Benatar's Fallacy

+ Rethinking the Consequences of Draper's Asymmetry

In ‘Benatar and Beyond: Rethinking the Consequences of the Asymmetry’, Kaila Draper offers a neat critique of Benatar’s most famous argument for anti-natalism. But from Draper’s starting point, I’ll suggest a stronger critique is possible.

Benatar’s Asymmetry

Benatar advocates the following four claims:

(1) the presence of pain is bad,

(2) the presence of pleasure is good,

(3) the absence of pain is good, even if that good is not enjoyed by anyone,

whereas

(4) the absence of pleasure is not bad, unless there is someone for whom this absence is a deprivation.

I’ve argued elsewhere that (4) is false and unmotivated — we should recognize the unconditional value of good lives as yielding impersonal (or “undirected”) reasons as well as person-directed ones. Only the latter depend on the person’s existence. Other reasons yet remain to regard the prevention of a good life as a loss to the world.

But that isn’t my focus today. Following Draper, let’s grant these four claims for sake of argument and see what follows. Benatar concludes that it’s “better never to have been”: all life is regrettable, for it contains some pain (the absence of which would be preferable) and some pleasure (the absence of which would be no worse). Combining some preferability with no outright dispreferability suggests that the absence of life is all things considered preferable.

The Fallacy: Conflicting Normative Standards

The basic problem is that Benatar’s asymmetry implies that reasons are modally variant — they depend upon which world we’re in — but his “better never to have been” verdict depends upon applying the standards of a world that is not our own.

From the standpoint of the empty world, premises (3) and (4) suggest that a life-containing world is worse.¹ But we are not in the empty world. From the standpoint of the actual world, and assuming that our lives contain more good than bad, premises (1) and (2) suggest that it’s a good thing that we’re alive. The empty world is not preferable from our normative standpoint: the absence of pain looks like an improvement, but this prospect is more than counterbalanced by the greater loss of pleasure (since there are people—us—for whom the imagined absence constitutes a deprivation).

At least, that’s my diagnosis. Draper’s is more complex. They first argue that Benatar’s asymmetry reflects the following Fundamental Asymmetry:

For any possible subject S of pain and pleasure, S's pain has unconditional intrinsic disvalue; but S's pleasure has only conditional intrinsic value: it has intrinsic value only on the condition that there is a time at which S exists.

Draper next notes the first part of the “basic problem” identified above: “the very standards for evaluating and ranking outcomes vary across the two outcomes to be evaluated and ranked.” Strangely though, Draper does not note that, when questioning the value of our own existence, it is already settled which of the worlds in question is actual. Instead, they focus on a purely prospective form of evaluation, from which it is an open epistemic question whether or not S will turn out to exist.

Conflicting normative standards

Three Options for Dealing with Conflicting Standards

Draper compares two candidate principles for evaluating outcomes with varying standards. According to the “shared-standards-only approach” (to which Benatar seems implicitly committed), the only applicable reasons are ones that apply unconditionally, regardless of which outcome eventuates. This approach (given our prior assumptions) yields the anti-natalist verdict that it’s better not to bring a new life into existence.

Draper instead advocates for the “all-standards-are-equal approach”, which—in cases of conflicting verdicts—implies that the two outcomes are evaluatively incommensurable. Draper’s main reason for preferring this approach is that it respects the following common-sense judgment:

CSJ: If someone’s life is a very good one in the sense that the value for them of the good things in their life far exceeds the disvalue for them of the bad things in their life, then it would not be better for them never to have been.

That’s fine as far as it goes, but common-sense surely goes further. Not only is it not better never to have been (for someone with a very good life), most of us surely think that we have good reason to be positively glad of good existence, all things considered. So good existence is, in this sense, outright better than non-existence. We may call this strengthened common-sense judgment CSJ*.

A third approach to divergent standards, and one that successfully secures CSJ*, we might call the “some-weight-to-each-standard” approach. You might give equal weight to each standard, or perhaps weight each one by your credence in its associated outcome. Whatever the method, once weights are assigned you can blend the standards by maximizing expected choiceworthiness across your available options.

For example, suppose we assign 50% each to the world in which S exists and the world in which S never exists. Then the badness of S’s pain is weighted at 100%, and their pleasure is weighted at 50%. So long as S’s life contains at least twice as much pleasure as pain, we end up with sufficient reason to rank S’s existence strictly above her non-existence, even without assigning unconditional value to her positive welfare.

Besides securing CSJ*, a second reason to prefer this some-weight-to-each-standard approach is that it deals better with trade-offs. As others have noted, it would seem misguided to treat two outcomes (A and B) as incommensurable when one standard regards A as vastly better than B, and the other regards B as just faintly better than A. Differences in normative stringency or magnitude need to be taken into account, and a weighted “maximize expected choiceworthiness” approach would seem the obvious way to achieve this.

Coda on Unconditional Value

Though it’s not the main focus of the paper, Draper suggests that we should endorse the Fundamental Asymmetry as essential for explaining other normative data. But it isn’t needed, as I argue at length in my 2017 ‘Rethinking the Asymmetry’.

For example, Draper writes that “At a minimum, we have a much stronger positive reason to cause an existing person to have good experiences than to cause someone to have good experiences by bringing them into existence.” I agree with this claim about our reasons. But any form of partiality towards the antecedently actual, such as the commonsense hybrid view on which we have both person-directed and undirected reasons, will suffice to secure this datum.

It’s a common fallacy in defenses of the asymmetry (found also, e.g., in Frick’s 2020 ‘Conditional Reasons and the Procreation Asymmetry’) to begin by noting the inadequacy of impersonal totalism, and inferring from this that good lives ground no unconditional/impersonal reasons. But this simply doesn’t follow. The inadequacy of impersonal totalism gives us reason to posit person-directed reasons: special reason to care about existing individuals. But that no more undermines our more generic reasons to favour better futures than do our special reasons to care about friends and family undermine the idea that we also have (more generic) moral reasons to help the global poor.

Moreover, as I’m constantly reminding people, it’s a myth that rejecting the value of creating good lives resolves the paradoxes of population ethics. The same quantity-quality tradeoffs simply re-emerge within a life, so we’re still on the hook for a more robustly axiological solution. And even on the population level, noncomparativists can’t secure the desired result that Parfit’s Utopia A is better than the repugnant world Z.

So I think there’s really nothing going for the strong asymmetry (on which good lives have merely conditional value). We do better to begin with recognition of the unconditional value of good life. There are deep puzzles in ethics, to be sure, but they aren’t helped by denying the full value of each other’s lives.

1. A good reason to question those premises!