Journal of Controversial Ideas

What’s the aim? Principally, and as stated in the Abstract to my paper ‘Anti-Natalism and the Asymmetry’,1 to argue against the Asymmetry, and then to make a case, in its place, for a version of Anti-Natalism. So, of course, I’m surprised to read in Jeff McMahan’s provocative and stimulating commentary2 that in fact I defend the Asymmetry. But the puzzles here are easily solved. McMahan sees what he calls the Neutral Claim – starting good lives is permitted – as external to the Asymmetry; I’ve seen it as pretty much at the core. So reject this – and we agree that’s an aim – and, for me, the edifice crumbles. And two reasons for this – first, my gloss on the Asymmetry, at the outset of Section I, was taken from McMahan’s 2009 paper. Indeed, there are two components, but the second has two parts. Second, if we don’t do this then the difference between the Asymmetry and Anti-Natalism is at the least blurred – the Anti-Natalist will accept both (1) and the austere version of (2). As we both believe there is no issue of real substance here, so then I can move on.

McMahan’s commentary, I suggest, might be seen as a subtle blend of exposition and criticism but then with an admixture of further material, taking things forward. I’ll separate these out.

Exposition. I propose, McMahan says, a radical asymmetry between pleasures and pains. Indeed. There are both intrinsic goods and bads, I claim (obvious candidates – pleasures and pains) but only intrinsic disvalues. As is noted, and unsurprisingly, there’s a fair emphasis on reasons hereabouts; so while I think that given its negative value there are reasons always to reduce pain, there aren’t correspondingly always reasons to increase pleasures. When are there such reasons? Simplifying a little, when desired. So should we make people happy and/or make happy people? Only the former.

As a counter to this asymmetry I elsewhere downplay a difference. A common view will have it that even though there aren’t reasons to bring people (with overall good lives) into existence, there are reasons to keep them in existence once their lives are under way. I deny this. Just as we should prevent both the starting and the continuing of lives which are throughout worse than nothing, so we should neither start nor continue the overall good lives, apparently far better than nothing. Moving from nothing to something, or no one to someone, doesn’t shift the goalposts, presenting us, all of a sudden, with a thing that matters. (This no change view has affinities with the much-touted view that existence isn’t a predicate.) But I need to say more.

First, contrast this change – coming into existence – with one that often succeeds it – becoming a person. Now this does make a difference. Only persons, I claim, are things which can desire future pleasures, or want more life. But only non-persons, as a matter of fact, are things we can bring into existence. So then when I refer to the lives we have no reason to continue I exclude the likes of you and me. Second, step this up a gear. As a further matter of fact no lives are perfect. They all contain some pain. This pain gives us (some) reason first to prevent the starting of, and then, if prevention is missed, later to end, a life. Where persons are concerned this pain is not only outweighed but, as I say, often compensated by future pleasures – there is often reason, and a greater reason, to tolerate the pain. Where babies and animals – non-persons – are concerned, though there is easily enough the outweighing, there is never the compensating. So not only is there no reason for, there is reason against continuing the lives of babies and animals.3 But, again, there’s not much difference between present and future lives. So there is reason also against starting these lives. Thus not only is such starting not required, neither is it permitted.

Criticism. McMahan notes that several of my claims will be viewed as implausible, highly counterintuitive, in conflict with common sense. Others have gone further – such claims are outrageous, their defender immoral or evil. I need to be careful here. Argue too successfully against these charges and I forfeit any justification for being in this journal. So I aim only at a partial defence.

Some qualifications. The claim that death is not bad for babies and animals is too blunt as it stands. I allow a sense in which death can be bad, not only for babies and animals but also for trees and plants. It can interfere with the natural run of things, end a healthy life, prevent flourishing. But I deny that death in such cases is bad in a way that matters, that gives us reasons to prevent it. And of course I allow also a sense in which there are very often reasons to interfere, not only to prevent deaths, but also to start lives. But these reasons are all of them other-regarding; we need plants for food, animals similarly, people for companions, co-workers, carers. What I deny is that there are still further reasons for bringing or keeping any of these things in existence. So then I claim that it is not good (in a way that matters) either for those people, or for the universe, that new people should come into existence.4

Now McMahan senses a whiff of complacency hereabouts, and thinks I risk jumping out of the frying pan into the fire. It is altogether too easy, he fears, for me to justify bringing into existence wretched lives, simply, say, because I like having animals or babies around. But I claim only that other-regarding reasons are often weighty. There are no grounds for thinking either that these reasons will always be selfish (this is a charge he levels not only against me here, but also, in the 2009 paper, against Anti-Natalism more generally) or that they can’t be effectively opposed. I might be a utilitarian. Start a life, I’ll say, when, but only when the benefits to existing people (say, pains avoided) outweigh the costs to the new person (say, pains that will need to be endured).

Some grey areas. Several of McMahan’s examples mention an 18-month-old child. But here the waters are muddied. I make, and make much of, a person/non-person distinction, but don’t claim that this is sharp. This more or less walking, talking (and pointing) thing may well be thought of as neither one thing nor the other. So I’ll avoid making any contentious claims about it. Similarly with those animals – whales, apes, perhaps the brighter dogs – that might be described as quasi-persons. Leave them out of the discussion. Similarly, again, with time. Set aside the smartest animals. And I claim of the rest that they have no desires for future life or future pleasure. But imagine a sparrowhawk anticipates its prey’s next move. Is it having thoughts about the future? I’ll play safe – (typical) animals and babies don’t have plans for the day after tomorrow.

Desires. Taking my cue from McMahan’s discussion, I need to say more, and in more detail, about some further claims regarding desires, and their relation to the badness of death.

Do I hold that there is reason to produce future goods only if these goods are desired? McMahan appears to think so. And he considers a baby who might or might not live beyond the age of 30, claiming that I can have reason now to ensure that the person this baby will become receives some good after the age of 30, even though it has no desires for these goods. But, of course, I agree. Given the baby will become an adult, we have reasons to promote its future as an adult. But we don’t, I say, have reason to ensure it becomes an adult. McMahan is far from convinced:

For, if there is a reason to enable it to have more intrinsic goods in the future rather than fewer, there should also be a reason to enable it to have some rather than none.

This isn’t perspicuous. The comparison more rather fewer seems to involve someone who exists either way. But some rather than none involves no such commitment. Someone might exist, desire goods, but have none. Or she might fail to exist, and lack both desires and goods in equal measure. There is reason to want goods for someone who does, will, or might exist, no reason to want goods for someone who won’t exist, and no reason, I maintain, to bring into existence someone who might exist (whose existence depends on what I do) and who will then have concerns for their supply of goods. Bluntly, it’s one thing to anticipate there will be a person having some future desires, another to create such a person. McMahan’s baby is unimportantly different here. It exists but isn’t yet a person, and has no desires for the future. There is no reason, I maintain, to sustain its life, help it become a person.

There’s then a distinction made between moral and prudential reasons for the generation of future goods. McMahan doubts I make a good case against the former, but has no doubts about the latter – my argument here is holed below the waterline.

He claims I (can) have reason now to secure for myself some future (intrinsic) good even if I have no desire now for this good, either now, or later. Take as candidate goods, health (good in itself, not merely a means to life) or life. Suppose I am temporarily depressed, and have right now no desire for health or life. Still, it can be bad for me to die, even though, if I do, I won’t regret it. If I don’t, I’ll recover, welcome this, and then regain my desire for more life. More clearly it can be bad for me to become seriously unhealthy. I most likely will regret it. (Whether someone who is seriously depressed has reason to desire life is moot. Does someone in a coma have reason to come round?)

A little more about desires below, but first a point which, although it comes at the end of McMahan’s commentary, is perhaps best dealt with here. Neither of us, he says, can lay claim to views which are throughout both coherent and compelling. But of the two, my position is allegedly the worse. Well, perhaps, but am I really at odds with common sense, or challenging our fundamental moral beliefs, in insisting that we might kill babies and animals with the promise of worthwhile lives? Babies are a special case, but I’m not sure how many people, and for how long, have thought it wrong to subject animals to an unanticipated and painless death. We shouldn’t assume recent and local commonplaces have everywhere been set in stone.

Further Issues. Let me go back to the opening of this, and the question of what the Asymmetry claims. I want to resist the Neutral Claim, or the claim, in my gloss, that starting is permitted. But – and this is important – only within the context of the Asymmetry. Earlier drafts betrayed more ambition – I wanted to defend a version of Anti-Natalism against all-comers. But, now more cautious, I claim only that if we agree with the Negative Claim – there is no reason (absenting other-regarding reasons) to start an overall good life – then we’ll lose all hope of defending the Neutral Claim – the seemingly common-sense view that starting such lives is at least permitted. But the conditional is important – if there are reasons to start good lives then uncontroversially this will (in some circumstances) be permitted. In the 2009 paper on which I mostly draw, McMahan is evidently a supporter of the Asymmetry. Not so in his commentary:

I believe, though I cannot argue for it here, that there is a moral reason to confer a noncomparative benefit by causing an individual to exist. That is, I reject the Negative Claim.

So then he rejects here the Asymmetry. This represents what is surely an important shift in position, and one that we might have expected to be more fully acknowledged. And I do want to insist on this shift. In the earlier paper McMahan is at pains to distinguish between the Asymmetry proper, and what he calls the Weak Asymmetry – there is some, but perhaps not much reason to start good lives – and at pains also to articulate his reasons for backing the former.

This rejection of the Asymmetry surfaces, even if somewhat less clearly, earlier in the commentary on my paper. McMahan claims (again) that I rub up against common sense. A woman can bring into existence either child A, with a compromised sense of smell, or child B who is in this respect quite normal. She chooses, for trivial reasons, child A. Most will think she ought to have chosen B. But won’t I think that if there is no reason to start a good life, there is no reason to start the better of two good lives? Indeed, but I am not at all embarrassed by this. Consider another and well-known example of the so-called Non-Identity Problem. According to Parfit we ought to stint ourselves now if we thereby ensure that those in the future will be population B, whose lives are very much better than ours, rather than population A, whose lives are only moderately better than ours.5 But I see no reason to agree. After all, depleting resources now might be aimed not at our benefit but rather at that of others – perhaps strangers to us – alive now. And going for A rather than B improves intergenerational equality. Obviously there’s more to say, though not here.

McMahan later links this claim about choosing the better life with one about starting a good life. Though the first is more widely supported the second is the more compelling. For surely the advantages of a good life over no life are greater than those of a good life over one a little less good. Now this is a point I made some years ago, but spinning it in quite the opposite direction. Whereas McMahan thinks that as there are reasons to choose the better life, so there are even greater reasons to start the good life, I claimed (and still do) that as there are no reasons to start the good life, so, more clearly, there are no reasons to choose the better life.6

Why, though, deny there are reasons to start good lives? McMahan says, correctly, that I never explain this. Let me be clear. I believe this is true. But I haven’t argued the point in the paper, and haven’t there done much more than allude to McMahan’s discussion in the 2009 paper of the Asymmetry’s advantages over the Weak Asymmetry, with its claim that there are some such reasons. The principal target is, again, the defender of the Asymmetry. But here’s just a little more. You think there are reasons to create happy people on some distant planet. And you’d skip lunch in order to achieve this. Well, that’s fine and dandy but surely if you think there are reasons, you’ll be prepared to impose, if necessary, some cost on unwilling people, in order to achieve this. Suppose you can create thousands of happy people. Then the cost may be considerable. I’m just not inclined to think this is acceptable. Contrast this case with one in which we impose costs in order to save lives. This – think of taxes – is acceptable.

McMahan and I appear to be poles apart, then, on this issue of new lives. Elsewhere, and on a topic he discusses at some length in the later pages of his commentary, the distance between us is considerably less.

We agree that the Deprivation Account of death needs amending. It implies that where good lives are concerned the earlier death is worse. I lose more future goods by dying at 16 weeks than at 16 years. But intuitively the death of the teenager is more regrettable, distressing, tragic. And we both think the badness of death has to take into account not only goods lost but, in some sense or other, one’s relation to these goods at the time of death.

What is this relation? I appeal to desires. At 16 we typically have desires, hopes and plans for our future. At 16 weeks, none of this. And so the early death is, in the way that matters, not bad at all. McMahan’s explanation, via his Time Relative Interest Account (TRIA) is subtler. At 16 weeks, in fact from birth, the infant has some, even if a markedly weak, interest in its future. So its death is to some degree bad.

I’ll make just two points about this. First, interest is ambiguous. Contrast being interested in something (listening to more Purcell) and something’s being in my interests (protecting my hearing). I take it McMahan has the second sense – not a close cousin of desires – in mind. But plants, as well as animals, have interests in this sense. Does he want to imply we have reasons, just for their sakes, to keep plants alive? If so, the point is in the shadows. For prominent here is a concern with psychology. An infant at 16 weeks is, he claims, psychologically connected with its later self. Suppose it weren’t. It would still be physically connected, still be one and the same creature. So it seems that these psychological connections in particular play a role in explaining why even the very early death is to some degree bad. Psychological connections figure also in my account, of course. Death is bad at 16 only if I have then some desires to live on and do such and such in the future; and so to be at least one and the same person. So McMahan must have different sorts of connections in mind. What sorts? And how does the argument go?

I confess to first coming across this TRIA probably decades back. Evidently, I’ve some way to go in understanding it. And so there’s more work to be done.