We know how to insult one another. Popular insults reduce the insulted person to specific parts of his or her body. We reach the endpoint of reductions when we say that the other means or counts for nothing. And when I moved to the UK, I learnt that we can go further still. You are not nothing, you are less than nothing. Space, though in plentiful supply, would be, according to this insult, preferably bare and empty: You are a waste a space.

Anti-natalism, in the form considered here, raises this insult to a doctrine: Independently of your effects on others, i.e. purely on your own terms, it would have been better for anyone like you (or me) to have never existed at all. In its ideal state, our world would contain no humans, no animals, and, arguably, nothing else. Anyone and anything are, in this sense, a disaster. Everything would better not be.

Anti-natalism is different from nihilism, the view that nothing really matters. Things do matter. It is on account of the misfortunes that can befall humans (and anything else) that it would be better for them never to have existed. Non-existence saves us from pain, illness and disappointment. Annihilation awaits everyone and everything. Annihilation is an unavoidable and an almost universal misfortune. Not so never to have been: what never is can never be the subject of any misfortune.

The idea that everyone is, on his or her own terms, a waste of a space, will play different roles in the two parts of this paper. In the first part, I consider whether it is true and I will argue that it is not. In the second part, I consider whether such views, views which are or come close to being insults, should be advanced within philosophy. I advocate a position according to which awareness of the insulting nature of a view is an important part of its discussion. Some debates within the anti-natalist tradition deserve, I will argue, moral disapproval. The appropriate form of disapproval will fall far short of moral outrage. When others engage in these debates, it is, in my view, appropriate to raise one’s eyebrows and shake one’s head.

Before I embark on these two tasks I need to take a step back. Anti-natalism is the view that procreation is morally wrong (in most cases). In its most prominent form, it is motivated by the concern to prevent suffering. In terms of ethical theory, the view that procreation is wrong arises within a person-affecting framework. This means that it deals with what is good for someone and eschews any fundamental role for a notion of simple goodness. We are facing a question of personal value – is it better for a person to exist rather than not? – and, in answering this question, we won’t refer to facts about what is good impersonally. According to the view under discussion, the main reason why people should not procreate is that doing so would be bad for the person coming into existence. (This reason could, on occasion, be outweighed by the benefits the person’s existence brings to other people.) ‘The central idea of this book’, says Benatar (2006, p.1), the most renowned anti-natalist philosopher, ‘is that coming into existence is always a serious harm’.

Benatar’s views have been widely discussed. In addition to conferences and journal articles, his views have entered public discourse.Footnote 1 He has engaged with many of the challenges that have been raised.Footnote 2 His responses have removed misunderstandings; they have done much to clarify his argument and to make his reasoning transparent (at least to me). Nevertheless, a fundamental objection remains. In developing this objection, I will go along, as far as I can, with the assumptions and premises of Benatar’s argument. My objection will be that, once we fully understand why he accepts the premises of his argument, we will also understand why these premises fail to support his conclusion.

When it comes to my second objection, I need to mention that the target shifts. In the second part of this paper, I don’t object to the anti-natalist doctrine itself or to the underlying evaluative claim. I rather object to the way anti-natalism is being presented, namely as a fascinating truth that invites ideas about its best implementation. Within the anti-natalist framework, the ambition to minimize suffering leads to the following question: How can we ensure, with as little suffering as possible, that the human race will die out as quickly and as painlessly as possible? Or, more generally, how can we ensure that there is no more life on earth and in the universe?

Such discussions make me feel uncomfortable. If my parents (or my children) would find out about my presence at a discussion of how to end all life, I’d feel genuinely embarrassed. In thinking about how to end all life, one oversteps a boundary and, I suppose, it is a boundary of moral decency. In the second part, I will try to develop a moral objection to this aspect of the anti-natalist project. Such an objection needs to respect the idea that not just any conflict between one’s own moral views and what philosophers are arguing for could be the basis of a moral objection. Nevertheless, doing moral philosophy is an activity that is itself part of its subject matter. I admit, it is not a very important part. However, those who have subscribed to reflection as part of their professional life should be willing to apply such reflection to their own work.

1 The First (Philosophical) Objection

Benatar wants to argue that, for any existing person P, it would have been better for P, had P not existed. Any claim about such comparability in terms of what is or would be better for P, be it Benatar’s claim that existence is worse or the more common claim that, for most people, existence is better for them than their non-existence, faces a challenge. We do know (or, at least, we think we do) what our existence is like for us, but we can’t really think about what our non-existence would be like for us. So, how can we compare our existence and non-existence in terms of what it would be like for us when this very idea fails to apply in one of the cases we want to compare? On similar grounds, John Broome argues against the very idea of comparing P’s existence and her non-existence in terms of what would be better for P. Here is his argument.

If it were better for a person that she lives than that she should never have lived at all, then if she had never lived at all, that would have been worse for her than if she had lived. But if she had never lived at all, there would have been no her for it to be worse for, so it could not have been worse for her’ (Broome 1999, 168).

Let us call being better for one, being worse for one and being of the same value for one positive personal value relations. Broome’s argument takes us from the assumption that a positive personal value relation holds between one’s existence and one’s non-existence, via a principle pertaining to the logic of comparisons, to a claim he regards as obviously false, namely that things can be better or worse for people even if they fail to exist.

I agree with Broome that nothing can be good or bad for one who fails to exist. What is good for a person stands in the relation of being a benefit to this person and without a person no such relationship can hold. Similarly, nothing can stand in the harming-relation to someone who fails to exist. Nor can anything be neutral for a person without this person existing. If there is nothing that is good or bad or neutral for a person who fails to exist, then nothing can be better or worse or be of the same value for the non-existent. Broome’s principle, I take it, is secure: any statement of positive personal value relations will be false if the person in question fails to exist. If there is no Santa Claus, nothing, not even the cancellation of Christmas, will ever harm him.

What I called a principle pertaining to the logic of comparisons seems equally secure. If A is better than B for P, B will be worse for P than A. Whatever the standards are that tell us what is better for P, they will, thereby, also tell us what is worse for P. Note, however, that it can happen that the compared alternatives, A and B, will themselves incorporate different standards of comparison. Kierkegaard famously said in Either/Or, ‘Marry, and you will regret it; don’t marry, you will also regret it; marry or don’t marry, you will regret it either way’ (Kierkegaard 1843). An unmarried person might value companionship and shared experience and, on this basis, regret not being married; a married person might miss their freedom and the time when less responsibility was placed on them. Being married, if Kierkegaard is right, looks differently from the perspectives of the married and the unmarried person. To the unmarried being married looks better and, consequently, being unmarried looks worse, but for the married, being unmarried looks better and being married worse. So, there are two questions an unmarried person can ask regarding these two alternatives. How does being married look from here – it looks better – and how does being married look from there – it looks worse.Footnote 3

In order to appreciate the lesson I’d like to draw from Kierkegaard’s example, we need to reduce the space between how things look from a perspective and how things really are. If one’s preferences determine what is better for one, what looks better and is, thus, preferred, is thereby better. The same is true on hedonistic premises. If how one feels about situations determines personal betterness, then the presence of regret indicates that the alternative is better. Depending on the example used, we may also assume that being in one of the situations changes one in ways such that being in the other would actually be better given one’s current dispositions. The change that is for the better given one’s current dispositions may change one’s dispositions in ways so that a reversed change would be better given the dispositions one will then have: true tragedy from which there is no escape. Of course, for those of a more optimistic bent, we can also find alternatives which are such that whichever one chooses it will be for the best. Prospective undergraduates who agonise over which university offer to accept will often be happy wherever they go. They will identify with the values and practices of their chosen institution; they will find friends and feel supported and at home wherever they choose to go. Generalising, we can say that it happens that people who choose A will become A-people for whom it is best to be at A and, had they chosen B, they would have become B-people for whom it would have been best to be at B.

When the alternatives themselves affect how standards apply to them, questions of comparisons become ambiguous. How alternatives compare given one standard may well differ from how they compare given a different standard. We can directly apply this insight to existence/non-existence comparisons. Leaving Benatar’s anti-natalism aside for a moment, I prefer existing (in one of many ways) to never having existed at all. From here, i.e. from the perspective of my current interests and concerns, existing, I’d say, is better for me than to have never existed at all. The reverse holds as well, i.e. from my current perspective, non-existence would have been worse for me. However, I cannot make a positive comparison between the same alternatives from there, i.e. under the assumption of non-existence. Had I never existed, nothing would be good or bad or better or worse for me. This includes my existing in my present form when judged from there.

According to this picture, it would not be better for Santa Claus that he came into existence as there is no Santa Claus. When we say that it would have been nice for Santa Claus that he existed what we mean is not that his non-existence would have been improved by coming into existence, we rather mean that, had he existed, his existence would have been better for him than his never existing. Both these evaluations, that coming into existence would not have been better for him (as he does not exist) and that had he come into existence, his existence would have been better for him are true. Their difference is the following: From there, i.e. under the assumption of his non-existence, his coming into existence would not benefit him, as there is no him. However, under the counterfactual assumption that what is now there would have been here, we can make a counterfactual from-here comparison, according to which he would have enjoyed his existence.

Broome’s argument starts from the assumption that one’s existence is better for one than one’s non-existence. This does entail that one’s non-existence would have been worse for one. Both these evaluations are evaluations from here and they don’t commit one to any particular evaluation from there. Thus, they are compatible with the idea, which I accept, that had one not existed, nothing (including one’s coming into existence) would have been good or bad or better or worse for one. Thus, if a person exists (or is hypothetically or counterfactually assumed to exist), a positive comparative value relation can hold between this person’s existence and this person’s non-existence. Showing that such comparisons are possible clears the way for Benatar’s argument that one’s existence, as long as it includes some harm, will always be worse than one’s never having existed at all. Footnote 4

Benatar’s argument comes in three stages. In the first step we evaluate, in the second we compare, in the third we aggregate. In the first stage of his argument, Benatar evaluates the presence of the things that are good or bad for us and their absences. As he aims at a comparison of existence and non-existence, he is particularly interested in how to evaluate the absences of the good and the bad things when these absences are due to non-existence. Footnote 5 At the second stage, we compare. On the basis of the evaluations of the first stage, Benatar makes two comparative claims, comparing the presence and the absence of the bad things and, then, the presence and the absence of the good things. At the third stage, we put the comparisons of the second stage together to come up with a comparison between a person’s existence and her non-existence. The table below is the table Benatar uses to present the first stage of his argument. There Benatar talks about pleasure and pain but he makes it clear that these terms are simply placeholders for whatever is a benefit or harm to a person.

Scenario A (X exists)

Scenario B (X never exists)

(1) Presence of Pain (BAD)

(3) Absence of pain (GOOD)

(2) Presence of Pleasure (GOOD)

(4) Absence of pleasure (NOT BAD)

  1. (1)

    The presence of pain is bad.

  2. (2)

    The presence of pleasure is good.

  3. (3)

    The absence of pain is good, even if that good is not enjoyed by anyone.

  4. (4)

    The absence of pleasure is not bad unless there is somebody for whom this absence is a deprivation.

Claims (1) and (2) are, in their general form (when pleasure and pain are replaced by what is good or bad for X), trivial truths. Whatever is good for X is good for X and whatever is bad for X is bad for X. Claim (3) has puzzled Benatar’s commentators. How can the absence of pain be good for X if this absence is a result of there being no X?Footnote 6

Note that we can call something good on the basis of a comparison with how things would have been had what we call good not happened. I can say to a person who fell off the roof, ‘Good that you just broke a leg’. What I mean here is that such falls often result in injuries which are much worse. Although breaking a leg is bad for the one who suffers it, we can call it good because it is better (or considerably less bad) than what, in these circumstances, could have happened. Here is another example. Assuming that there are no posthumous harms, we can say that it is good for him that he was not around anymore when his bad deeds came to light. We mean that it was better for him to die before the discovery of his bad deeds than to have witnessed the moral outrage they have caused. A similar comparative thought underlies the idea that the absence of pain is good for X: it would be better for X not to suffer what actually harms him than to be harmed. Being harmed is worse (and worse for the person who suffers the harm) than not being harmed, so not being harmed is better than being harmed even if the harm doesn’t befall one because one fails to exist or, as in the previous example, fails to exist when what would have been harmful occurred. Benatar’s thought in (3) is simply that when we consider all the bad things that happen in a life, non-existence would be, in terms of the avoidance of these bad things, better for the person in question.Footnote 7

(4) recognises that absent pleasures are bad if there is someone who misses out on them. For example, when none of the expected guests show up, one will be disappointed. Even in cases in which there is no disappointment when, for example, the absence of the good fails to be noticed, we can still call this absence ‘bad’ on comparative grounds: it would have been better for an existing person, had the beneficial event happened. When it comes to non-existence, however, it would be wrong to claim that anything can be a harm or benefit (or be better or worse for) a non-existent X. This was the principle we found in Broome’s argument. I endorsed it and Benatar seems to rely on it, in his claim (4). Like with (3), I suggest to read (4) comparatively. The claim that absent pleasures are not bad when X fails to exist, means that the absence of pleasure, if there is no X, is not worse for X than the presence of pleasure. In the following passage, Benatar agrees with this comparative reading of (4).

Just as absent pleasures that do deprive are ‘bad’ in the sense of ‘worse’, so absent pleasures that do not deprive are ‘not bad’ in the sense of ‘not worse’. They are not worse than the presence of pleasures. (Benatar 2006, 41f.).

The second stage of Benatar’s argument is the comparative stage. I have read the two claims (3) and (4), which are the only source of any potential disagreement, comparatively. Thus, there is no more work to be done at this stage. I make the comparative nature of (3) and (4) explicit in the following two comparisons (3’) and (4’).

(3’) The presence of pain is worse for X than the absence of pain (even if no one enjoys the absence of pain).

(4’) The presence of pleasure is not better for X than the absence of pleasure (if there is no X who is deprived of the pleasure).

The qualifications in the parentheses of these two claims make it clear that we have been comparing pleasure and pain with special cases of their absence, namely with absences due to non-existence. We can make this reference to non-existence explicit and rewrite (3’) and (4’) in terms of characterisations of existence and non-existence, respectively.

(3*) In terms of the bad things in life, non-existence has an advantage over existence (because the absence of pain, if X doesn’t exist, is better for X than its presence).

(4*) In terms of the good things in life, existence has no advantage over non-existence (because the absence of pleasure, if X doesn’t exist, is not worse for X than the presence of pleasure).Footnote 8

A simple dominance rule helps us to aggregate these two judgements to an all-things-considered comparison of existence and non-existence for any person X: what has an advantage over something else without having any disadvantage must be the better of the two for X. This leads us to Benatar’s conclusion.

(Conclusion) Non-existence is better for X than existence and, thus, coming into existence is always a harm.

This presentation of Benatar’s argument does, I hope, do justice to what has moved his thinking.Footnote 9 There is, however, one divergence between his and my own presentation of his argument. In the end, my presentation avoids what I take to be a mistake on his part. The issue arises at the very start of the debate, namely with whether one’s existence and one’s non-existence are comparable. I’ve argued for a view we could call semi-comparabilism or, more precisely, semi-comparabilism about positive personal value relations between one’s existence and non-existence. Positive personal value relations hold between one’s existence and one’s never having existed at all, only if one exists. For an existing X, X’s existence can be better or worse than (or of the same value as) X’s non-existence. If X fails to exist, however, nothing can be better or worse for X, i.e. all statements of positive personal value relations are false.

In contrast to this picture, Benatar thinks about existence-non-existence comparisons in the following way. Take two possible worlds. X exists in only one of them. We can make the required comparison in terms of what better serves the interests of X regardless of whether X exists in the actual world or not. The evaluative standard provided by the interests of X, whether they are actual or not, can always be used to compare these two worlds. We can call this view a full-comparability view. Even if X does not exist, his existence (or anything else) can be better or, if Benatar is right, worse for X.Footnote 10

Full comparability is, however, implausible. It violates the idea that nothing can be good or bad for someone who fails to exist and this idea is plausible. Suppose I am married with two children and suppose that our house burns down. This is bad for my wife, for me and for my two children. We loved our house and so our interests have been negatively affected by the house burning down. Although this event is bad for us, it is not bad for my third child because I only have two children. On Benatar’s view about comparability, in contrast, it would also be bad for my third child, a child I never had. The child I could have had would have been affected by our house burning down in the same way as my actual family has been. So, its interests would be negatively affected by the burning down of the house even though these interests are the interests of a child which doesn’t exist. On Benatar’s view, the loss of my house is not only bad for my actual children but for all the children I could have had as long as their interests remain aligned with ours.Footnote 11 As I said, I find this implausible. I accept the view that nothing can be good or bad for someone who fails to exists. Benatar must reject it. This is a weakness of his presentation of the asymmetry argument and it has further ramifications.

Remember that I supported (4), the idea that absent pleasures are not bad by a comparative reading. For one who fails to exist, nothing can be worse, and this includes the absence of pleasures. Given full comparability, however, (4) comes out as false. Given the interests the person who could have existed would have had, not to exist has a disadvantage: it does not give one the pleasures of life.Footnote 12

In reaction to this criticism of (4), Benatar tries to shift the debate. He appeals to explanatory advantages of accepting (3) and (4). According to this new defence, his basic asymmetry – absent bads are good but absent goods are not bad – helps to explain other views we hold, most notably the procreational asymmetry according to which we have a duty not to create unhappy lives whereas we have no duty to create happy lives.

… it has been suggested that if we can claim that absent pain is good for a person then we can make the symmetrical claim that absent pleasure is bad (Harman 2009, pp. 781–782). According to this argument, we should deny (4) of my basic asymmetry. The suggestion here is that it would “be bad, for the non-existent person we might have created, that his pleasure not occur, because it would have been good for him if it had occurred” (Harman 2009, p. 782). The mistake in this objection is that it misconstrues my basic asymmetry as a logical rather than axiological claim. We certainly can (logically) state that just as the absent pains in Scenario B are good, so the absent pleasures are bad. The problem, I have suggested, is that we should not claim this. Among the reasons for this is that we would then not be able to make all the value judgments we do in the four asymmetries that I say are explained by the basic asymmetry. (Benatar, 2013, 126)

Here I won’t follow Benatar’s lead for two reasons. First, I have argued that we do not need the full comparability view to defend (3) and (4). Thus, we already have an anti-natalist argument to be discussed. Secondly, as long as Benatar holds on to the full comparability view, (4) is not only doubtful, it simply is false. The possible X has to think of the good things in life that he would enjoy were he to exist as being better for him than their absences. (It is only if all claims of positive value relations for X are false if there is no X that we can claim that the presence of good things is not better for X than their absence as there is no X.) Thus, whatever else can be said in defence of (4), it won’t help as long as we work with full comparability. Once we abandon this idea, however, we will come back to the argument that I’ve developed earlier. For this reason, if there is any argument for Benatar’s anti-natalism, it is the one outlined earlier.Footnote 13

The core of Benatar’s argument are the following two comparisons.

(3*) In terms of the bad things in life, non-existence has an advantage over existence because the absence of pain, if X doesn’t exist, is better for X than its presence.

(4*) In terms of the good things in life, existence has no advantage over non-existence because the absence of pleasure, if X doesn’t exist, is not worse for X than the presence of pleasure.

I’ve argued that comparisons between alternatives become ambiguous if the alternatives themselves invoke their own diverging standards of comparison. There is a sense in which the notions of personal value do not get a grip on non-existence. What I mean is that if P fails to exist nothing is good for P or bad for P or neutral for P (nor is anything better for P or worse for P or of equal value for P). Whenever I make an evaluation from there, i.e. under the assumption of non-existence, all I get are negations of personal evaluations. The house burning down is neither good nor bad for my third child, if, as we have assumed, there is no such child. But I can also look at non-existence from here, i.e. under the assumption of my existence, with my set of present concerns as the standard of comparison. (I will briefly indicate how to think of this comparison later on. At the moment we should focus on the ambiguity that arises in cases of such comparisons.) If I look from here at one of the bad things in life, an episode of intense pain, for example, which was not part of any meaningful process, like in grieving or in feeling regret, or had positive effects later on, then the absence of this episode of intense pain would be better for me. It is this thought that underlies (3) in all its variants.

In contrast to (3), which expresses an evaluation from here, (4) expresses an evaluation from there, i.e. under the assumption of non-existence. Whether the house burns down or whether we win the lottery, nothing matters to the third child I never had. For this third child, the pleasures of life are no advantage. (If I had a third child, i.e. under the counterfactual assumption of its existence, the third child would suffer from the loss of our home.) The evaluations (3) and (4) arise from two different perspectives or, if one objects to talk about ‘the perspective of non-existence’, we can instead claim that the two evaluations arise under different assumptions, the assumption of existence in the case of (3), the assumption non-existence in the case of (4).

Let us first make these assumptions explicit in (3*) and (4*).

(3*from here) In terms of the bad things in life, non-existence has an advantage over existence because the absence of pain, if X doesn’t exist, is better for X than its presence.

(4*from there) In terms of the good things in life, existence has no advantage over non-existence because the absence of pleasure, if X doesn’t exist, is not worse for X than the presence of pleasure.

Now we can add two more comparisons. A comparison regarding the presence and absence of the bad things from there and a comparison regarding the presence and absence of the goods things from here.

(3*from there) In terms of the bad things in life, existence has no disadvantage over non-existence because the presence of pain one would experience if existing is not worse for one who fails to exist.

(4*from here) In terms of the good things in life, existence has an advantage over non-existence as the pleasures of life are better for X than their absence.

On the basis of the above two comparisons alone, one could conclude that existence, as long as it contains something positive, is always better than non-existence because existence, we said, has no disadvantage when it comes to the bad things and an advantage when it comes to the good things. However, both the anti-natalist argument, according to which existence is always a harm, and the pro-natalist argument, according to which existence is always a benefit, make the same mistake: they apply one set of standards, the standards inherent in existence, to one comparison and a different standard, namely the lack of any positive standard in cases of non-existence, to the other comparison. But when we aggregate to an overall comparison of existence and non-existence, we need to have the same standards in place for both component comparisons that determine the overall comparison. Thus, we can only coherently think about the comparison between existence and non-existence if we either adopt the perspective of existence and its standards or if assume our non-existence and its lack of any standards. The second comparison is easy and its result is trivial: nothing matters to those who do not exist because there are no such people. The first comparison is often easy in practice – we are glad to be alive – but understanding on what basis we make these comparisons is a philosophical challenge.Footnote 14

However, the aim of this part of the paper does not require us to tackle this issue. We wanted to show what goes wrong in Benatar’s argument. I have dismissed full comparability. I offered an anti-natalist argument which uncovers Benatar’s thinking whilst improving on his presentation by relying on semi-comparability. We uncovered the mistake which underlies anti-natalist thinking. The anti-natalist argues for an overall assessment of existence and non-existence by relying on partial comparisons which employ different standards. The bad things are evaluated from the perspective of existence and so they do matter; the good things are evaluated from the perspective of no-existence and so they don’t matter. If such a procedure were allowed, we could, with equal legitimacy, produce an argument with a conclusion that life, regardless of how bad it is, is always a benefit as long as it contains some good. (From the perspective of existence the good things matter; from the perspective of non-existence the bad things don’t matter as nothing does.) Instead of endorsing both arguments, we should endorse neither. This form of argument is invalid.

It is, I think, worth pointing out the mistake in Benatar’s argument because it is a mistake that isn’t easy to spot. If it were, it would have gained prominence in the anti-natalism debate by now. The underlying analysis points to a general fact – comparisons can become ambiguous if the compared alternatives invoke their own standards of comparison – which has not yet received the full attention it deserves.

2 The Second (Moral) Objection

‘Sudan’ was the name of the last male northern white rhino. Now there are only two females left. They need constant protection to save them from poachers who hunt them for their horns. David Attenborough tells us how sad he is to see this great species of animals become extinct. The anti-natalist, however, who happens to care about the white rhino, tells us how glad he is: finally, they are about to become extinct. No more pain for the white rhino. The white rhino is just one example. The ideal population size for any species, Benatar tells us, is zero. The ideal world is a world free from humans, animals and plants. The ideal world, for the anti-natalist, is empty of all life.

In order to ensure a proper moral reaction, let me develop anti-natalism in the direction of what is called pro-mortalism. We apply something like Benatar’s reasoning to the future lives of living beings, including our family and friends. If they stop existing now, nothing bad is going to happen to them anymore and, furthermore, once they stopped existing, they won’t miss the good things that would have happened to them. So, according to such reasoning, it would be better for everyone, including those we love to stop existing now.Footnote 15 On this view, and bracketing effects on others, we ought to kill those we love and welcome it if they kill themselves. Human history is full of atrocities. Think of genocide. It feels improper to spell out the consequences of such a view any further.

How should we react to an argument which concludes that, when considering just what is good for them, one ought to kill one’s children? One might feel disgust and revulsion, ‘How can anyone say such a thing? Philosophy is not only absurd it is also dangerous.’ This reaction is of the same kind but a step up from the embarrassment I felt when anti-natalists discussed how to end all life. My attitude to life is different. I am glad to be alive and I am glad to live in a world full of life. Mine is a common attitude. ‘In all things of nature’, says Aristotle in Parts of Animals (645 a4-6), ‘there is something marvellous’. I will call this the pro-life attitude.Footnote 16 To clarify it I make two observations. First, this attitude neither denies the conflict between forms of life nor that we will take sides in some such conflicts. Secondly, in the context of this discussion, I’m interest in this attitude as something that may come before philosophical reflection and need not be its result. It is an attitude that provides material for and may guide philosophical reflection. In this sense, I consider it in its pre-philosophical or pre-reflective form. There are other firm views that are common and widespread and that come in such pre-reflective forms (in which they provide material for philosophical reflection). The idea that there is a mind-independent external world, the idea that there are other people around me who are in important ways similar to me which provides them with moral standing, the idea that we can be held to account for what we do, that choices are up to us and that we continue to exist to witness the result of our choices. The idea that our senses provide an accurate picture of the world around us and so on.

When I object to the discussion of how to end all life, I say more than that this discussion rests on a false presupposition. I say that it’s wrong to claim that we ought to end all life – we ought to protect life – and I say this in a distinctively moral tone. But what role can my pre-reflective pro-life attitude play in discussing these matters?

I will distinguish between three options of how to answer this question. First, our pre-reflective moral attitudes may set barriers to the content of philosophical reasoning. It simply cannot be true that we ought to end all life. Any argument to the contrary is not only pointless, as it tries to undermine what we know to be true, it also deserves moral condemnation. Moral attitudes play, according to this view, an external role: they are independent arbiters of the results of philosophical reasoning. These attitudes are not themselves part of philosophical reasoning; they simply condone or condemn its results. According to the second view, our moral attitudes and intuitions are part of the philosophical debate. They play, in this sense, an internal role. They are part of the mix which philosophy deals with. Like everything else, they need argumentative support and they need to engage with any argument launched against them. After discussing these two approaches, I will develop an intermediate third position, according to which the role of our pre-reflective attitudes is not restricted to their internal role; they remain important independently of the results of philosophical argument. However, they don’t set barriers to the results of philosophical debate; their effect is not to censure and limit philosophical debate to views which agree with these attitudes. This intermediate position accepts the internal role of pre-reflective attitudes. It rejects the external role as characterised above but wants to capture a different kind of external role for our pre-reflective attitudes.

Let me start by considering the external view: our moral attitudes set boundaries for moral philosophy. Anscombe famously said that ‘if someone really thinks, in advance, that it is open to question whether such an action as procuring the judicial execution of the innocent should be quite excluded from consideration—I do not want to argue with him; he shows a corrupt mind’ (1958, 17). Anscombe says that someone for whom everything is on the table – even the hanging of the innocent – is morally corrupt. Not only is a view without moral absolutes – never and under no circumstances may one punish the innocent – false, adhering to such a view bars one from being a respectable interlocutor. Both the view and its defender are subject to moral criticism.

Sami Philström has launched a critique of this kind against Benatar’s anti-natalism. ‘Perhaps there are philosophical ideas and arguments that ought to be left aside precisely because they violate some human values and ideals that are cherished more deeply than the idea of sound argument itself’ (Philström, 2009, 657). Benatar’s view, he says, is, ‘too far off the humanly possible moral scale even to be seriously disagreed with’ (ibid.) ‘There are’, he says, ‘moral limits to what we may think and argue, even in philosophy’ (Philström, 2009, 665). Like in Anscombe’s case, this looks like an external criticism, in the sense above: firmly held moral attitudes stand outside the philosophical debate and, in the cases at hand, condemn opposing views.

If deeply held moral convictions could be a barrier to philosophical thinking in ethics, other deeply held views would bar much of philosophy altogether. Think about the denial of the external world or the denial of other minds. Consider the sceptic’s claim that we don’t know what we think we know, or the moral error theorist who argues that all ethical statements are false. The outcry directed at the sceptic that one can see the world does not by itself undermine the sceptic’s argument. It is an important part of philosophy to raise challenges to deeply held views. If these views functioned as legitimate barriers against philosophical thought, there wouldn’t be much philosophy left. So, I am not sympathetic to the external view.

‘I can see my hands’, I said, doesn’t, by itself, undermine scepticism. It is not an external condition to which philosophical thought has to align itself. It can, however, be part of a response, a so-called Moorean response, to scepticism. The details don’t matter here. A firmly held conviction – I can see that I have two hands – becomes itself the starting point of a philosophical view. In this way, firmly held convictions are part of and thus internal to philosophical discourse. Not wanting to dismiss Anscombe or Philström out of hand,we can apply this strategy to their views. According to this internal reading, Philström (2009, 666) offers a transcendental argument against Benatar’s claim that all life ought to end along the lines that thinking about what ought to be requires a positive attitude to ourselves. Missing that attitude, we could not even take ourselves to argue for anything: ‘The departure from our common standards of rightness is so radical that no such evaluation [of any argument] is possible anymore.’ I have my doubts. However, here we need not check whether Philström’s argument succeeds. Like the Moorean response to scepticism, I use it simply to illustrates the internal perspective where firmly held views are part of philosophical arguments against challenges to these very views. Anscombe can be read along similar lines: her criticism of consequentialism as condoning what according to her pre-reflective attitude are simply evil deeds is part of a larger philosophical argument which crucially draws on the claim that consequentialism lacks a plausible moral psychology.

On the internal view and apart from their role in wider arguments, our pre-reflective attitudes play no further role. One such wider argument uses the idea that we would not have a view were it not true so that, with this additional premise, the fact that we hold an attitude supports the attitude itself. This argument is criticized by so-called debunking arguments which aim at showing that there are better explanations for why we hold certain views and attitudes than their accuracy. Once such a strategy of defending pre-reflective views is dismissed, any philosophical argument for a conclusion in tension with a pre-reflective view will be seen as undermining this view. Benatar makes this dialectical move in the following response to the claim that his view is counterintuitive. ‘We should not put too much store on these intuitions, for they are unreflective intuitions formed under the influence of well-demonstrated biases. Once we begin to think about it, there should be nothing counterintuitive about a view that, if acted upon, would eliminate all (human) suffering.’ (Benatar, 2012, 164).Footnote 17

The intermediate view about the role of pre-reflective attitudes in philosophical debates tries to secure a place for these attitudes that goes beyond the internal view in which they are simply part of the mix that enters philosophical reflection. These attitudes should count for something independently of their role in philosophical arguments. I find the intermediate view plausible. I will start building a case for it by distinguishing between the results of philosophical arguments on the one hand and their implementation in practice (or the consideration of such implementation) on the other hand. I will argue that intuitions play an external role, not as barriers to what we can argue for in philosophy, but as legitimate obstacles to the implementation of whatever it is that a philosophical argument seeks to establish.

Having read Parmenides and Zeno, someone might come to think that motion is impossible. Philosophical arguments support a view which, intuitively, we find implausible. If a modern admirer of Zeno excused himself from attending meetings on the basis of the impossibility of motion –an example of an attempted implementation of a philosophical view – this would meet with ridicule. The pre-reflective view that motion is possible as well as the appearance of motion play a role in rejecting any such excuse. In order to be an obstacle to their implementation, this view need not engage with Zeno’s Paradoxes. One doesn’t need to find fault in Zeno’s reasoning to object to this attempt to implement this form of philosophical scepticism.

Structurally, my reaction to Benatar’s claims is analogous. I happily discuss whether existence is a harm or a benefit. However, I feel unease when it comes to discussing the implementation of his view, namely how to end all life. Like in the case of motion scepticism, the feeling that such a discussion is inappropriate has some standing even if it fails to engage with arguments to the contrary. When it comes to the implementation of a philosophical view, we take an extra step and it is here that pre-reflective views legitimately push back against the implementation of the results of philosophical reasoning. Note that in contrast to the external reading of Anscombe’s anti-consequentialism, these views and attitudes pose no limitation on the results of philosophical reasoning, they only affect their implementation.

I want to support this picture of the role of what I called pre-reflective attitudes, firmly held views or intuitions with two examples. One is provided by David Hume in whose writing we find two very different reactions to philosophical argument. The intermediate view allows for such differences. The other is the debate about how to react to what for some is a metaphysical discovery, namely that we don’t satisfy the conditions required for ultimate responsibility. I start with the latter.

The minimal implementation of any philosophical reasoning would be the adoption of the attitude which it supports. In cases in which philosophical argument delivers radical conclusions, like when we think about moral error theory or various forms of scepticism, the view which is undermined plays a different role when it comes to the philosophical argument itself and when we consider its implementation (even in the minimal form of implementation which is attitude revision). Our belief that people are sometimes fully responsible for what they do, for example, plays no direct role in incompatibilist arguments which try to show that the control necessary for responsibility is illusionary on metaphysical grounds. Nevertheless, the usefulness of the practices built on the acceptance of responsibility makes a case for arguing against adopting the attitude supported by philosophical argument. Smilansky (2001) has argued for illusionism about free will. Though we are not free and we are not ultimately responsible for what we do, we need to think that we are. Most people, according to Smilansky, live in this illusion. ‘Humanity is fortunately deceived on the free will issue, and this seems to be a condition of civilized morality and personal sense of value. Illusion and ignorance appear to be conditions for social and personal success.’ (Smilansky, 2001, 85). Despite a strong philosophical case for a view, it ought not to be implemented. Part of the explanation is our strongly held view to the contrary (and the resulting benefits). This is my first illustration of the intermediate view.

Here is the second. In the Treatise of Human Nature (Book 1, Sects. 7, 1978: 263–274), Hume reflects on the intellectual situation his philosophical reflections have put himself into. ‘For I have already shewn that the understanding, when it acts alone, and according to its most general principles, entirely subverts itself, and leaves not the lowest degree of evidence in any proposition, either in philosophy or common life.’ Hume, however, doesn’t attempt to put this sceptical conclusion into practice. There is, for example, no demand to close all schools and universities and other institutions whose very idea is the generation and transmission of knowledge. Quite the opposite: he’s happy about being distracted from what he, with his normal sensibilities, regards as dire conclusions. ‘Most fortunately it happens, that since reason is incapable of dispelling these clouds, nature herself suffices to that purpose, and cures me of this philosophical melancholy and delirium, either by relaxing this bent of mind, or by some avocation, and lively impression of my senses, which obliterate all these chimeras. I dine, I play a game of backgammon, I converse, and am merry with my friends; and when after three or four hours’ amusement, I would return to these speculations, they appear so cold, and strained, and ridiculous, that I cannot find in my heart to enter into them any farther.’

At the end of the Enquiry Concerning Human Understanding (Hume, 1975: 175), we find a different Hume, one who eagerly recommends implementing his philosophical principles. ‘When we run over libraries, persuaded by these principles, what havoc must we make? If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.’

Depending on the importance and the firmness of our pre-reflective views and attitudes, we can either see scepticism as a public policy programme – stop teaching as there is nothing to teach – or, which is more common, we see it as a philosophical problem for our everyday claims to knowledge. Seeing the results of philosophical reflections as posing problems for common-sense views falls short of adopting the attitudes which would be in line with the discoveries of philosophy. This strategy – respecting philosophical research by recognising that it sets problems – is always available. When we take it, we won’t have the attitude that recommends burning books. (Even if we were fully committed to Hume’s principles, there would be other reasons not to implement philosophical thought in this way.)Footnote 18

Many people value and admire life in its many forms. Like in Hume’s case, we can regard Benatar’s conclusions either as the basis of a public policy aiming to eradicate all life or as outlining a problem for our pre-conception of life as something marvellous. This choice itself is not decided by philosophical argument alone. If we opt for adopting some radical conclusion, we show a certain character: we might not share to the same extent as others the sensibilities that give rise to these pre-reflective attitudes or we might overestimate the force of philosophical argument and, if they are of our own making, the force of our own ingenuity. Thus, in adopting a particular philosophically supported conclusion, we may either lack in common sensibilities or in intellectual humilityFootnote 19 or in both. This is my moral objection against the anti-natalist discussion of how to end all life.Footnote 20

I end with three clarifications. First, the idea that some views are important to us does not imply that the benefit of holding such views is a reason to adopt it. The importance of a view might simply set a high threshold for arguments which aim at changing deeply held convictions.Footnote 21 Secondly, what I say about the force of philosophical arguments – namely that it is a question of character whether we adopt the attitudes they support – will apply to arguments of all kinds unless one provides some feature of philosophical reasoning that makes it especially susceptible to doubt. Although it is true that any philosophical view has its supporters and arguments in its favour, I will have to leave this issue open here. Thirdly, and most importantly, what I called a lack of sensibility or a lack of intellectual humility, others, who start from a different point of view, may call obstinacy, dogmatism and intellectual inflexibility. De Grazia seems to take this attitude when it comes to Benatar’s opponents. Benatar’s discussion, he writes, ‘… demonstrates a willingness to stare uncomfortable truths in the eye and accept them as reality. It is, I think, part of philosophers’ shared self-image that we are pretty good at this – that we are good bullshit detectors – and that we care enough about the truth to accept it and speak honestly about it even in the face of significant unpleasantness.’ Benatar’s discussion ‘displays the virtues of intellectual honesty and courage’ (DeGrazia, 2010, 320).

My moral objection to the implementation aspect of Benatar’s work, shows the opposite reaction. The pro-life attitude is not, to use DeGrazia’s term, ‘bullshit’ which Benatar has exposed. I accept that my objection takes the pro-life attitude outlined earlier as a given. If you share this attitude, then you may join David Attenborough in lamenting the forthcoming extinction of the white rhino. I admit, however, that this conclusion is only conditional. If you don’t share this attitude, and you follow the view that every life would better not be to its conclusion then, out of your concern for living beings, you will try to work out how best to rid this planet of all life. Is embracing this project bold and courageous, as DeGrazia has it, and its rejection dogmatic and timid, or is its consideration reckless and dangerous and being present when it is discussed embarrassing and unpleasant?Footnote 22

According to the intermediate view, it depends, in part, on your commitment to our pre-reflective attitude to life. If you really fall on the other side, and you hail Benatar’s view as a discovery that, when fully implemented, will have rid the world of all suffering, please note that there is a problem for this programme. It is not a moral problem for you as you don’t share the sensibilities on which my moral objection is based. It is a purely internal, philosophical problem for Benatar’s view. I have devoted the first part of this essay to it. Benatar’s asymmetry argument for his anti-natalism does not succeed.

3 Conclusions

In this paper, I have argued that once we understand the reasons for the premises of the asymmetry argument, we realise that the argument trades on a general ambiguity that affects evaluative comparisons whenever the compared alternatives contain their own diverging evaluative standards. In this way, the symmetry between good and bad and its presence or absence can be restored. Given the standards of existence, the good contained in a life is an advantage over non-existence and the bad is a disadvantage. Given the standards of non-existence, nothing matters to X, because there is no X. Furthermore, I tried to show that our firmly held pre-reflective attitudes, whilst not restricting the content of philosophical debate, play a legitimate role when it comes to the implementation of philosophical findings. If one is attracted to Aristotle’s pro-life attitude, one will shake one’s head in disbelief and disapproval when one encounters debates about how to end all life. This reaction, I have argued, does not require philosophical validation in order to be legitimate.