Summary: Do not go gentle: why the Asymmetry does not support anti-natalism

This is a summary of the GPI Working Paper "Do not go gentle: why the Asymmetry does not support anti-natalism” by Andreas Mogensen. The summary was written by Rhys Southan.

Many people believe that it makes the world worse to create miserable lives, but that it doesn’t make the world better to create happy lives. This is one way of expressing “the Asymmetry” in population ethics.

If we go on creating new people, many will be happy, but some will be unhappy. If we accept the Asymmetry, the continued existence of humanity therefore involves something which is no better than creating no one and something which is worse than creating no one. This seemingly implies anti-natalism—the conclusion that creating new people is worse than creating none.  

In “Do not go gentle: why the Asymmetry does not support anti-natalism,” Mogensen argues this impression is mistaken. In the strange realm of population ethics, we should not assume that something worse plus something not better is necessarily something worse. So, this interpersonal interpretation of the Asymmetry need not entail anti-natalism. Mogensen argues this as well for Benatar’s intrapersonal interpretation of the Asymmetry—that the bad things in our lives count against creating us but the good things in our lives do not count in favour of creating us.

Interpersonal Asymmetry

Advocates of the interpersonal Asymmetry say it is not better or worse to create good lives rather than to create no one. It is important to recognise that here, “not better or worse” does not mean “precisely equal in value.” If it did, there would be an obvious problem. If creating a slightly good life was precisely as good as creating no life, and if creating an excellent life was also precisely as good as creating no life, then by transitivity (if X is as good as Y and Y is as good as Z, then X should also be as good as Z), creating a slightly good life would be precisely as good as creating an excellent life. However, it certainly seems better to create excellent lives than to create mildly good lives.

To avoid this contradiction, advocates of the Asymmetry say that instead of the transitive “as good as” relation, a different, intransitive, relation holds between creating good lives and creating no lives. Broome (2005) tentatively proposes “incommensurateness” as this intransitive relation. For Broome, incommensurate just means not better, not worse and not equal to. If the creation of happy lives is incommensurate with not creating any lives, there is no contradiction in claiming that creating an excellent life is better than creating a slightly good life but neither is better than creating no life. 

Incommensurability may also allow the Asymmetry to avoid anti-natalism. We can see how by considering three possible populations (A, B and C). In A, there would be one individual with high positive wellbeing, named Poppy, and no one else. In B, Poppy would exist with the same high positive wellbeing as in A, but there would be two additional individuals, named Diego and Eva, who would have slightly positive wellbeing. In C, Poppy would have the same high positive wellbeing as in A and B, but Eva would have high positive wellbeing too. Diego would have slightly negative wellbeing in C, however. 

The following table illustrates these three populations in numerical form, with Ω referring to the two missing individuals in A:

          A = Poppy (5 wellbeing)                   Ω                                Ω

          B = Poppy (5 wellbeing)    Diego (1 wellbeing)     Eva (1 wellbeing)

          C = Poppy (5 wellbeing)    Diego (−1 wellbeing)   Eva (5 wellbeing)

A and B are precisely equal in value with respect to Poppy, and they are incommensurate in value with respect to Diego and Eva, since A lacks those individuals altogether. This suggests A and B are incommensurate in value overall. 

A is precisely equal to C with respect to Poppy. C is worse than A with respect to Diego, since a slightly negative life is worse than no life at all. And, C and A are incommensurate with respect to Eva, since A entirely lacks a corresponding individual.  

Moving from population A to population C involves one change for the worse (adding the slightly negative Diego) and one incommensurate change (adding the highly positive Eva). The most obvious conclusion, given our assumptions so far: C is worse than A.

Yet population C seems better than population B, which we have said is not worse than A. In one respect, B and C are equal, as they both have a very happy Poppy. Although B is better than C with respect to Diego, since Diego has slightly positive welfare in B and slightly negative welfare in C, it appears C more than makes up for this with its very positive Eva, which is a significant improvement over B’s slightly positive Eva.

If C is better than B, which is incommensurate with A, C cannot be worse than A. So, like B, C is incommensurate with A overall—even though it is equal to A in one respect, worse in another and incommensurate only in the third.

How can C be incommensurate with A even though C adds something worse and nothing better? Perhaps the incommensurability of creating good lives is “able to swallow up bad things and neutralize them” (Broome 2005: 409). Broome calls this hypothetical phenomenon greedy neutrality, and while he rejects it (which leads him to ultimately reject the Asymmetry), this argument gives those who accept the Asymmetry a reason to believe that something worse plus something incommensurate can lead to something incommensurate. Creating some mixture of good and bad lives need not be worse overall if the incommensurateness of creating good lives can cancel the worseness of creating bad lives by making the entire outcome incommensurate with creating no lives. 

So, accepting this interpretation of the Asymmetry does not mean it is necessarily better if we stop having children and go extinct. 

Intrapersonal Asymmetry

Benatar (2006) defends an intrapersonal interpretation of the Asymmetry when he claims the bad things in our lives are worse for us than our never existing but the good things in our lives are not better for us than our never existing. Does greedy neutrality also block the apparent anti-natalist implication of Benatar’s intrapersonal Asymmetry? 

Benatar implicitly denies greedy neutrality without acknowledging the concept. He claims that because existence involves something worse than never existing (the bad aspects of each individual existence) and something which is not better or worse than never existing (the good aspects of each individual existence), existence is worse than never existing for everyone. But this assumes something worse plus something incommensurate is necessarily something worse, which we now have reason to doubt. 

Benatar does not seem to mind the anti-natalist implications of his views, and so is not compelled to accept greedy neutrality to avoid them. However, his implicit rejection of greedy neutrality has an additional result that is arguably even more unintuitive than anti-natalism. Mogensen demonstrates this by imagining three possibilities for an individual (or a merely possible individual) named Davina:

          D =    𝜔       𝜔

          E =    1        1

          F =    5       −1

In possibility D, Davina never exists.

In possibility E, Davina has two distinct life phases, and in both of them, she has a slightly positive wellbeing. Because her existence in E is entirely positive, Benatar must treat E as incommensurate with Davina never existing. E is not better for Davina than her never existing, but it is also not worse. 

In possibility F, Davina has two distinct life phases. In the first life phase, she has high positive wellbeing. In the second life phase, she has barely negative wellbeing. The first period of her life is incommensurate with her never existing and the second period of her life is worse than her never existing. Benatar believes F is worse for Davina than if she never existed because he implicitly assumes a worse thing plus an incommensurate thing is a worse thing.

Yet F is almost certainly a better life for Davina than E is. Most of us would welcome a small pain in our lives if it were necessary for getting significantly more pleasure as well. The problem is that if F is worse for Davina than her never existing, and F is better than E, then E must also be worse than Davina’s never existing. But it is highly implausible that a purely positive existence is worse than non-existence. Furthermore, Benatar agrees, writing, “About such an existence I say that it is neither a harm nor a benefit and we should be indifferent between such an existence and never existing.”

As such, Mogensen concludes that Benatar is left with only one plausible option if he wishes to maintain his intrapersonal Asymmetry: greedy neutrality. If something worse plus something incommensurate is not necessarily worse overall, Davina’s more pleasurable life can be better than her mildly pleasurable life, and both can be incommensurate with her never existing. Incidentally, this means Benatar must abandon anti-natalism and concede that a life which contains some bad is not always worse than no life at all.

References

Benatar, David (2006) Better never to have been: the harm of coming into existence. Oxford: Oxford University Press.

Broome, John (2005) Should we value population? Journal of Political Philosophy 13, 399-413