Against Clean Water

Fish sex is still a really bad thing.

Yesterday, a Substack user left a hilarious comment on my article about Tyler Cowen’s argument for eating fish. Cowen says that, for fish, being eaten basically isn’t any worse than living in the wild, so it’s fine to eat them given utilitarianism. I think that’s wrong, because killing fish means that a lot more of them are going to be spawned to replace the ones who died, and fish younglings usually live very short lives filled with intense suffering. Hence (and if you haven’t caught on, this is the motto of the blog) fish sex is a really bad thing.

In Search of Fish Contraceptives

The commenter had this to say:

The Hudson used to be thoroughly polluted. Swimming in it was seen as insane, and it was very difficult to catch anything for lack of fish.

Now I look out the window and cringe. The water has been cleared, pollution reduced and the fish have returned. These poor creatures, with their net-negative existence, have another environment to reproduce and flourish in. It must have been far more moral when elevated pollutant levels and CO2 kept the river sterile of most animal life.

That gave me a good laugh. I think the comment is supposed to be sarcastic, but I can’t tell. The poster follows some accounts that might be woke on the fish sex question, like Michael Huemer and Scott Alexander. But it also has the cadence of a joke.

No matter, I’ll bite the bullet. Take out “and flourish” and I agree with the comment even at face value. It is wrong to reintroduce certain forms of wildlife, including every species of fish of which I’m aware, to environments that were previously uninhabitable to them. This is true even if human activity is what made the environment uninhabitable in the first place. In fact, rewilding is basically analogous to factory farming, and the arguments against it are nearly as strong.

Take this neat little thought experiment as an explanation of my view.

Let’s imagine if, instead of merely cleaning up the Hudson so that fish could live there, we had built an underwater settlement to simulate what fish lives are like in the wild. This settlement would include a wide variety of structures, each one dedicated to simulating a different process or condition in the natural lives of fishes.

Some of our structures would allow fish to do the things they like to do. We would have a nest-building area, a place with aquatic plants for the fish to eat, and a comfortable place to sleep.

In other structures, we would starve the fish to death, injure and mutilate them, infect them with natural diseases and parasites, kill their younglings en masse shortly after they spawn, and have the fish fight each other to the death so that some of the fish can eat the other fish. The youngling structure would be the biggest by far, since individual fishes spawn as many as a few million fry each year, all of which except an average of one per parent end up dying in some pretty gruesome ways. Fish fighting would be the second largest, since predation is the most common cause of death among mature animals in the wild. (And if you don’t like the idea of fish fighting—perhaps because it’s illegal—we could instead build a structure where we suffocate the fish to death and burn them in acid, which is what happens to most fish after they’re eaten alive by predators.)

Would it be ethical to populate this settlement with fish?

Of course not! What I’ve just described is a fish torture facility. And we have every reason to believe that the conditions of life in the settlement would be far more bad than good. If the fish would experience the same things in the wild and thus have the same quality of life as they would in the simulation, then rewilding must also be wrong.

There are a few objections you could make to this comparison. I don’t think they’re particularly good, but they’re still plausible.

First, you might say that it’s different if we torture the fish in the simulated environment than if we simply allow the fish to suffer the exact same conditions in nature. That would be a fine argument if we were considering whether to intervene in a natural environment that already exists. But that’s not the case here! The question is whether to create the environment at all or to leave it uninhabitable. Even if we didn’t build the settlement, and we just allowed the fish to live out their equivalent lives in the wild, we’d still be responsible for the fishes’ welfare because we’re the ones who brought the fish into being in the first place.

This is why I say rewilding is analogous to factory farming. In both cases, we’re bringing vast numbers of animals into existence who wouldn’t have existed otherwise, and these animals live such terrible lives that they’re almost certainly not worth living. Just as factory farming is wrong, it’s wrong to cause most wild animals to exist.

Second, you might say that wild fish ostensibly enjoy a great degree of autonomy that’s not enjoyed by the fish in our simulation. Whereas our fish would live a very structured life, wild fishes can move about freely and make their own choices. So shouldn’t rewilding be better?

No. For one, most wild fish actually don’t enjoy greater freedom than fish in our simulation. Consider: What sort of autonomy could possibly be enjoyed by the overwhelming majority of fish who are spawned, exist for a few seconds, and then either starve or are eaten alive by other fish? That doesn’t sound like a very autonomous life!

Moreover, any plausible moral theory must at least balance autonomy with other important values, like welfare. We tend to think that people have a right to do things that might be risky or harmful, but we don’t think that autonomy is without limits. And we don’t tend to think it’s worth living like Lord of the Flies. So why should we think it’s a good thing that a small minority of the fish we’re bringing into existence might have some degree of autonomy when the fish in toto are going to live really awful lives?

Finally, you might say that we have an obligation to restore the Hudson River because it was wrong for human beings to make the river uninhabitable in the first place. I’m not sure if I think destroying the river ecosystem even was wrong, but if you do, that doesn’t tell you anything about what you should do after nature has been destroyed.

As I mentioned in the original post, Tyler John and Jeff Sebo have argued that we should refrain from destroying nature because destroying nature might lead us to act on a generalized belief that it’s acceptable to harm animals. But in our case, we’re not in a position to destroy nature, only restore it. In fact, John and Sebo’s argument might tell us that rewilding is even worse than I give it credit for, because it promotes the idea that it’s a good thing to bring animals into existence who are going to live terrible lives—and that could cause us to believe that factory farming is good as well.

So, do I look out my window and cringe at the now sparkling, pristine water of the Hudson River, a single tear rolling down my cheek as I think of the countless fish suffering a net-negative existence? Yes, actually, I do. We should expect reason to sometimes lead us to counterintuitive conclusions, and we shouldn’t be afraid to set our intuitions aside when it does

The good old days.