TO BE OR NOT TO BE: THAT IS THE PARADOX
This Version: August, 1995
A remarkable paradox has become apparent in recent philosophical discussions of population policy. This paradox follows upon the intuitively plausible claims that (a) there is a moral duty not to deliberately produce a potentially miserable child, and (b) there is no comparable duty to produce a potentially happy child. Following the evolving convention, we will call this "the asymmetrical position."
The paradox arises as arguments in support of (a) are found to refute (b) and to support the contention that there is a moral duty to produce a potentially happy child. An affirmation of both the duties — the duty not to produce a miserable child, and the duty to produce a happy child — is called "the symmetrical position." The difficulty with the symmetrical position is that "the duty to produce the happy child" seems intuitively odd, and worse still bears troublesome and counter-intuitive implications.
My task in this paper will be to explicate the paradox and to identify its elements; then, by locating the operative factors in the paradox, suggest a resolution which supports the asymmetrical view. The key to this resolution, which I call "the personalization rule," states that happiness is morally significant as it applies to persons, while misery is morally significant per se. "The Personalization Rule" will be recommended by (a) its apparently straightforward capacity to solve the paradox, (b) the apparent unavailability of any rival explanation, and (c) some indications of independent support from the analysis of moral language and institutions, and reflections on moral experience. Because deep investigation of these "independent indications" would radically shift the grounds of this essay from metaethics to linguistic analysis and phenomenology, I will let this "suggestion" suffice, to be addressed, I would hope, at a later time. This essay will then close with some clarification and qualifications of my conclusion and a defense against some anticipated criticism.
Actual, Eventual and Possible Persons -- A Terminological Note: The task of the reader will be greatly facilitated if it is kept in mind that by "actual person," I simply mean a person now alive. A "possible person" is an individual whose future existence or non-existence is the result of a deliberate "procreative choice" (rationally calculated in light of the presumed net-utility to these persons of being born). "Eventual Persons" are the class of indeterminate persons whose future existence may be assumed regardless of a given procreative decision. For simplicity, I will occasionally employ the terms "actuals," "eventuals," and "possibles."
A noteworthy moral difference follows upon
the distinction between "possibles" and "eventuals." For "possibles,"
in this stipulated sense, eventual existence as persons
follows-upon a calculation of utility prospects. On the other
hand, acts by the living in behalf of future persons who, it is
presumed, will exist in any case, are acts for the sake of
eventual persons. In the former case, utility determines the
eventual existence of persons; in the latter, the assured
existence of persons entails obligations to make utility
provision. In short, the moral order is reversed. This
distinction, we shall see, is crucial for my argument.
"Shall we have a child, or shall we not?" Consider this question as it is faced by two hypothetical couples:
In the first case, both prospective parents know that they are genetic carriers of Tay-Sachs disease, and thus that there is a high probability that their child will be born with this unspeakably horrible condition, whereupon the child will endure a few brief years of unremitting pain and loss of sensory and muscular function, yet retain sufficient mental awareness to contemplate his dreadful and declining condition. To carry the case further: the wife does in fact conceive, and it is later determined by amniocentesis that the fetus has the disease. We then have two questions: Should they have decided to conceive in the first place? Faced with this dreadful news, should the woman nonetheless carry the pregnancy to full term? To most moral philosophers, the clear and unequivocal answer to both of these questions would be "no!" Presumably, conventional moral opinion would support this judgment.
The second case presents a somewhat more cheerful prospect. The couple fully expects that the potential (but as yet unconceived) child will lead a very happy life. And yet the couple, and their two present children, greet the prospect of another child with collective indifference or even with regret. Having the child will cause a net utility change of zero among the present family, and perhaps even a slight drop in utility. (For simplicity, we assume no utility changes beyond the family). What is the morally proper choice in this case of "the reluctant parents?"
Concerning the second case, there seems to be some controversy among moral philosophers, especially within the ranks of the utilitarians. We will have much to say about the philosophical disputes presently. But what do our immediate intuitions and common social practice tell us? They tell us, generally, that this second case presents no great moral difficulty. Most would say, "Even if the child would be happy, don't go through with it if you don't feel like it. After all, who gets hurt? Certainly not the would-be child, for, if you don't have him 'he' cannot be said to 'get hurt' or to 'lose out'. In this second case, the choice to procreate is generally regarded as permissible but is not morally required.
A conjunction of these conventional responses to the two cases denotes what I will call (following current practice) "the asymmetrical view" regarding the morality of procreative choice. To recapitulate, this view contends that in contemplating parenthood, one is morally required not to conceive a child who one knows will be miserable. However, the choice to have a happy child is morally neutral. (The effects of the choice upon others besides the potential person are, for the purposes of this analysis, to be disregarded.) For the sake of brevity, I will speak of "the duty not to conceive" (or bear) potentially miserable children), and "the duty to conceive" (or bear) potentially happy children. (Alternatively, these positions are referred to as "the right (of the potentially miserable) not to be born," and "the right (of the potentially happy) to be born").2 Thus, the "asymmetrical position" is that while there is a duty not to conceive potential persons who would be miserable, there is no corresponding moral duty to conceive potential persons with good prospects. Leading proponents of this view include Jan Narveson, Joel Feinberg, and Mary Warren.3 In contradistinction, the "symmetrical view" contends that anyone who holds that there is a duty not to conceive potentially miserable persons is morally required to affirm a corresponding duty to conceive potentially happy persons. The proponents of this view include L. W. Sumner, R. I. Sikora, R. M. Hare and James Sterba.4 (The symmetrical view is apparently implied by classical or "total" utilitarianism.)
Treated separately and casually, the two cases do not appear to provoke deep moral reflection or to involve basic principles. Surely the first couple should take extraordinary means not to bear the unfortunate child. Moreover, it would appear that the tastes and inclinations of the second couple should dictate their choice, and that their choice should not be burdened with an overwhelming sense of duty to procreate. But when these cases are placed side by side and analyzed with care, a deep and disquieting paradox emerges. In what follows, I shall examine this paradox and the attempt by some defenders of asymmetry to resolve it. These attempts have been, at best, incomplete and unconvincing and, at worst, fallacious or irrelevant. For all that, I will offer yet another argument for asymmetry.
The paradox arises as the asymmetrist attempts, in turn, to account for and defend (using an alternative terminology) "the right not to be born" and to deny "the right to be born." Feinberg thus argues for the right not to be born:
There may very well be future interests of the child whose eventual fulfillment has already been blocked [before his birth], so that if the child is allowed to be born, his rights to the protection of those interests will have already been violated, and that fact seems to me to be a very good reason for not permitting it to be born. Moreover, it seems a reason for saying that non-birth is something we now owe it, that it is something that can now be claimed on its behalf as its due. . . .
. . . If, before the child has been born, we know that the conditions for the fulfillment of his most basic interests have already been destroyed, and we permit him nevertheless to be born, we become a party to the violation of his rights...
"In such circumstances," Feinberg concludes, "a proxy for the fetus might plausibly claim on its behalf, a right not to be born."5
Narveson focuses upon the potential for suffering: "If it is our duty to prevent suffering and relieve it, it is also our duty not to bring children into the world if we know that they would suffer or that we would inflict suffering upon them."6
A prominent argument against the putative duty to produce a happy child (alternatively, a "right to be born"), employs the notion that if this supposed "duty" is "violated," there is no victim or "deprived" beneficiary. But, the argument continues, if there is no "claimant" of the "duty" (or "bearer" of the "right"), then there is no "duty" (or "right") at all. The point was succinctly made by Narveson in private correspondence with the author:
There is nobody who suffers from not being born. Nor, more importantly, is there anybody who fails to realize the "benefit" of being born, even if being born would have been a benefit . . . . There just isn't any person there whose lack of well-being need bother me. There is no "he" to whom existence is a possible but, in the event, unrealized benefit, leaving "him" less well off than he might have been.7
Note also, that according to the foregoing formulation, if the supposed "right" to be born is "honored," even so there is, strictly speaking, no beneficiary, for as Narveson writes:
The child cannot be happier as a result of being born, since we would then have a relative term lacking one relatum. The child's happiness has not been increased, in any intelligible sense, as a result of his being born; and [if we assume that] nobody else's has either, directly, there is no moral reason for bringing him into existence.8
We shall later call this the "no reference argument."
To recapitulate: the asymmetrist view holds
that (a) it is morally wrong to deliberately bring miserable
children into the world, since doing so would create unhappiness
that might voluntarily be avoided. However, (b) it is not
morally required (though it is permissible) to produce children
who, it is known beforehand, will be happy, since, if a choice
is made not to reproduce, no one will be deprived by this
decision -- precisely because that decision was made.
When first encountered, the asymmetry position seems quite plausible. However, as Sterba and Sprigge skillfully point out, the asymmetrist's initial plausibility is accomplished by stating only the favorable half of the procreative choice case. But when the other side of the coin is examined, we find complements of the asymmetrist argument that threaten to undermine this superficially sensible position. Worse yet, the asymmetrist appears to be hard-pressed to supply reasons for not accepting these troublesome complements. Sterba's presentation of the problem is especially elegant. First, he summarizes the "favorable half" of the symmetrist's argument as follows:
If I bring into existence a person who lacks a reasonable opportunity to lead a good life, there will be a person who can reproach me that I did not prevent his leading an unfortunate existence. But if I do not bring into existence a person who would have a reasonable opportunity to lead a good life, there will be no person who can reproach me for preventing his leading a fortunate existence. Hence, only the person who lacks a reasonable opportunity to lead a good life can claim a right not to be born.
But then he presents the complements:
Notice that, if I do not bring into existence a person who would lack a reasonable opportunity to lead a good life, there will be no person who can thank me for preventing his leading an unfortunate existence. And, if I do bring into existence a person who had a reasonable opportunity to lead a good life, there will be a person who can thank me for not preventing his leading a fortunate existence. Thus whatever failure of reference there is, it occurs in both cases, and therefore, cannot be the basis for an asymmetry between them.9
Next, Sterba directs our attention to Narveson's "no reference argument," which assumes:
. . . that a person's life cannot be compared with his nonexistence unless the person already exists. This means that, if one allows a fetus to develop into a person who has a reasonable opportunity to lead a good life, one does not make that person better off than if he never existed. And it also means that if one allows a fetus to develop into a person who lacks a reasonable opportunity to lead a good life one does not make that person worse off than if he never existed. But what then justifies a right not to be born in the latter case? According to the argument, it is simply the fact that unless the fetus is aborted a person will come into existence who lacks a reasonable opportunity to lead a good life.
Sterba then explicates the troublesome complement:
But if this fact justifies a right not to be born, why, in the former case, would not the fact that unless the fetus is aborted a person will come into existence who has a reasonable opportunity to lead a good life suffice to justify a right to be born? Clearly no reason has been given to distinguish the cases.10
Moreover, as Sterba astutely observes, it seems that the choice not to produce an unhappy child must involve a comparison between existence and nonexistence, despite Narveson's insistence that being and not being are not comparable. For, without such a comparison between non-existence (before conception) and subsequent existence (after birth) how could one judge that it would be better if the child were not conceived? And if that judgment is allowed, why not then allow the judgment that it would be better if the potentially happy child were conceived?11
We are now able to identify the central structure of the controversy:
(a) If a potentially miserable person is deliberately conceived, there will be a victim to rightfully complain of it.
(b) If a potentially happy person is deliberately not conceived, then there will be no one to complain of lost benefits.
It follows, then, that it is morally impermissible to produce a miserable child, but not morally required to produce a happy child. However, the critics point out, by the same line of reasoning it follows:
(c) If a potentially miserable person is deliberately not conceived, then there will be no one to be grateful for the "favor" (i.e., no one will have "benefited" from the decision).
(d) If a potentially happy person is deliberately conceived, then there will be a "beneficiary" to thank us for it.
Furthermore, strict adherence to "the no reference rule" (employed in (b)) would seem to entail the troublesome conclusion that the decision not to deliberately produce a miserable child (c) is morally neutral, (i.e., not praiseworthy); and that a decision to produce a happy child is morally praiseworthy.12 These conclusions utterly undo the advantages sought by the asymmetrist position!
To the best of my knowledge, these difficulties in the asymmetrist position were first identified by Timothy Sprigge in his 1968 review of Jan Narveson's book, Morality and Utility (published the previous year, as was Narveson's influential paper, "Utilitarianism and New Generations"). And so, Narveson, Feinberg, and others defending asymmetry have had time to come up with a rejoinder. And yet, as I read their recent arguments, I am struck by how much they seem to be not responses but reiterations of earlier positions. In virtually every case, the asymmetry position can readily be converted, with suitable negation and substitution, into the troublesome complements suggested by Sprigge and Sterba. (The reader is invited to try this out as he reads the arguments of Narveson, Feinberg, Warren and others). Furthermore, I have encountered no clear or persuasive reasons not to make such extensions. Consider, for example, Narveson's statement of asymmetry (with my emphasis of the operative words):
Suppose we decide, in the end, not to bring [a new person] into existence. Then even if [he] would have been perfectly happy, still nobody misses anything; or anyway, it is only we, the people he would have had desirable effects upon, who are missing something. But there isn't any loss of anything by the contemplated party since he doesn't exist. On the other hand, if we actually do bring him into existence, then of course we must be as concerned for his welfare as anyone else's. Suppose we could foresee that his life would be miserable. Then it seems obvious that a way to avoid that would be to avoid conceiving him in the first place, and that that, accordingly, is what we should do.13
Does this formulation of asymmetry place Narveson on the slippery slope, at the bottom of which are to be found Sprigge's unwelcome complements (i.e., (c) and (d) above)? I believe that it does, as we can readily determine by restating the above passage with a suitable substitution (in brackets) of each of the italicized "operative words":
Suppose we decide, in the end, not to bring (a new person) into existence. Then even if (he) would have been perfectly [miserable], still nobody [gains] anything [i.e., by this decision]; or anyway, it is only we, the people he would have had [undesirable] effects upon, who are [gaining] something. But there isn't any [gain] of anything by the contemplated party, since he doesn't exist. On the other hand, if we actually do bring him into existence, then of course we must be as concerned for his welfare as anyone else's. Suppose we could foresee that his life would be [happy]. Then it seems obvious that a way to [cause] that would be to [conceive and bear] him in the first place, and that that, accordingly, is what we should do.
The reader is urgently reminded that the preceding passage is my alteration of Narveson's text into a formulation of the complement of Narveson's statement of asymmetry. Narveson, of course, would reject this alteration. My point is that there is no apparent means at his disposal to do so.14
In the few occasions where the arguments might resist such conversion, Narveson, Feinberg, and others hint at the idea that happiness/misery and benefit/harm are not, morally speaking, symmetrical opposites. However, perhaps because such avenues of analysis threaten to cause deep theoretical difficulties (especially to Narveson), these suggestions are not given careful consideration.15 That oversight, I believe, is a fundamental mistake, for, as I hope to demonstrate presently, it is precisely along these lines that the paradox of asymmetry can best be resolved, and in a manner which preserves our intuitive disposition to accept the asymmetrical view that there is (a) a "right not to be conceived" and (b) no "right to be conceived."
However, before we endeavor to reconstruct a
defense of asymmetry, we should devote some attention to the
claim that the symmetrical view is the acceptable resolution of
There are, in fact, two possible symmetrist positions. However, the first is so morally ghastly that it can be quickly disposed of. The two positions are:
a. The right to be born and the right not to be born -- alternatively, the duty to/not to conceive -- are morally neutral. (More accurately, there is no "right"/"duty" in either case.)
The first interpretation would condone the casual production of miserable children (as in the Tay-Sachs case). For this reason alone, it can be dismissed. The second symmetrical position holds both that the potentially miserable child has a right not to be conceived, and that the potentially happy child has a moral claim to be brought into existence. Both sides of the dispute accept the right of the potentially miserable person not to be born. The dispute centers upon the alleged "right to be conceived" of the potentially happy child (or, alternatively, the duty to conceive a potentially happy child).
James Sterba, as we have noted, has expertly explicated the "symmetrical complements" of "the right not to be born." He does so in defense of the symmetrical position - the claim that there is a duty to produce potentially happy children. His lucid and artfully argued paper thus presents us with a convenient statement and definition of the symmetrist position, and we will devote most of our attention to it.16
Among the more common criticisms of the "right to be born" is the charge that it lends itself to strongly pro-natalist policies that will adversely affect the living conditions in the future. Thus some familiar objections to "total utilitarianism" are said to apply against the symmetrist view. For if a potential person with on-balance positive prospects has a claim to be brought into existence, then are we not morally required not to "deny" him that opportunity -- providing, that is, the "gain" of his net happiness is not offset by a consequent net loss among others? In short, might not such a policy imply a maximization of total utility at the cost of a considerable loss to the average prospects of each?17
Sterba acknowledges this implication and seeks to avoid it by proposing a middle course -- a "compromise between the policies of average and total utility."18 This is his proposal:
Given that the welfare rights of future generations require existing generations to make provision for the basic needs of future generations, existing generations would have to evaluate their ability to provide both for their own basic needs and for the basic needs of future generations. Since existing generations by bringing persons into existence would be determining the membership of future generations, they would have to evaluate whether they are able to provide for that membership. And if existing generations discover that, were population to increase beyond a certain point, they would lack sufficient resources to make the necessary provision for each person's basic needs, then it would be incumbent upon them to restrict the membership of future generations so as not exceed their ability to provide for each person's basic needs. Thus, if the rights of future generations are respected, the membership of future generations would never increase beyond the ability of existing generations to make the necessary provision for the basic needs of future generations.19
But just what are these conditions of "necessary provision for the basic needs of future generations?" For if the "basic needs" are not met due to resource depletion by overpopulation, then surely this will mean that we will have recklessly overshot the peak of the "total utility" calculations. (What, in fact, is a "basic need"? Life sustaining nutrients? A forty-hour work week with paid vacations? How do we answer this, if not through utility calculations?) In short, if Sterba proposes to regulate current procreative policy by linking it with the perceived capacity to meet "basic needs" in the future, it is difficult to distinguish his policy from that of total utility, except, perhaps, to suggest that total utility might call for even a smaller population than that indicated by Sterba's position. (Billions of people now alive might be said to have their "basic needs" met, and yet to lead lives of unacceptably low utility). At the very least, Sterba's policy seems to demand far too much of "actuals" and "eventuals" for the sake of "mere possibles." In short, Sterba is saying that potential persons should be produced so long as the gain that these potentials provide, through their birth, is not offset by a greater loss in the capacity to provide for the basic needs of others. If by "basic" Sterba means "necessary for life-sustenance," I confess that I find little difference here between his principle and the "total utility principle") -- with its repugnant conclusion that we are obligated to produce an overpopulated world of on-average slightly well-off individuals who, due to their enormous numbers, sum up an "optimum," if impersonal, total. But if by "basic" he means "productive of a fulfilled life," then surely some rather non-compelling considerations can justify a decision not to procreate -- considerations by no means complementary, to strong the moral requirements that entail a duty not to conceive an eventually miserable person. Thus, according to this latter interpretation of "basic need," we have what amounts to a qualified asymmetry.
How then are we to keep the "mere possibles" from crowding the "actuals" and "eventuals" at the table of limited resources and utility, if not by restricting their admittance? And how are we to justify this restriction except through a claim that the "actuals" (who now exist) and the "eventuals" (who will exist regardless of our personal procreative decisions) have a greater claim to these resources than those who we might bring about by deliberate choice (i.e.,the "possibles")? However, once this much is admitted, then the issue has been conceded to the asymmetrist. In short, the symmetrist position may lead inexorably to "the repugnant conclusion" of total utility -- an over-crowded world of individuals with barely tolerable lives. Or if not this conclusion, then another that is not recognizably preferable. Attempts to avoid this conclusion apparently lead to an abandonment of the symmetry.
While Sterba, Sumner and other philosophically sophisticated symmetrists have spared us, one often hears that tiresome taunt of the pro-natalists: "What if your parents had used contraception or had an abortion when your 'turn' arrived?" Well, it happens that in my case they did not, and I will admit that I am retrospectively grateful for it. But, of course, that fortunate circumstance is a precondition of the question. It will never be asked of the sisters that I never had -- nor of an infinitude of other unactualized potentials. And if I had never been conceived, that "deprivation" would have "mattered" no more nor less to "me" then to that multitude of other "unfulfilled possibles," because, of course, there would have been no "me" for it to matter to. But all this is tediously familiar by now to those with even a casual interest in the philosophical issue of procreation policy. I raise it just once more to make an essential point, which even now seems to be overlooked by many capable philosophers; namely that a crossing of the line from prior possibility to subsequent actuality produces a point of perspective and reflection, a moral reference, a locus of desire, interest and motivation, which makes a literally incalculable moral difference. Thus the fallacy of the "retrospective gratitude argument" above (i.e., "What if your parents. . ."). Furthermore, the emergence of personhood from prior nothingness is also fundamentally different from the extinction of personhood at death. Thus the "merely possible" person "looses" nothing by never being conceived, while a person, at death, "looses" everything. Birth and death are not simply to be compared as boundaries between being and non-being, distinguished merely by the reversal of time's arrow. Thus the asymmetrist's "no reference argument" is not effectively refuted by drawing repugnant analogies between the denial of the "right to life" of a potential person (through contraception and abortion) and an actual person (through homicide).20 But we are now at the frontiers of some difficult logical, even ontological, issues, the pursuit of which might lead us hopelessly astray from the developing argument of this paper.21
And so, we return, at length, to the central insight of the asymmetrical position -- often reiterated but never explicitly justified by the asymmetrist (perhaps because it is a simple, unanalyzable, intuitive given), yet never refuted with satisfaction by the symmetrist. This is the simple point that a decision not to procreate deprives no one (except, irrelevantly to this consideration, the potential parents and would-be associates). For what sense could be made of claiming that it does? Are we to presume some sort of pool of subsistent and permanently deprived and frustrated might-have-beens? (A moment's reflection will indicate that the population of this Meinongian realm will be virtually infinite). The suggestion seems unworthy of a moment's pity or moral anguish. Accordingly, in our ordinary moral judgments, we do not condemn priests or other celibates for "depriving" potential children of existence. Contraception is not deemed comparable to infanticide or murder (except by a few fanatics), and an additional child deliberately not had is not generally regarded as a child "lost."
Much more might be said against the symmetrist view, but, if I pursue that task, insufficient space will remain to carry out the intended purpose of this paper. So, in behalf of that original purpose, we shall hereafter assume that the symmetrist position bears unacceptable and even absurd implications. Given a choice between the contending positions, we provisionally accept the asymmetrical view, pending a resolution of the paradox therein. The remainder of this paper will be devoted to an attempt to resolve the paradox and thus to preserve the asymmetrist position.
Let us assume, then, that we will not find a resolution of the paradox through a simple acceptance of the symmetrist's position -- of an affirmation of the duty to conceive potentially happy persons. If we wish to preserve the intuitively attractive asymmetrical position, how can we do so without being drawn toward the unwelcome absurdities and paradoxes skillfully explicated by Sterba and Sprigge? In brief, how can we affirm a duty not to conceive potentially miserable persons without, by implication affirming a duty to conceive a potentially happy person (the symmetrist view). Alternatively, how can we deny a duty to conceive the fortunate without, at the same time, denying a duty not to conceive the miserable? That, in essence, is the paradox. I believe that the first step toward a solution is to identify the operating variables. This we accomplish by setting down the bare bones of the controversy:
1. Concerning the "right not to be born" (of potential miserable persons):
2. Concerning the "right to be born" (of potential happy persons):
The asymmetrist regards (a) as morally obligatory, (b) as morally reprehensible, and (c) and (d) as morally neutral. (Recall that the term "right" is intended broadly and metaphorically; thus, there is no need to be involved here in metaethical analyses of the concept of right). From this table, the following three pairs of variables emerge:
A. "Honored"/"Violated": These terms simply identify that there is a choice involved -- a condition (necessary, but not sufficient) of moral significance.22
B. "Beneficiary"/"Victim": Behind this distinction lies "plus and minus utility"; i.e., happiness/misery, pleasure/pain.
C. Being/Non-Being ("there is"/"there is no"): This means "eventual actualization as a person," and "permanent non actualization." (These alternatives are contingent upon the choices indicated in (A), above).
I will argue that asymmetry may be gained by treating only the second pair (beneficiary/victim) as nonsymmetrical (i.e., not polar opposites) -- at least in its application to "possible persons." Thus, in the asymmetrist view, with respect to possible persons, the avoidance of misery has lexical priority over the production of happiness. Accordingly, since the Tay-Sachs case involves a potential victim, but no potential beneficiary, it is morally significant. And, since the case of the reluctant parents involves a potential beneficiary but no potential victim, it is morally less significant--to the point, perhaps, of being morally neutral.
The resolution is appealing in its simplicity, but unless some qualifications are added, we will find that this cure is worse than the ailment -- that our solution involves more paradoxes and absurdities than it resolves. One important qualification is to be derived from the third pair, "Being"/"Non-Being". But more about that later.
According to classical utilitarian thought, pleasure and pain are normally treated as "polar opposites". (This conception, I believe, is reflected such non-philosophical thinking as economic cost-benefit analysis). Thus, goes the assumption, these opposites should, in principle, be regarded as commensurate, as represented on opposite ends of a bipolar scale, like minus and plus degrees Celsius or dates A.D. and B.C., or the credit and debit sides of a ledger. And yet, in this essay I seem willing to set aside this concept of polar opposites of utilities in order to rescue the asymmetry position.
The notion that happiness and misery are not polar opposites is appealing. Karl Popper writes:
There is, from the ethical point of view, no symmetry between suffering and happiness.... Human suffering makes a direct moral appeal, namely, the appeal for help, while there is no similar call to increase the happiness of a man who is doing well anyway. . . .instead of the greatest happiness for the greatest number, one should demand, more modestly, the least amount of avoidable suffering for all.23
Popper's point deserves careful reflection. Surely his observation that "suffering makes a direct moral appeal" strikes a responsive note in most of us. The alleviation of anyone's (undeserved and inflicted) misery, demands attention and is the business of everyone (save in special cases such as the receiving of just punishment). On the other hand, "the pursuit of happiness" is a private matter. This is a crucial point, about which I will have much to say presently.
The idea that asymmetry might be rescued through a rejection of the polarity of happiness and misery appears to be appealing to Feinberg, though one wishes he had given this notion more attention.24 Narveson, however, does not pursue this lead at all, perhaps because of the threat that it poses to his general ethical theory. And so he takes care to point out that the issue of asymmetry (regarding potential persons) "is not to be identified with the other issue of whether our duties, strictly speaking, concern only (say) the reduction of suffering and not the increase of well-being."25
But a simple rejection of polarity "rescues" asymmetry at much too great a cost. First of all, as J. N. Smart has observed, if the alleviation of pain is to count for everything, then an alien astronaut would be permitted, even required, to destroy the world instantaneously, without warning, so as to abolish all pain.26 The consequent loss of pleasure would have no bearing upon this decision. Similarly, a denial of polarity might lead to the conclusion that one should never have children. For, as Sprigge points out, "Their pains would be present on the debit side without their pleasure being there on the credit side."27 But even more commonplace difficulties follow upon the supposition that pleasure and pain are not polar opposites and thus commensurate. For, if they are not commensurate opposites, what then becomes of policy planning, cost/benefit analyses of industrial and governmental projects, intra-personal utility calculations? How then, for example, is a student to determine whether four more years of graduate school is worth the privilege of searching for an appointment in a philosophy department? Don't we, in fact, just assume that pleasures can be measured against pain? How could we do otherwise?
Very well, we can and do make plus/minus utility calculations. In fact, I will concede that we must do so. Does this concession demolish the promising notion that non-polarity might rescue asymmetry? Not quite. I have only conceded that polar opposition of pleasure and pain obtains in the utility calculation of existing persons. But we are manifestly not dealing with existing persons when we face the question of "the right to be, or not to be, born." The distinction is crucial, for if it can withstand analysis, it will bring us close to the resolution we seek. My proposal, then, is as follows: Pleasure and pain, and happiness and misery, are polar opposites when they apply to the utility calculations of and for persons who do exist or who surely will exist. (Under carefully restricted conditions, such calculations may also apply in the case of hypothetical persons, but such conditions must be treated with great circumspection; otherwise I shall lose my distinction and, with it, my argument. I will discuss this in the final section.) However, for persons whose future existence is potential and contingent upon the choice of an existing moral agent, happiness and unhappiness are not commensurable. In this latter case, the prevention of misery takes overwhelming, and perhaps absolute, precedence. But how am I to argue this point so that it will be more than an ad hoc "fix" for a threatened intuition? I do so by defending the principle that happiness is morally significant as it applies to persons.28 Misery is morally significant per se and thus the avoidance of misery claims greater moral urgency than the promotion of happiness prior to, and apart from, its localization in the life of particular persons. I will call this "the Personalization Rule."29
If we can find warrant for the Personalization Rule, then we might, with this rule, restrict the polarity principle to the utility calculations of existing persons (or in behalf of "eventuals" -- future persons who, it is assumed, will exist in any case), thus avoiding manifestly absurd and impractical implications of rejecting polarity in these areas. And yet, with this rule, we might also exempt the polarity principle from procreational decisions. If we can accept this restriction, then the "right to be born" will in fact be morally negligible -- in short, our asymmetrical view will be sustained without paradox.
To illustrate this claim, consider Mary Anne Warren's alternative expression of asymmetry: "Creating happiness for people is morally good, but creating people for happiness is morally neutral."30 Warren's maxim might be thus paraphrased and tabulated:
However, as we have found, if the maxim is carelessly combined with the polarity principle, we get this unwelcome symmetrical "completion":
Yet Warren and Narveson clearly assert, and most moral intuitions confirm, that it is morally desirable "not to make miserable people." Clearly, our qualified "personalization rule" allows us, at the same time, and as desired, to affirm (a), (b) and (c) and to deny (d).
And so, of course, our remaining question is simply this: Is there reason to accept the Personalization Rule? Lest we be found guilty of ad hoc reasoning, we must find warrant for the personalization rule that is independent of its apparent capacity to resolve the paradox of asymmetry.
As a courtesy to the reader, I should like to pause for a brief space to recapitulate the concepts, review the argument and anticipate the remaining course of the paper. The essential aim of this essay is to explicate the symmetry paradox and to present a defense of the asymmetrical position that will serve to resolve the paradox. The "asymmetrical position" is essentially as follows:
a) Potentially miserable persons have a "right" not to be born.
b) And potentially happy persons have no converse right to be brought into existence, and thus that there is no entailed claim upon existing couples to conceive and bear happy persons.
My first move was to introduce and explicate "the principle of polarity" or "polar opposition" by which I mean the assumption of many utilitarians that pleasure and pain (or happiness and misery) are opposite values on a uni-dimensional scale, and thus commensurate. Thus, for instance, it is possible to assess whether or not the willing endurance of certain pains is adequately compensated for by the resulting "payoff" of eventual pleasures, etc. An unqualified acceptance of the polarity principle, we found, entails that if the potentially miserable have a right not to be born, then the potentially happy have a right to be born, thus refuting the asymmetry position and confirming symmetry. And yet, if polarity is denied, the foundation of much routine moral deliberation and policy planning is undermined, and such deliberation is made incoherent, unavailing and pointless. Thus asymmetry can be rescued only if the polarity principle is somehow qualified.
But might we not, following Narveson and Parfit, insist that utility be "actual-person regarding" i.e., that utility calculations apply only to existing persons? If so, then there is no moral obligation to produce possible happy persons. Unfortunately, as we have noted, this qualification accomplishes too much, for it also follows that there is likewise no moral requirement to refrain from producing possible miserable persons.31 To avoid this result, I have suggested first that the polarity principle applies within and among existing and identifiable persons, or persons who surely will exist; and secondly that among potential persons the avoidance of unhappiness takes primary, and perhaps exclusive, priority. This proposal follows from a basic and widespread intuition that while happiness is normally promoted for the sake of identifiable persons, misery, per se, demands our moral attention and response. I have called this intuition "the Personalization Rule."
The Personalization Rule offers a ready and intuitively appealing account of the two hypothetical cases which opened this paper. In both the case of the potentially happy child of the reluctant couple and the potentially miserable Tay-Sachs child, there is no present referent -- i.e., the potential utilities are not personalized. But our rule states that happiness needs a referent (a person) if its promotion is to be morally praiseworthy, but that misery needs no referent to require us to prevent it This accounts for the asymmetry of the cases. The potential happiness of the first child is insufficient to require the actual couple to endure even a minor loss of net happiness. On the other hand, the potential (though unpersonalized) misery of the Tay-Sachs child weighs conclusively against allowing him to be born.
Consider next the alien astronaut case,
mentioned earlier. Following J. N. Smart's suggestion, we recall
that a strict "negative utilitarianism," which gave all moral
credit for the alleviation of pain and no moral credit for the
promotion of pleasure, would permit, even require, the instant
destruction of all worlds containing any misery. But in our
personalization rule, we have the warrant to stay the hand of
the destroying alien. For now the positive utilities of actual
persons here on earth count for preservation. (Are there enough
positive utilities to morally offset the misery in the world?
Perhaps the alien requires further argument. But that,
thankfully, is not our task at the moment).
Our carefully qualified "Personalization Rule" apparently does the job we want it to do. For according to the rule, potential happiness counts for little while potential misery is of overwhelming moral significance. Thus we have the desired conclusion that there is a duty to prevent potentially miserable children, but no converse duty to produce potentially happy children. Moreover, we account for the asymmetry without paying the unacceptable price of rejecting the validity of utility calculations among existing persons.
But is this personalization rule a justification of asymmetry? Not if it serves only to prop up our intuitions concerning asymmetry. For in that case, we have merely concocted an ad hoc device and not an argument. And so, some independent warrant is required. We must, that is, find in various other areas of moral reflection and experience the evidence for believing, first, that happiness is most appropriately promoted for the sake of persons, while misery, as such, calls for relief, regardless of the personal identity or even the present existence of the victim. (This latter, of course, is a prima facie claim, subject to overriding considerations such as just punishment, the cost and hazards of relief efforts, etc.).
In the remainder of this paper I will attempt to indicate that the experience of happiness and misery are such as to lead us (a) to affirm polarity within and among actual persons, but (b) to deny polarity as it applies to potential persons. (And most significantly, persons who are "potential" in the sense that our calculation of their utility prospects has bearing on our decision to bring, or not to bring, these potential persons into existence). I say that I wish to "indicate" this. The word was carefully selected. I do not presume to "demonstrate" or "prove" this "personalization rule." That is a task beyond my present ambition or competence, and surely beyond the scope or available space of this essay. Thus I shall not attempt, as may eventually be required, a phenomenological analysis of pleasure and pain nor a conceptual analysis of motivation. But while I will not attempt to explain why it is true, I might attempt to indicate that it is true. I shall do so on the basis of this assumption: If the personalization rule (i.e., specifying "polarity" for actual persons, and "negative utility priority" for potential persons) is, in fact, a reflection of normal human experience and judgment of pleasure/pain and happiness/misery, then that fact should be widely manifested in conventional language, moral experience and judgment, and institutional practice -- i.e., in the manner in which ordinary persons in ordinary experience, with ordinary judgments, regard happiness and misery. I believe that a citation of a few cases will so indicate.32
a). Consider first the ancient, yet familiar, "hedonic paradox" which asserts that the person who seeks happiness directly is less likely to "find" it than another who cares not for "happiness" per se, but rather seeks experiences, activities, companionship, etc. "for their own sakes" -- things which, as it happens, also make him happy.33 Thus it would appear that happiness is not a thing apart to be "grasped"; rather, as Aristotle and Butler so astutely observed, happiness is a condition which comes upon the individual in the course of the successful and fortuitous pursuit of his life's activities and goals. Thus it seems that happiness can not sensibly be regarded apart from the person who possesses it; in fact, this very locution is faulty, for it wrongly suggests that happiness is some "thing" which someone "has." (Of course, strictly speaking, neither happiness nor misery exist apart from a being who is happy or miserable -- i.e., a subject of these conditions is necessary and presupposed. But when I deny that happiness can be sensibly abstracted from a person, I am making a stronger claim which, as we shall see, does not apply to misery).
Thus we don't normally say "I want to give John some happiness." Rather we say, "I'd like to do something to make John happy," or even better, "give him something he will enjoy." The subject of the happiness is appropriately prominent. But note that we do commonly remark, "I wish that I could do something about Jim's unhappiness" (as if the unhappiness "itself" were the focus of attention and of the difficulty). And how are we to make John happy? Well, that depends upon what sort of a person John is. Unless we know this, we are at a loss. First there is the person -- his tastes, interests, goals, etc. Given this, certain things may or may not happen to him to make him happy.
In contradistinction, there seems to be no "anhedonic paradox." When one is in misery, the misery "itself" is the focus of attention, and generally the best strategy for relief is to identify the cause straightaway and, if possible, to attack the cause directly.
b) Common linguistic usage tends to attach happiness to persons, while misery seems more "detachable" and objective. For example, we might say "it is good for Jones that he is so happy," but not "a lot of happiness has come to Jones." We would say, however, that Smith has seen a lot of misery in his day." And what of those who cause happiness and misery? Do we not say, for instance, that "Stalin caused a great deal of misery," and in so doing, we feel that this was contemptible in itself, without feeling compelled to ask "misery for whom?" On the other hand, the expression "Groucho Marx has caused a great deal of happiness" seems much less appropriate than "Groucho Marx has made many people happy."
c) The traditional "right" to "the pursuit of happiness" is fundamentally a right of a person. In other words, it is the responsibility of the mature, autonomous agent to seek after his own good, without meddling "do gooders" and "busybodies" imposing their "goods" upon him. (Of course, children, whose personalities are "in the making" are appropriately "guided" in their choices of goods). The right of the person to pursue his own happiness appears to be a requisite condition for personal integrity and dignity; i.e., given that there is a person, then his good is his business. The person precedes the benefit -- i.e., the happiness pursued.
In contrast, misery is much less a private matter. If a person is needlessly and innocently suffering, a claim for assistance is upon anyone in a position to alleviate that suffering. The suffering "itself", apart from the victim, demands attention and moral concern. The responsibility to ease the suffering is primary, while the personality of the sufferer is secondary.34
d) We have been examining the claim that plus and minus utilities may be commensurable when applied to persons who exist (or will exist), but not when applied to persons who might exist. But now a complication arises. The weighing of plus and minus utilities is affected by whether the calculation is intra- personal or inter-personal. In the case of intra-personal calculations, negative utility seems to count the least. Thus the agent is entitled to assume higher risks and endure greater pains for the sake of his own personal payoffs. But he is not entitled to assign comparable risks and costs to others -- even for their sakes. (Thus, while a soldier may volunteer for a dangerous patrol, he does not have the same moral right to assign the mission to another.). But there is more. Justice requires that risks and costs be assigned and shared by the community that stands to gain collectively by these disutilities. (Thus it is unjust for "outsiders" downwind of an industrial complex to suffer, without compensation or benefit, the result of air pollution). Furthermore, risks and costs are not appropriately assumed by those not competent to make an informed, deliberate choice. Children and incompetents deserve protection from exploitation, and all have a right not to be defrauded. In short, negative utility counts least in intra-personal utility calculations within primary groups, and progressively more as the individuals involved are either (1) less identifiable, (2) less capacitated, (3) less informed, or (4) less mature. Is it not, then, plausible that the limiting case in this progression concerns persons whose very future existence depends upon a present choice whether or not to bring them into existence? And in such a case, would not negative utility be decisive and positive utility negligible in the utility calculation?
(e) Further support for the Personalization Rule might be found in our legal traditions. Thus the law requires one to refrain from causing harm -- to anyone in general. (Thus harm is "abstracted" from identifiable persons). In contrast, the law generally does not require one to benefit others. The exception to this latter rule is instructive: we are, it seems, legally required to benefit those particular, identifiable persons with whom we have reciprocally agreed to binding contracts.
The Personalization Rule will, I believe,
sufficiently qualify and restrict the principle of polarity to
allow a resolution of the paradox of asymmetry. That, at least,
is the essential claim in this paper. In this section I have
attempted to gather a few straws in support of the
personalization rule, but that rule remains the weakest
component of my argument. How is that rule to be explicated,
demonstrated and proven, or, conversely, refuted? In that
question, and beyond, lie some deep and crucial issues of moral
philosophy and philosophical anthropology. If this paper
provokes useful investigation along these lines, it will have
accomplished it purpose.
As I have developed and discussed the ideas expressed in this paper, a few objections have emerged. While I find none of them to be fatal to my defense of asymmetry by means of the Personalization rule, I believe that many of these objections and comments offer opportunity and grounds for clarification and qualification of this account.
a) The rejection of polarity in the case of future persons implies that no children should ever be allowed to be born. This stark conclusion seems to follow rather directly from an acknowledgment that all children, even the happiest, have some moments of unhappiness in their lives. But if this is so, and if the "potential" happiness of a merely possible person counts for nothing, then the residue of potential unhappiness, however, minuscule, suffices to disqualify the occasionally not-happy child (in fact, any child) from being conceived or born.
The conclusion is evaded by denying the embedded premise that "the potential happiness of a merely possible person counts for nothing." But how is this presumption to be avoided? Simply by identifying an ambiguity and keeping our logical sequences in correct order.
The ambiguity resides in the terms "potential happiness" and "potential unhappiness" as they appear above. Do we refer there to the hedonic quality of various moments in the anticipated life, or the net utility summation of the life-as-a- whole? Obviously, few if any lives, however miserable, are wholly without moments of fleeting pleasure; neither are happy lives free of occasional moments of pain. Just as obviously, a decision not to have a miserable child, or conversely to bear a happy child, is properly made on the assessment of the "life-on- balance."
All this is simple common sense. "But," the critic may continue, "that just makes it all the worse for your theory. For according to your account, potential happiness just doesn't count against potential misery -- however bright the prospects for a preponderantly happy life." In response, we briefly turn to logic.
The asymmetrical view, we will recall, claims first that a potentially miserable person (M) has a right not to be born (RN), i.e.:
Furthermore, the potentially happy person (H) has no right to be born (~B). I.e.
Obviously, due to an excess of uncoordinated terms, we can get nowhere by means of formal inference. But that is not my purpose here. Instead, I wish only to point out that before the fate of any given "x" (RN or ~B) can be decided, it must first be determined which predicate applies to "x". Put simply, the "H" and the "M" come before the "—>". In other words, the utility assessment ("net-happy" or "net- miserable") is hypothetical -- which means, of course, that in making such an assessment, one is not (yet) committed to make a choice regarding its actualization. Furthermore, it is the hypothetical child, entire, regarded as "timelessly actual," whose net utility prospects are being evaluated. We wish to understand, that is, whether the potential child, not his individual hedonic states, should be actualized. Once this is determined, the next step is to draw inferences from these assessed prospects of the (integrated and summed) potential child.
In the preliminary stage of net utility evaluation, the positive and negative factors are eminently commensurable in that (a) they are intra-personal,35 and (b) the "mere potentiality" of the individual is not at issue -- it is properly "bracketed," to be brought to bear at the next stage of analysis. At that next stage, the "brackets" are dropped as one asks, "What follows from this (prior) assessment that this potential child's prospects are net-negative (or net-positive)?" According to the asymmetrist, a net-minus assessment entails a duty not to procreate, while a net-plus assessment permits, but does not require, procreation.
b) The same objection can be answered by a slightly different approach. The objection, we will recall, follows from the Personalization Rule and the observation that all future persons will surely experience some unhappiness. And if, according to the personalization rule, the anticipation of future happiness provides no requirement for procreation, and an anticipation of future misery (however negligible) entails a positive requirement not to procreate, it then follows that we are morally required to have no more children, and thus to bring the human race to an end.
The objection is based upon the assumption that futurity itself implies non-personalization. If we can show that this is not the case, then we might answer the objections by denying the premise that "the anticipation of future happiness provides no requirement for procreation."
Why might one conclude that future happiness and misery are not "personalized"? One argument might be that future individuals are indefinite and indeterminate.36 That is to say, while we might be prepared to describe future persons by some of their shared general qualities, we are unable to identify them as discrete, individual persons.
This point can be readily countered with the observation that among these general qualities that we can distinguish is that the admittedly indefinite future individuals will be persons., and thus, as required, their net utility (misery and happiness) will merit moral consideration.
But didn't I earlier define "personalization" in terms of the "localization" of happiness and misery "in the lives of particular persons"? Indeed I did. But note that future persons can be particular, though "indefinite" and "indeterminate." The latter terms describe epistemological limits. The term "particular person" identifies a selected "level of concreteness," appropriate to moral consideration.
Might not this rejoinder accomplish too much? Doesn't "personalization" now apply to all persons, actual and future, with a resulting reinstatement of unqualified polarity -- and all the attending difficulties thereof? I think not, for it is not futurity, per se that has bearing on the personalization rule Here I would reiterate the familiar distinction that appeared at the opening of this paper; namely the distinction between "eventually future persons" and "possible future persons."37 The Personalization Rule applies to the latter category and not to the former. Eventual persons, we will recall, are those who might be assumed to exist, regardless of our deliberate decisions.38 In contradistinction, possible persons are those whose existence is contingent upon the deliberate choice of those making a procreative decision. (Thus, if abortion is a "live option," fetuses are "possibles"). Accordingly, eventual persons may be regarded as "timelessly actual" - i.e., since their eventual actuality is assured, our deliberative actions will have both positive and negative effects upon their "eventual, actual, personal lives. Time alone distinguishes "actuals" from "eventuals," and polarity considerations thus apply.
This point might be clarified if we recall that, by definition, the "symmetry paradox" involves "possible persons" exclusively. (Again, "possibles" are defined as individuals whose future existence is contingent upon a deliberate procreative choice). According to the Personalization Rule, prior to a procreative decision and a commitment thereto (i.e., while the potential person is still a "possible"), potential misery matters and potential happiness does not. Why? Because until the decision and commitment is made to procreate, there is no person (either actual or eventual), to which the atemporal predicates "is happy" and "is miserable" apply. This is the sense of "potentiality" to which the Personalization Rule" applies - i.e., the sense in which a decision not to bring a potentially happy person into being deprives no existing, or to-be-existing, being of happiness, while a decision to bring an unhappy person into being creates unhappiness per se. With a decision to procreate (either by conceiving or, later, by electing not to abort) the "possible person" becomes an eventual person and, with time, an actual person. Time itself, we have noted, is the only important distinction between eventual and actual persons.39
c) These considerations lead us, at last, to
the resolution of a difficulty, suggested by Derek Parfait, of
the case of the woman who is able to choose between having a
mildly handicapped (but "net-happy") child this year, or another
much happier child the next.40 Our critics might
charge that, since these children are both indeterminate "possibles,"
their happiness counts for nothing, and that the choice is
indifferent. I reply that the woman has no obligation to produce
either. However, once she decides to bear a child, the child to
be chosen will be, not a "possible" but an "eventual" and then
an "actual." The woman, having chosen to conceive, moves on to
the (logically) forced choice of one "eventual" from a roster of
"possibles." This "impending actuality," brought about by the
decision to have a child, imports an urgency to "make the better
choice" that is not to be found when the choice is whether or
not to have a child at all; when, that is to say, the choice is
still one logical step "behind" the urgent choice between
children with differing prospects. The operative rule, once
again, is that "the Personalization Rule" applies not to "possibles"
but to "eventuals" and "actuals".
Population policy raises a number of troublesome moral issues that simply do not arise when procreational decisions are not being taken into account. For instance, practical choices between total and average utility, and the puzzles and paradoxes associated therewith, arise in this context. So too with the problem of asymmetry. Furthermore, like so many philosophical problems, the asymmetry issue displays paradoxes on the surface that might eventually lead to significant resolution in depth. That has been the hope and expectation of this essay. We have found herein that there is no simple resolution to the asymmetry paradox. But symmetry too entails unacceptable consequences. On the other hand the complement of one half of the symmetrical position ("the right not to be born") seems to be a right of the potentially happy to be born, while the complement of the other half (the denial of the alleged "right of the happy to be born") leads to the repugnant conclusion that there is nothing morally wrong in willfully procreating miserable persons. The paradox is deep, and so too must be its resolution. It was not enough, we have found, to deny the polarity of happiness and misery, for that denial also yielded unacceptable implications. So we looked deeper still. Polarity, I suggested, had to be qualified by the Personalization Rule -- the principle that "happiness is morally significant as it applies to persons," and that "misery is morally significant per se" At that point, with a few "indications" of supporting evidence and a brief reply to criticisms, we left the matter, with the hope that the Personalization Rule" might be examined through other modes of philosophical investigation such as conceptual analysis and phenomenological research.
It may be charged that this approach to the problem employs ad hoc reasoning. The charge is not without merit. Indeed, I not only admit to it, but I would also defend, this approach. The Personalization Rule has, in fact, been brought forth to do the job of resolving the paradox of asymmetry. But recall the essential guiding question of this investigation: "Just what would it take to escape the paradox?" I approached this question by attempting to isolate the elements of the paradox and thus to identify the only possible avenue(s) of escape. If this analysis has been sufficiently rigorous and complete, then I may hope to have identified, in the Personalization Rule, the only avenue of escape. To put it briefly, I chose this Rule, not because it was a convenient and personally satisfying way out, but because, for reasons carefully derived and displayed, it seems to be the only way out. As a preliminary step, I believe that this is legitimate, provided this method is not presumed to settle the matter. And so I then proceeded to enlist support from a review of a few moral and legal intuitions and invited further investigation. Perhaps such investigation will not bear out my proposal. But win or lose, I hope that I have furthered inquiry by at least suggesting an answer to the paradox of asymmetry and inviting further examination of both this troublesome puzzle and this proposal of a resolution.
NOTES AND REFERENCES