Summary
In “The Worseness of Nonexistence”, Theron Pummer makes an interesting argument that suggests that a failure to create new people can be as bad as cutting an existing person’s life short. I here briefly sketch out a reply to Pummer that can be made, in some version, from a variety of different views.
Outline of Pummer’s argument
The primary aim of Pummer’s essay is to defend comparativism, the view that things can be better or worse for merely possible persons. I agree with Pummer that we should accept some version of comparativism — for example, it seems obvious to me that a state of affairs in which a single person is brought into existence only to be tortured for their entire life is worse than a state of affairs in which no individual is brought into existence (non-comparativists cannot say this, as they hold that the two states of affairs are not comparable).
Yet Pummer seeks to establish worseness in the other direction, as he argues that a state of affairs in which a (possible) person does not come into existence can be worse than a state of affairs in which this person does come into existence. (Note that various views accept comparativism without accepting the proposed worseness of nonexistence, e.g. Benatar, 1997, 2006; Fehige, 1998; St. Jules, 2019; Frick, 2020; Vinding, 2020, ch. 2).
Pummer’s argument is roughly that it seems plausible that death — more precisely, the prevention of future life — can be bad for a being who has just barely come to fully meet all the criteria for being a person. Yet it seems implausible that the badness of preventing future life should disappear completely in the case of a slightly less developed being, i.e. a being that just barely fails to fully meet a given criterion required for personhood. Hence, if we accept that it can be worse for an existent being to have their life cut short, we should also accept that it can be worse to fail to create future life for beings who do not yet exist (Pummer, 2019, sec. 4).
I believe there are several plausible lines of response to this argument that can resist the proposed worseness of nonexistence. One such line of response is to rely on conditional interests.
(Alternatively, one could reply along Epicurean lines by arguing that death is not worse for the person who dies. Arguments for this Epicurean position can be found in Rosenbaum, 1986; Hol, 2019. Note that on an Epicurean view, death can still be bad for instrumental reasons, e.g. due to the loss of the positive roles that a life has for others. And a proponent of the Epicurean view may argue that such instrumental factors are a significant confounder in our evaluations of the supposed worseness of nonexistence for a life in isolation.)
Conditional interests
Conditional interests have been defended as a plausible basis on which to rest our views on population ethics, and for defending the Asymmetry in population ethics (St. Jules, 2019; Frick, 2020). On these views, interests only matter conditional on existing, and hence there is no moral value in creating and satisfying additional interests that otherwise would not have existed (this is related to the antifrustrationist axiology defended in Fehige, 1998).
Gradual emergence of interests
Yet a defense of conditional interests per se is arguably not sufficient to address Pummer’s argument, or at least an adapted version of it. For it still seems strange to say that there should be some point at which a being goes from having absolutely no interest in continued existence to suddenly having a very strong such interest. What is further needed to avoid Pummer’s discontinuity is that interests emerge gradually.
On such a view, it may be that, say, a human fetus that is three months old has no interest in continued existence, a fetus of four months has a very slight such interest, at five months the interest is somewhat greater, and so on. This view seems in line with the fact that the brain structures that mediate sentience and sentient interests develop gradually (Tawia, 1992). It is also consistent with the widely held view that it is worse to abort a fetus later rather than sooner, e.g. after eight months compared to three, which is supported by various views in ethics and value theory, such as Jeff McMahan’s time-relative interest account of the badness of death (McMahan, 2002).
On the view sketched above, there is no point at which an individual’s interest in continued existence goes from being wholly unimportant to being as important as, say, a strong interest in continued existence held by a fully developed person; continued existence is only wholly unimportant as long as no rudimentary such interest has developed. Thus, there is no sudden discontinuity, nor any worseness of nonexistence before an interest in continued existence has emerged.
It is worth noting that some views based on conditional interests may imply the rejection of what Pummer calls “weak deprivationism”, namely that death can be bad even if there is no desire or preference for continued existence — a premise that I think can reasonably be questioned. Yet conditional-interest views can also be consistent with this premise, as one may hold that the creation of a person creates an interest in continued existence that is not reducible to mere desires or preferences, while still maintaining that no such interest exists for merely possible persons.
Other views with similar replies
One may reply to Pummer’s argument in a similar vein based on other views. For example, one may hold that death is an intrinsic harm (as, e.g., W. D. Ross did), and further hold that this harm can vary depending on how maturely developed the being in question is. Such a view, too, would avoid the worseness of nonexistence and the discontinuity implied by Pummer’s argument, as the intrinsic harm of death for a being that is (in a morally relevant sense) just barely existent could be very slight, and then increase gradually rather than discontinuously.
Acknowledgments
I’m grateful to Michael St. Jules for useful comments.