1. NEUTRALITY
The intuition of neutrality [N] is, Broome says (143), one that he and many others find attractive:
[A]dding a person to the world is very often ethically neutral . . . [A] person's existence is neutral in itself, setting aside its effect on other people. There is no consideration stemming from the wellbeing of the person herself that counts either for or against bringing her into existence. (143–4)
These quotations contain two versions of N: that adding a person is very often neutral, and that is always neutral. Broome's view is the former. N has an actual limit, and a possible limit. The actual limit is in relation to lives ‘at a poor level’ (144; cf. 146). It is tempting to think that any life clearly worse than nothing is at the poor level, but Broome doesn't say. The possible limit relates to lives which go ‘very well’ (235).
Given these limits, it is clear that Broome is wrong to say that N is ‘part of the broader way of thinking known as the “person-affecting view” . . . the view that only benefits or harms that come to people can be ethically significant’ (145). N is a hybrid view: person-affecting within the limits, but beyond the limits non-person-affecting. Broome sharpens the above statements of N into the following principle:
Principle of equal existence [PE]. Suppose two distributions have the same population of people, except that an extra person exists in one who does not exist in the other. Suppose each person who exists in both distributions is equally as well off in one as she is in the other. Then there is some range of wellbeings (called ‘the neutral range’) such that, if the extra person's wellbeing is within this range, the two distributions are equally good. (146)
PE is a problem for Broome, since it conflicts with the idea of the separability of lives, grounded on the principle of personal good (120, 146). He accepts the plausibility of counterexamples to PE based on the principle of personal good (146–8), and then spends a good deal of time in chapters 11–12 testing responses to these counterexamples on behalf of PE. In the end (14.2), he gives up N because of its ‘greediness’. But he is clearly unhappy to have to do so. So if N could be dealt with immediately, it would save him time and make him happier.
Broome does not offer an epistemology. He may be claiming that N is self-evident, in the sense that it is plausible in itself. If he is saying this, then he must face up to the fact that many people find the opposing principle equally plausible: that adding a life worth living to a population is, other things being equal, positively good. Further, anyone who asserted either of these views as self-evident would be asked to specify what they mean with examples. Broome does indeed offer an example:
The Couple. A childless couple decide that their own lives will be better on balance if they have no children. So they have no children.
Broome says that we do not think that this couple are acting wrongly. But note the following.
First, the example would have to be fleshed out for common sense to have a clear view on it. Many people will think that this couple are excessively absorbed in their own concerns, or that they have the wrong view of wellbeing (another view of common sense might be that children, in our society, always increase their parents’ wellbeing). As the example becomes more detailed, it may lose its apparent force.
Second, as mentioned, there is a question about the place of examples in Broome's epistemology. The conclusion he reaches in his book is broadly utilitarian, and it is well known that utilitarianism conflicts with many common-sense intuitions. So if denying N makes Broome unhappy, it is unclear why much else in common-sense morality doesn't do the same.
Third, common-sense morality, once we have further specifed the case in question, on Broome's understanding of it seems to be in internal tension. One assumes that Broome is talking about a couple now, in a world with a population of about 6 billion. But now consider:
Adam and Eve 1. Adam and Eve are the first and only people in the world. If they have children, they know (because God has told them) that many billions of people will live lives well worth living (though no one will reach Broome's ‘very high’ level). They decide that their own lives will be better on balance if they have no children. So they have no children.
Common-sense morality's deliverance seems to be that Adam and Eve are selfish in giving no thought to lives that might be lived by their dependants.
Broome may object that this is a case in which numbers are large, and our intuitions cannot be trusted. When he makes this point about numbers, it is in a context where we cannot appreciate the value of large populations, whereas in this context it seems, of course, as if common sense can do exactly that. (It is also worth pointing out that, since 6 billion is a pretty large number, Broome's point about large numbers seems to count against his own example.) But let me anyway change the example:
Adam and Eve 2. Adam and Eve are the first people in the world. If they have children, they know (because God has told them) that 1 people will live lives well worth living (though no one will reach Broome's ‘very high’ level). They decide that their own lives will be better on balance if they have no children. So they have no children.
Even when the numbers are easily comprehensible, the verdict of common sense seems the same as in Adam and Eve 1.
When he is discussing whether the neutral level might depend on the number of people in a population, Broome considers the suggestion that, if humanity has a special value in itself, the first few people will be especially valuable (197). He deals with this by stipulating that there are already at least a few thousand million in the population. Again we find Broome appealing to intuitions about large numbers. But note anyway that he could not use the strategy of assuming a large population to deal with either Adam and Eve case. In both, the existing population is only two.
Finally, it is significant that, if we accept that common-sense morality has evolved to solve certain societal problems, we should expect the common-sense view on The Couple to emerge. Usually people want children, and there is no need for the sanctions of morality. When childlessness is causing problems for society, people's views change. In the time of the Emperor Augustus, for example, childlessness was seen as one of the vices of the late Republic, like sexual immorality. This point is not sufficient to debunk the common-sense view of The Couple, but it does throw doubt on it.
So there are problems with Broome's use of the case of The Couple. Another of his arguments to support N is the fact that economists seem to accept it when they are setting a value on people's lives. But unless we assume that economists have some special normative insight this point cannot add any weight to N. These economic evaluations rest on common-sense morality, and the plausibility of the deliverances of that morality are what is at stake.
Let me end this section by mentioning what I think is the most serious problem with N: its asymmetry. As Broome says, many people accept something like N about good lives. But when a few decades ago certain people realized the implications of a person-affecting view for bad lives, most of them adopted a non-person-affecting principle. To adopt a hybrid view, especially one which allows very good lives to count, seems unmotivated. Indeed, given Broome's general reliance on systematic and abstract theory in Weighing Lives, I was surprised that he didn't go further into this prima facie odd aspect of N, seeking to spell it out and to offer some explanation of its hybrid nature. If bringing a life into existence at a poor level is bad, then for anyone sympathetic to utilitarianism it is a mystery why failing to bring a good life into existence is not equally bad.
2. PLEASURE
After reluctantly giving up N, Broome turns to the question of what the level of wellbeing is such that it makes no difference to the overall value of some outcome whether that person comes into existence – the neutral level (208–10). Though he himself has no answer, he does reject one theory that implies a particular, and particularly clear, neutral level: hedonism.
Broome understands personal hedonism – that is, hedonism about wellbeing – as the view that wellbeing consists in the balance of good (i.e., pleasant) experiences over bad (i.e., painful) experiences. This is a theory of ‘goodness for’. General hedonism is a theory about general goodness, and it says that the value of a distribution depends on the overall balance of pleasure and pain in the lives of all in that distribution. According to personal hedonism, a ‘blank’ life, such as one lived in a coma, has neither positive nor negative value for the person living the life. Therefore, according to general hedonism, the level of the blank life is the neutral level. Further, general hedonism gives us not only a neutral level, but a non-vague one. There is nothing fuzzy about the boundaries of the blank life. So what does Broome have against it? His first objection is to the move from personal to general hedonism:
No doubt good experiences benefit the people whose experiences they are. But I do not see why we should think that, independently of this benefit to people, good experiences are in themselves good to have in the world. (209)
I suspect many people would find this view hard to understand. Consider the following case:
Two Futures. There are two possible futures. In one, a billion people fall into a coma. In the other, these same people live extremely pleasant lives.
Most would say that the second future is much better – much better ‘to have in the world’ – than the first.
It could be, however, that Broome is thinking of goodness in a more substantive way. On a ‘formal’ understanding of goodness, the goodness of an outcome consists merely in the reasons to bring it about. So a strict deontologist, for example, may say that the outcome in which she keeps a promise is better than one in which she breaks it. But that is just because she has a duty to keep a promise, and duties give reasons to act. It's not as if that outcome contains some substantive value, such as beauty, constituted by the promise's being kept or having been kept. If Broome is thinking of substantive goodness, then it will indeed seem hard to understand what value the world can contain, according to general hedonism, over and above the goodness of pleasure for those experiencing it. But general hedonism says that the substantive value of any world is to be understood solely in terms of the pleasure in that world. There is no substantive goodness over and above the goodness of pleasure for those who experience it.
Broome's second doubt about general hedonism is its implication that, because it doesn't matter in whose life an experience occurs, longevity has no value. He discusses the value of longevity as an objection to separability of times in Section 7.2, and the structure of his claim there is analogous to that concerning N. The view about longevity is stated, and then an example is offered. The claim is that it is an advantage of one distribution over another if it contains one long life rather than two (or presumably more) short lives.
I suspect that there would be a good deal of disagreement about this as a general principle. As Broome allows (108), if the first life lasted 200 years, some would see longevity as a disadvantage. So, as with N, we again have a principle with limits, and some work needs to be done to explain both the principle and its limits.
Broome's example in this case is:
Replacement. A baby is ill. There are two alternatives: Save the baby's life, or let her die and replace her with another baby.
His claim is that we think it better to save the life of the baby. Note first that this view does not imply any value in longevity itself. The two babies will, we may assume, live lives of equal length. Rather the thought is that continued life is better than replacement, which in certain distributive options would lead to our preferring to extend a life rather than end it and start a new one.
Again, however, questions may be raised about the example. First, what is the role of such examples in Broome's overall epistemology? He is appealing to the common-sense bias in favour of the actual over the potential, expressed in views such as that it would be heartless of the parents or the doctors in Replacement to leave an existing baby to die to be replaced by the actualization of a potential baby. If Broome wishes to give weight to such common-sense intuitions, then it is unclear why his theory of distribution does not end looking more like the ‘dogmatic intuitionism’ of Frances Kamm or Judith Thomson, who see the role of ethics as an attempt to sharpen common sense. Further, how do such intuitions sit alongside his broadly utilitarian conclusions? Yet further, we should note that both broad utilitarianism, and more specific general hedonism, can provide a two-level explanation of the existence of such intuitions and their merely instrumental value.
Second, the general hedonist may again provide examples which, if we assume that the case against hedonism is not already proven, suggest that common sense is in internal tension and/or that replacement is in fact neutral. Consider:
Two Worlds. God creates two worlds. In the first individuals live lives of continuous mild sensual pleasure for 100 years each. In the second it appears that the same is happening. But God entirely replaces all the matter in the world with new matter, identically situated, at the end of every year.
Although I can see why being in the first world might be worse for one of its inhabitants than being in the second, since at any time each has at most one year to live, it is hard to see what special addition could be made to general goodness by the first world rather than the second.
Broome's third doubt about general hedonism is that it is intuitively implausible that a blank life is neutral. A person who comes into the world and lives out her life in a coma lives a ‘meaningless’ life, and ‘the occurrence of a meaningless human life is intuitively a bad thing’.
Here the general hedonist is entitled to more argument in support of the intuition of meaninglessness. There would again be a good deal of disagreement about it (myself I find the very idea of ‘meaning’ independent of wellbeing hard to understand and hard to accept). Where is meaning in human lives to come from? If from God, then Broome needs a theology. If from goals accomplished or from human relationships, then Broome needs to explain why the value of meaning here does not collapse into the value of wellbeing, so that his objection would amount to no more than the assertion of a contrary theory of personal wellbeing. (Broome does indeed go on to interpret his own claim as that ‘a blank life is worse than not living at all’ (209). But I think this must be a slip. He means that the living of a blank life makes a world worse than a world the same but without that life.) Broome would also need to explain why, even if meaning does add general value to a world, lack of meaning is not just neutral – in the same way that as, say, beauty adds value, mere lack of beauty is neutral, and only ugliness is bad. A human being without consciousness is plausibly seen merely as a biologically functioning system, and it is not clear why the existence of any such system, in itself, could make a world worse. Could it be that, again, Broome is trading on common-sense intuitions – in this case about people who are in comas as a result of accidents, whose projects have been interrupted, whose friends and relatives care about them deeply, and who are objects of general concern within common-sense morality? If so, then the same set of questions as that above arises about the place of these intuitions in his overall epistemology.
Let me note a further point. Broome says (209–10) that he has the same doubts about general hedonism as he has about any theory according to which the general goodness of a distribution depends only on the quantities of the features that make lives lived in those distributions good or bad. This claim is hard to understand. First, imagine someone who thinks that the meaning of a life is part of what makes it good. If we apply a version of Broome's first argument against general hedonism to that claim, then we find him committed to the view that meaning, independently of its value to people, does not add to general goodness. But his third argument against general hedonism depends on exactly this claim. Second, someone who thinks that meaning is important may spell out that value in terms of what is accomplished in the life as a whole, and so deny separability of times in her account of personal wellbeing. So Broome's second argument against general hedonism would also be inapplicable. (The weight of Broome's objection to the meaning-view could indeed rest on a version of his third argument. But the proponent of the meaning-view might plausibly argue that the blank life, though it lacks meaning, does not make a world worse, since non-meaning is mere lack of meaning and not in itself bad.)
Note also that someone who thinks that the blank life is represented by the neutral level is not committed to the broader theory to which Broome is here objecting. For example, someone may think that certain principles of distribution are historical, so that understanding the value of a distribution requires understanding how certain goods have been acquired and transferred, and that unfairness in such processes makes a distribution worse independently of its effect on wellbeing. There is nothing to stop such a person accepting that the blank life represents the neutral level when it comes to the question of whether adding such a life to a population would make the world in question better or worse.
There is one final argument of Broome's I wish to discuss. Towards the end of the book, he includes a sub-section (242–3) which criticizes personal hedonism. He cites a passage of Nagel in his support, and, having repeated the longevity objection, appears to take seriously the view that a life with experiences, none of which is either good or bad, is above the neutral level. If that were so, of course, personal hedonism would be false, since according to it the value of such a life is equivalent to that of the blank life and therefore neutral. Such a view might follow from, say, an objective list theory of wellbeing, according to which accomplishment is of value. We might imagine someone with a form of anhedonia preventing any kind of enjoyment or suffering who accomplishes something significant. We would have to rule out the notion that their experiences of their accomplishment are good for them, since we are postulating a life with neither good nor bad experiences. On the face of it, this view of wellbeing strikes me as quite implausible. I can see why one might want to say that such a life is a good life, or a good human life; but it is somewhat controversial to claim that it is good for the person living it. Anyway, if something like this lies behind Broome's argument here, then it needs to be brought out and argued for. At present, the suggestion amounts to little more than the assertion that hedonism is wrong.
