Coming to exist is always a harm. Or so argues David Benatar in his provocative book, Better Never to Have Been.
A central pillar of Benatar's defense of this offputting 'anti-natalist' thesis is what he calls the asymmetry argument (BNHB, p. 30):
Pleasure benefits us and pain harms us.
(1) The presence of pain is bad.
(2) The presence of pleasure is good.
So far, pleasure and pain are symmetrical in their goodness and badness. But they are not symmetrical with respect to their absence. More specifically:
(3) The absence of pain is good, even if that good is not enjoyed by anyone, but
(4) The absence of pleasure is not bad unless there is somebody (an actual somebody) who is deprived by its absence.
If sound, the upshot of Benatar's asymmetry argument is that by not coming to exist, a prospective person is benefitted by not suffering the pains that inevitably color human existence, but a prospective person (being merely prospective, not actual) is not harmed by being denied the pleasures that come from actual existence. Thus, all other things being equal, coming to exist is a net harm.
I'm not sure if I'm convinced by the asymmetry argument as it stands, but I would like to raise a different worry. Does the argument generalize to other harm/benefit or good/bad pairs? In other words, is the following reasoning sound?
G benefits us and B harms us.
(1) The presence of G is good.
(2) The presence of B is bad.
(3) The absence of B is good, even if that good is not enjoyed by anyone.
(4) The absence of G is not bad unless there is somebody (an actual somebody) who is deprived by G's absence.
Now even if the pleasure/pain version of the asymmetry argument is sound, it weakens Benatar's case for his anti-natalist position if it does not generalize to other good/bad pairs. If it doesn't generalize, then perhaps, once the other goods and bads are taken into account, coming to exist will be neutral or a net benefit to us. In other words, the asymmetry argument can be more readily dismissed if its soundness requires that a hedonistic theory of value or well-being (a philosophically controversial theory) be assumed.
(A sidenote: It's not clear what Benatar would say about my generalization worries. He does not explicitly endorse hedonism, and later in the book, he argues that whichever plausible view we take human welfare or "quality of life", hedonistic or otherwise, life goes worse than we thought. So I'm not trying to pin hedonism on Benatar. I am instead questioning whether the asymmetry argument is persuasive only on the condition that hedonism is true.)
And my own intuition, inchoate though it may be, is that the asymmetry argument does not generalize. It strikes me as far less obvious that, for instance, virtue and vice, wisdom and ignorance, or friendship and friendlessness are asymmetrical in the way Benatar proposes pleasure and pain are asymmetrical. At this point, I don't have much to support this beyond my inchoate intuition, but here's one small idea: I dimly recall that Scholastic philosophers were fond of pointing out that opposing properties (or predicates, if you prefer) can oppose each other in two different ways. Some oppose each other because they are qualitatively antagonistic. Pleasure is not simply pain's absence, nor is pain pleasure's absence. They have opposing 'feels' or qualities. (Indeed, what makes pain bad is not its pleasantlessness, nor is pleasure good because of its painlessness.) But some oppose each other as presence and absence. Darkness is light's absence, rather than being antagonistic to lightness. (Forgive the imprecision of this. I'm sure I'm also getting the vocabulary wrong too!) I wonder if the pleasure/pain version of Benatar's asymmetry argument works because of the antagonistic relation between them, whereas it is less persuasive for good/bad pairs that are related as presence and absence (wisdom/ignorance, say). If so, then the argument would not generalize.
But again, I'm groping about for some insight into whether some good/bad pairs are not 'Benatar-asymmetrical,' and if so, why. Any help appreciated!
Wait a second. Stop the train, I want to get off.
Is what’s presented really the core of his argument? (Still haven’t gotten around to reading the book.) Because I think it’s simply invalid.
This is because once we admit the possibility of asymmetries of that sort (where X can be a good but ~X is not necessarily a harm, or where X can be a harm w/o ~X necessarily being a good), the same sort of asymmetry can apply to coming to exist itself.
That is, his argument, at most, establishes that failing to come to exist is a net good. But that does not entail that coming to exist is a net harm. Even if a prospective person suffers a net good by not coming to exist, this doesn’t exclude the possibility that their coming to exist is also a net good.
For my point to work, both coming to exist and not coming to exist would have to be measured relative to some indeterminate state, but that doesn’t seem to be a problem…
Further thoughts by way of elaboration on that last sentence: Benatar’s argument requires pleasure and pain to be matched with a third state that is neither pleasure or pain. Without that third state, the asymmetry is not possible — if pain is simply not-pleasure, then (4) contradicts (1).
I take it that he’d deny, in response to the objection I raised in the previous comment, that there is a similar third state between coming to exist and not coming to exist. But I think there has to be such a state. To see this, suppose that we establish a placeholder term P for a person who may or may not come to exist at point in time T. Before T, we can’t really say that P has come to exist, but, equally well, we can’t say that P has not-come-to-exist. It’s true, P has not yet come to exist, but the things that follow, on Benatar’s argument, from not coming to exist do not apply to P — P hasn’t yet been relieved from the pains of life, for example, because it’s not yet determined whether or not P will come to exist. Put differently, because it is possible that P will come to exist, he will never benefit from the goodness that comes from not coming to exist. Therefore, he is not in a state of not-coming-to-existence. We ought to understand him in a state of existence-not-yet-determined, and then we might understand both coming into existence and not coming into existence as improvements on that indeterminate condition. The former because it’s a guarantee of freedom from pain, and the latter because of the pleasures that an existing person can expect to get.
(I’m not sure whether being in a deterministic universe undermines my claim that it’s possible for there to be a not-yet-determined-to-exist state, but let’s leave that, and any Schrodinger-style dodges, that to the metaphysicians.)
(correction: “he will never benefit” in the previous comment should be “he will not necessarily benefit.” And since not-coming-to-exist necessarily entails coming to benefit from the goodness of not coming to exist, if it isn’t necessary that P will so benefit, then P isn’t in a state of not-coming-to-exist.)
(3) seems very suspicious to me. How could absence of pain be bad even if that good is not enjoyed by anyone. That would make any possible world that has no inhabitants good one respect. So, take a world in which only one atom exists for a second and nothing else. According to (3) this world would have value.
Also, the idea that non-existing things can be benefited and harmed seems bizarre to me.
Jussi,
I’m not sympathetic to the main argument, but I’m not convinced by your objection. Think of John Broome’s “Goodness is reducible to betterness”. Broome argues that all the facts about what is good are already present in the facts about what is better than what. Take a world just like the actual, except that a sentient being (Agonistes) exists there but not here and lives 24 hours in agony, with no redeeming good features of his life. That world, I would say, is worse than ours. It would be worse if Agonistes had existed (like that), so better that he doesn’t. But it isn’t better for anyone.
This may be wrong, but it isn’t incomprehensible, and to my mind it is pretty plausible.
Jamie,
I’m not sure that I’m convinced by that argument as an argument to assign positive value to absence of Agonistes’s pain in our world. I think I could understand the claim that it is better that Agonistes doesn’t exist as a way of saying that our world contains more good-makers and less bad-makers than the other world. But, to explain that this is the case, I don’t have to refer to lack of Agostines’ pain in our world – it’s enough that his pain is bad-maker in the other world.
I’m wondering about Paul’s comments on existence. I may simply be coloured by the fact I’m chipping away at Being and Nothingness but when Sartre isn’t getting distracted by Heideggerian shaped butterflies, I seem to notice him expressing the point that non-existence/nothingness logically follows from existence; a kind of “of what one cannot speak…”
Jussi,
I don’t follow you.
The absence of pain is good, because it is better than the alternative, viz., pain. That is a sufficient condition for its being good. The point of the Broome discussion is that there is no further issue of whether the absence of pain has positive value. The moral of his story is, look for elements that make things better. Once you have the rankings settled, there is no further question (except possibly some semantic question) of which things are good.
hmh. I wonder. Take two worlds which include only the following things:
a: block of wood.
b: block of wood and a very ugly painting.
Let us assume that the world b is better than the world a. How do we explain this? We could say that the ugly painting makes the world b worse. And, if the world b is worse, then the world a is better by definition – that’s just what worse and better mean. This means that we have an explanation for world a’s betterness that does not require assigning positive value to absence of the ugly painting in a. The point is that if we want to explain betterness relations – it’s enough if we have bad-makers on side. We don’t need both bad-makers and good-makers.
I think that is a good result. Your explanation for the betterness – that the ugly painting is missing from a requires saying that the absence of the ugly painting is a good-making property in a. That requires saying that the world a has positive value, it is good. That’s something I wan’t to resist. As far as I can see, a world with a block of wood is just a value neutral world.
Suppose that if Agonistes exists in a world, then there is no time in that world in which Agonistes exists and Agonistes is not in total (exclusively in) agony.
Suppose that if Orgasmo exists in a world, then there is no time in that world in which Orgasmo exists and Orgasmo is not in total (exclusively in) orgasmic bliss.
Suppose that if Nappo exists in a world, then there is no time in that world in which Nappo exists and Nappo is not insentient.
Further suppose the following three worlds:
W1 is such that, necessarily, Agonistes exists and is the sole sentient constituent.
W2 is such that, necessarily, Orgasmo exists and is the sole sentient constituent.
W3 is such that, necessarily, Nappo exists and is the sole constituent.
It should seems obvious that (ceteris paribus) W2>W3>W1, but it looks like there are two competing approach strategies.
Double-Dipping Strategy:
W2 has no bad makers and two good makers (orgasm, absence of pain).
W3 has no bad makers but only one good maker (absence of pain).
W1 has no good makers and one bad maker (agony).
W2>W3>W1
This looks a bit goofy, and so perhaps we ought to require a more substantive sense of ‘absence’ to be in play, namely that absence entails possibility of presence (adjusting possibility according to preferred strength), e.g., the absence of a moustache is an uncool making property only for things able at least in principle to be moustachioed. So, given that pain cannot be present in either W2 or W3, we could claim:
The absence of pain is a good (in a world) only if it is at least be possible for pain to be present (in that world).
This supports the familiar and straightforward…
Single-Dipping Strategy:
W2 has no bad makers and one good makers (orgasm).
W3 has no bad makers and no good makers.
W1 has no good makers and one bad maker (agony).
W2>W3>W1
Of course, depending on one’s mood, the constraint placed on the absence of pain being a good could range anywhere from the presence of pain being metaphysically possible to pain being actually present at some time.
Jamie,
Also, John Broome uses the non-existence of p in world A to define the evaluatively *neutral* level of well-being of p’s life in world B which is otherwise identical with A. The level of p’s life in B is evaluatively neutral iff world A is as good as world B. I cannot see how this could be if the absence of person’s pain in a world would be good (as you seem to think) when there was no on to experience that absence. (This does not commit him to saying that someone living a life of not pain is neutral as such a life can make the world better than the inexistence world.)
“take a world in which only one atom exists for a second and nothing else. According to (3) this world would have value.”
I think this world has at least a quantum of value (pun intended). For the sake of this discussion, let’s take as a proxy for the value of a world its expected impact on conceivable valuers given epistemic access to it. So our hypothetical world, call it W(a), consists of a unit of spacetime that contains, or is constituted by, the career of a single atom. Sounds a bit mundane, perhaps – but only if you ignore the rich structure of atoms. This structure would be of infinitely more interest to at least some conceivable valuer than would be nothing. Hence, W(a) has more value than nothing.
Jussi,
Have you read “Goodness is reducible to betterness”? I can’t tell whether you mean to be objecting to the conclusion and disagreeing with the arguments there.
These are not two different things:
1. The presence of the painting makes its world worse.
2. The absence of the painting makes the painting-free world better.
They are the same thing. The presence of my laptop makes this room warmer; its absence would make the room cooler. These are not competing explanations or separate facts.
That’s the picture. I think it’s correct. It is, of course, entirely consistent with setting the neutral value (as in Weighing Lives to the value of non-existence.
I have and I am just reading it again. I think I’m objecting to thinking that Broome’s view (which probably is right) has the consequences you think it has.
We started from the claim:
(3) The absence of pain is good [has positive value, right?], even if that good is not enjoyed by anyone.
I denied this by claiming that absence of pain enjoyed by no one cannot have value because of an absurd conclusion this would have of making near empty worlds good (as far as I see Broome agrees with this – if it had value, how could the absense of lives be the standard for value neutral lives?).
You seem to be claiming that my claim that absence of pain lacks value if there is no one to enjoy it cannot be true because facts about goodness are reducible to facts about better than relations as Broome says. This is what I don’t understand. Broome himself seems to argue in that paper that ‘goodness may have absolute degrees of a sort, whilst still being reducible to its comparatives’. Could you explain why absense of pain must have positive value if claims about good are reducible to comparatives?
Jussi,
The slogan, “The evil of death is the value of life” might help here. Recall that one of Broome’s main points is that the Stoic view, that death cannot be a bad because it is (either a noncorporeal continuation of life or) nothing, fails because the deprivation of a good is itself a bad. (Do you agree with this?) There is no further question, once it is agreed that death deprives us of the goods of life, whether it is also a bad thing.
1. Empty worlds good?
There is no substantial issue of whether empty worlds are good. They are better than some worlds (with lots of pain, say) and worse than others. Once we have the facts of which worlds are better than which, we have all the facts. There is no further fact of where the cut-off between good and bad lies.
2. Neutral Level
That nonexistence fixes the ‘neutral level’ is consistent with the rest of what I’m saying here. It is a point about what lives make the world a better or worse place. It is not about where to draw the line between good and bad.
Suppose someone wanted to insist, “I agree that adding a life of exactly that sort would not make the world better or worse, but an utterly empty world doesn’t have a value of zero, it has a value of seven.” This would be unobjectionable except for possibly making some theoretic bookkeeping less convenient. (Do you agree with this?)
3. Valuable Absence
Compare two worlds w and x, just alike except for the presence of Agonistes in x (whose life is described above). w is better than x. There is nothing more to the claim that the absence of Agonistes’ life has positive value than this comparative claim. Since the comparative claim is true, it is true that the absence of Agonistes’ life has positive value. (This is like the point about the absence of my laptop making the room cooler. We do not have to believe that there is a substance, coolth, that the absence of my laptop has to some positive degree, to believe that the absence of my laptop would make the room cooler.) The absence of a painful life makes the world a better place; that is all we are saying.
Thanks Jamie. You have been very patient with me again. I still have issues especially with 1 given the section 10.3 in Broome’s paper. It seems to me that they neutrality of the empty world is one of the facts of the rankings of the worlds. Any goodness seems to be betterness to this world.
Let me just note that if [3] in the original argument of Michael’s is true on these grounds, then [4] in the very same argument must be false on the very same grounds. It states that absence of pleasure is not bad unless there is someone to experience them. The world without pleasure and experiencers is worse than the world with experienced pleasure. To say that the absence of pleasure is bad is nothing more than to make this comparative claim, if you are right.
I guess I can put this as a dilemma:
Either Broome is wrong and (3) is false, or Broome is right and (4) is false. On no view, are both (3) and (4) true.
I guess my original interest was that if you drop the not being enjoyed part of (3) and (4), then it is easy to find counter-examples to the asymmetry of the type:
1. Presence of real notes is good.
2. Presence of forged notes is bad.
3. Absence of forged notes is good.
4. Absence of real notes bad.