The Profoundest Error in Population Ethics

Confused metaphysical dismissals of future-inclusive beneficence

Population ethics is one of those tricky areas where even extremely smart people¹ routinely believe clear-cut falsehoods (and may even mistake them for self-evident truths). Perhaps the most egregious of these errors is the idea that there’s a logical barrier preventing reasons of beneficence from counting in favor of bringing good lives into existence. Let me explain.

Existence Can Harm

You can neither harm nor benefit someone who doesn’t exist. This much is true. But too many mistakenly infer from this that we cannot harm or benefit someone in bringing them into existence. You might, for example, accept the combination of:

(Existence Incomparativism): An individual’s well-being cannot be compared between an outcome in which they exist and an alternative in which they do not. So, in particular, they can be neither better off nor worse off as a result of existing.

and

(Comparative Analyses of Harm and Benefit): An individual is harmed by an event iff they are worse off as a result of the event than they otherwise would have been. Similarly, an individual is benefited by an event iff it makes them better off than they otherwise would have been.

These two theses together entail:

(No Existential Harms or Benefits): Coming into existence, when you otherwise would not have been, can neither harm nor benefit you.

This conclusion is false, because existential harms are clearly possible. Consider:

Misery: If you press a button, this will bring into existence a person, ‘Misery’, who will have a terrible life, experiencing nothing but constant suffering (and other basic welfare bads). It will have no other notable effects. Should you press the button?

Few questions in ethics are as obvious as this one. You should not press the button, and the reason why you should not press the button is that pressing it would be very bad for Misery. It would, in short, be harmful.² So the principle (No Existential Harms or Benefits) is false, and undeniably so.

So Can Existence Benefit?

Consider the reverse case:

Joy: If you press a button, this will prevent the existence of a person, ‘Joy’, who would have had a wonderful life, full of the deepest happiness (and other basic welfare goods). It will have no other notable effects. Should you press the button?

This also seems an easy question, even if it is not as undeniable as the verdict about Misery. Alas, while everyone agrees that one ought not to create Misery, there are some philosophers who insist that they see no reason to prefer Joy’s existence. (Yikes.)³

As I explain in the above footnote, I think there’s a lot wrong with that verdict (to put it mildly). But that isn’t my main focus for today. Rather, in this post I just want to defend the strictly weaker claim that there’s no decent argument for excluding Joy from the scope of beneficence. If you just flat-footedly insist on the exclusion (as posited by proponents of the normative ‘Asymmetry’ in population ethics): whatever. But if you think there’s an actual argument (of the sort described below) against the future-inclusive view, I think you’re mistaken. So, to be perfectly clear on the dialectic: I’m not going to argue that existence can benefit. Rather, I’m trying to counter an argument that existence can’t benefit.

Souls waiting for a chance at life: not something you actually have to believe in, in order for procreative beneficence to make sense.

The (Bad) Metaphysical Argument

Ok, ready? Here’s the bad argument (drawing on the principles defined above):

1. Existence Incomparativism

2. Comparative Analyses of Harm and Benefit

3. Therefore, No Existential Benefits

The problem with this argument is that the premises equally entail No Existential Harms or Benefits, which we already saw is false. Given that there are existential harms—to bring Misery into existence is bad for her—we must already reject purely Comparative Analyses of the relevant normative concepts (or at least their conjunction with Existence Incomparativism). So you can’t use these rejected principles to argue against Existential Benefits.

In short: since bringing Misery into existence is bad for her (and for this very reason, bad per se), there can be no purely logical barrier to regarding Joy’s existence as good for her (and for this very reason, good per se).

There’s obviously a lot more to population ethics, including plenty that people can reasonably disagree about. But the above lesson may be the first and most important thing for anyone thinking about this topic to understand. You can try to make a normative case for not caring about Joy’s well-being. But you cannot reasonably think that future-inclusive beneficence is incoherent, or ruled out by basic logic or metaphysics. And once you appreciate the coherence of the future-inclusive view, you might next come to appreciate its normative correctness by reading just a little more: perhaps Joe Carlsmith’s ‘Against neutrality about creating happy lives’, or even just the quick thought experiment in the present post’s final footnote.

1. See, e.g., the famously sharp philosopher of science Tim Maudlin in our public Facebook exchange here. (But I don’t mean to be picking on Tim here: my sense is that the mistake he’s making is very widespread, hence my motivation to write this post!)

2. One might, like Maudlin in the above-linked thread, insist on using the word ‘harm’ only for comparative harms. But words don’t matter: the crucial concept that McMahan invokes via the term ‘non-comparative harm’ could be communicated using a different word instead. Call it ‘doing something bad for the subject’, if you prefer. We just need to then update our other normative inferences, since being “harmless” (in the narrow comparative sense) no longer means “not doing something bad for the subject”, and it’s the latter concept that normatively matters. After all, you would have to be morally insane to deny that we have moral reason not to create Misery, whatever particular word you want to use to describe the nature of the reason relating to concern for Misery’s well-being.

It’s certainly easiest to use the word ‘harm’ in a broad sense that allows for this non-comparative badness to also count as a harm. It seems a foolish kind of pedantry that insists on using words in less philosophically convenient ways that obscure what is important. But whatever. We can always use different words if someone insists upon foolish pedantry. Just mentally replace every mention of ‘existential harms and benefits’ in this post with the more tedious ‘bringing into existence in a way that is bad or good for the subject’, etc. And understand that we have welfare-related reasons not to create miserable lives even if you refuse to use the word ‘harm’ to describe how a miserable existence would be undesirable for the subject’s sake. Because again, a sane and decent person cannot reject the idea that I’m expressing here, even if you quibble about the verbiage.

3. For what it’s worth, I think that this neutral attitude towards good life is deeply immoral. To see why, imagine watching the “movie reel” of Joy’s potential life unfold before your eyes, and feeling nothing. You see her as a giggling toddler playing on the swings, and your thought is: “This life adds nothing of worth to the world.” Isn’t this attitude deplorably nihilistic?

It doesn’t seem to matter whether Joy already exists (and you’re watching the events “live”, so to speak) or not (and you’re instead granted a vision of the possible future from the Ghost of Christmas yet to come). Any decent person, upon fully grasping the potential of an innocent other’s life, must surely come to care about it—and to want the best for them, unconditionally. Alas, we tend to see others but dimly (and future persons least of all), and so we rarely grasp the full strength of our reasons of beneficence. That much is understandable. But I find it extremely disturbing that anyone could imagine this carefully and yet come to the settled conclusion that we have no reason at all to want Joy to exist. (It seems creepy in a similar way to other sentience-exclusionary views of beneficence: e.g., someone who vividly imagines wild animal suffering and reacts by saying “Nope, no reason at all to care about that!”)