Your Ethics Should Follow The Weirdness of the World

If the world is weird then so too should your ethical judgments be

I was recently talking with friend of the blog

Petrus

about whether factory farming is the worst thing in the world. Or, more precisely, whether it’s worse than all atrocities perpetrated against humans. My position is that factory farming is probably worse every few years than all human injustice has ever been in history. Petrus disagreed.

There are about 24 billion factory farmed land animals alive at any given time. They each undergo about an hour a day of suffering as intense as the most intense suffering most humans ever have in their life. I think numbers of factory farmed animals have gone up recently, so let’s be conservative and say that these factory farming numbers have been going on for 40 years. This means there have been about 3.5 x 10^14 hours of extreme animal suffering. If we assume that animals suffer about 10% as intensely as we do—again, quite conservative—and that animal suffering is only 10% as intrinsically morally serious as human suffering, then factory farming causes about as much badness as 3.5x10^12 hours of human suffering. In total, it’s about as bad as 400 million years of extreme suffering. That’s about as much as if the average human had to experience about 500 hours of suffering as bad as the most intense suffering the average human experiences.

Clearly that would be the worst thing in the world. It would be like forcing everyone in the world to give birth for an hour, every day for a year. And that’s a conservative underestimate of the badness of factory farming. Factory farming is, therefore, definitely worse than human atrocities.

But Petrus argued that this couldn’t possibly be right. If your view tells you that factory farmers are worse than those who carried out the holocaust, something has gone badly wrong. If your ethical intuitions about suffering conflict with other more firmly held ethical intuitions, then you should give up the more firmly held ones.

I think this represents a lot of what’s wrong with how people do ethics. In other words,

Petrus

is what’s wrong with America!

(Jk, jk, he’s great, and you should all read his very insightful blog—especially impressive because he just graduated high school!)

I think the core idea of what’s wrong with this line of reasoning was articulated really well by

Michael Huemer

in Dialogues on Ethical Vegetarianism.¹

Sometimes, we can identify the particular point in an argument where things become surprising. Take the example of the folded paper. First I say that the thickness of the paper after fifty folds is 0.001 inches times 2^50 . That statement isn’t weird or surprising or controversial. What’s surprising is just how enormous two to the fiftieth power turns out to be. That’s where the “craziness” of the final answer comes from.

That’s why it’s not reasonable to conclude that there must be something wrong with the argument.

Let’s review the major premises in my reasoning. Some of them are moral premises, and some are empirical, factual premises. First, I have the moral premise that suffering is bad. Anything surprising there?

The next step in my argument is just a factual, empirical premise: that life on factory farms is extremely unpleasant. Is that surprising?

Here’s my other factual premise: the number of animals killed in two years of factory farming is greater than the total number of humans who have ever existed. Were you expecting that?

That’s where the “craziness” comes from. My moral claims aren’t surprising; it’s the empirical facts that are surprising. It’s shocking that factory farming might be the world’s worst problem, not because it’s shocking that animal suffering might matter, but because the quantity of animal suffering is shockingly large.

I don’t think that would make sense either. That’s why I made the point about how the surprisingness of my conclusions is due to the empirical facts, not my moral premises. You shouldn’t reject an obvious moral principle based on your assessment of a particular case, if you didn’t know the facts when you made that assessment.

In short, if the facts of the world are very surprising, then our ethical conclusions should sound surprising. To rip off an example from

Silas Abrahamsen

, it would seem really surprising at first if people scratching their butts was the worst thing in the world. But then if you learn that people scratching their butts causes quadrillions of people to be tortured—so that it’s responsible for most of the world’s suffering—then you should simply give up on that initial intuition. You should realize that your initial intuition was a byproduct of faulty knowledge about the world. If you formed an intuition without knowing the facts, you shouldn’t dogmatically cling to it after learning facts that put it in jeopardy.

This is why I think that thinking insect suffering is the biggest problem in the world isn’t actually counterintuitive. Sure, it sounds crazy! But how something sounds to the ear is not a very good gauge of its ultimate counterintuitiveness. When you realize that:

1. It’s quite intuitive that pain and suffering are a source of a sizeable share of the badness humans experience.

2. Suffering experienced by other animals isn’t intrinsically less important than suffering experienced by humans. Certainly it’s not intrinsically many orders of magnitude less important than human suffering. Pain is bad because it hurts, not because the person who experiences it is smart.

3. There’s a sizeable chance that roughly 99.99999% of the suffering experienced between humans and insects is experience by insects.

Insect suffering being the worst thing doesn’t seem so counterintuitive. What’s surprising is the factual claim, not the ethical claim. Ethics should get weird results that sound odd to the ear when you realize that the creatures experiencing roughly 100% of the world’s misery are small weird-looking creatures that we don’t naturally empathize with.

We absolutely should not trust our intuitions about how major aggregate problems are. We should not trust intuitions of the form “surely the holocaust was worse than factory farming.” Humans display a well-known bias called scope-neglect—we’ll pay similar amounts to save 2,000 birds, 20,000 birds, or 200,000 birds. Our brains tend to register any very large number simply as “big,” without grasping the difference in scale. We have no intuitive grasp of the difference in size between 2 million and 2 billion, even though it’s much greater than the difference between 2,000 and 200,000.

In light of this, our intuitions about how bad problems are in the aggregate—e.g. the holocaust vs factory farming or poverty vs war—are not at all reliable. What we have is an emotional reaction to various horrors that’s largely a function of how soberly they’re treated in popular society. No one treats factory farming with the gravity with which they treat historical crimes. So for this reason, we have much greater psychological aversion to historical crimes than factory farming. But this aversion is not at all reliable—we should not trust it. It’s just a byproduct of societal stigma and emotional reactions. We are not the sorts of creatures who can have reliable direct intuitions about the badness of millions vs billions of egregiously terrible actions.

So for this reason, we shouldn’t treat the fact that some plausible ethical judgment implies something weird about our macro-scale evaluations of various problems as decisive.

Petrus is pro-life. So are lots of people. If you’re pro-life, you probably have to think that abortion is much worse than the Nazi holocaust. There are about 73 million abortions each year—and that’s not even mentioning IVF, for instance. Abortion, on the pro-life worldview is a lot worse than the holocaust.

Is this a good objection to the pro-life worldview? I don’t think so! It only has force because we have a stronger emotional reaction to the holocaust than abortion. This isn’t surprising—we tend to have stronger emotional reactions to things that we’ve been taught are the worst crime in history since we were small children than to socially-normalized behavior.

(Note: I’m pro-choice, at least in the early stages).

Instead of doing granular macro comparisons, you should try to do more limited comparisons in ways that remove bias. When I was arguing that insect suffering was the world’s worst thing, I noted that there are about 100 million insects for every person. I argued that if there were creatures with the minds of bugs but that looked like people, 100 million of them would obviously be more important than you. If 100 million of them starved to death, that would be a lot worse than random trivial problems you had.

This is a more reliable source of intuitions because it: 1) controls for bias (based on how they look) and 2) compares a small number to a big number rather than a big number to an incomprehensible number. Our brains are not good at multiplying or gauging the difference between very large numbers.

Similarly, if you were pro-life, I think you’d argue that rather than comparing 11 million deaths to 73 million annual deaths, instead compare 100 fetus deaths to one human death. When you do that, it doesn’t seem so outlandish—if you are pro-life—to think 100 fetuses dying is a greater tragedy than one non-fetus dying.

People sometimes object to utilitarianism on the grounds that it holds that the far future, factory farming, and wild-animal suffering are all way more important than present issues. I don’t think this is a good objection. Once you come to vividly realize that nearly all suffering in the world is experienced by farmed and wild animals—and that there are orders of magnitude more future people than present people—the conclusions seem less counterintuitive. The judgments follow the weirdness of the world. They’re weird, but only because the world is weird.

1. I’m going to not include the relevant ellipses because it breaks up the flow, but just know, there are relevant ellipses.