applying the NAP to animals would have outlandish implications in my view. That seems hard to imagine.