Superlongevity and Utilitarianism
Mark Walker
McMaster University
Penultimate Draft: Forthcoming in Australasian Journal of Philosophy
Abstract
Peter Singer has argued that there are good utilitarian reasons for rejecting the prospect of superlongevity: developing technology to double (or more) the average human lifespan. I argue against Singer’s view on two fronts. First, empirical research on happiness indicates that the later years of life are (on average) the happiest, and there is no reason to suppose that this trend would not continue if superlongevity were realized. Second, it is argued that there are good reasons to suppose that there will be a certain amount of self-selection: the happiest are more likely to adopt superlongevity technology. This means that the adoption of superlongevity technology will have the effect of raising the level of aggregate utility.
A number of technological developments on the horizon, for example, discoveries in genetics [Kenyon 1996], stem cell research [Shostak 2002], and the cessation of aging at the cellular level [de Grey 2005], point toward the possibility that sometime this century the length of the human life span could be radically lengthened. Should utilitarians promote or discourage such research? Let us think of ‘apologism’ as the view that it is wrong to extend the human life span beyond its current limits, and ‘prolongevitism’ as the view that we should seek to extend the human life span significantly beyond its current limits.[1] The term ‘significant’ is vague, so for our purposes let us understand it as meaning ‘superlongevity’: an average lifespan of at least 150 years. A life of this length would mean a doubling of our current allotment, that is, approximately 75 ‘extra years’—years beyond the average human life span (in the developed world). Presently a number of people live past 100 years of age, so we should assume a doubling of the average would mean that some will live past 200! Our question may be rephrased then as follows: should utilitarians reject superlongevity with apologists or should they endorse superlongevity with prolongevitists? I hope to show that a strong utilitarian case can be made for prolongevitism. We will use Peter Singer’s utilitarian and apologist paper as a foil: ‘Research into Aging: Should it be Guided by the Interests of Present Individuals, Future Individuals, or the Species?’[1991].[2] Singer argues that prolongevitism will lead to a lowering of aggregate utility, and so superlongevity ought to be rejected on utilitarian grounds. I will argue, to the contrary, that the empirical evidence available does not support Singer’s position. Furthermore, contrary to Singer’s predictions, there is reason to suppose that aging populations will tend to become happier on average through a process of self-selection.
1. Singer’s Apologism
Singer sets up his argument by invoking the following thought experiment. Suppose a scientist in the near future announces
she is ready to begin preliminary trials on an inexpensive drug[3] which dramatically slows the onset of old age. The drug will, the scientist tells us, have no noticeable effect until those who take it reach middle age; but after that it is expected to halt the rate of aging so sharply that the average life span will be doubled. The drug cannot restore the full health and vigour of our youth, but the onset of degeneration and senility will be very much delayed. Those who take it can expect to live to be about 150. Their health over the additional 70 to 80 odd years will not be quite as good as the average over a normal life span today, but it will be good enough—roughly that of people in their sixties and seventies today. Certainly it will not be so poor that we would regard those extra years as not worth living, or even come close to making such a judgment [133].
Singer’s question is whether we should recommend developing the drug given the ‘concern for what will lead to the greatest total amount of happiness, or welfare, over time [144].’[4]
An obvious way for utilitarians to defend apologism is if it can be shown that there is some reason to suppose that the extra years are ones of (aggregate) negative utility. For instance, many discussions of superlongevity consider the idea that the extra years will be filled with negative experiences like boredom, and so would not be worth living [Williams 1973; Kass 2001].[5] Clearly, if we had good reason to suppose that these extra years were years of overwhelming disutility then we would have a good (prima facie at least) case against prolongevitism. Singer, however, does not adopt this position. Indeed, Singer says that we should imagine that the extra years would be years of happiness or positive utility for the affected individuals, ‘since there seems to be no doubt that many individuals would want it…we must conclude that control over aging would be a good for present individuals [136].’[6] So Singer’s apologist conclusion, ‘we should recommend against any further development of the anti-aging drug [144]’, does not rely on the (common) thought that extra years will be ones of overwhelming disutility.
Rather, Singer’s argument turns on a point about opportunity cost; let the following simple example serve as illustration. Suppose we must decide between the ‘one-crop’ and ‘two-crop’ options for the future of our world. The superlongevity one-crop option is where 6 billion people live for the next 150 years. The two-crop option has 6 billion people live for 75 years at which time they are replaced by a second ‘crop’ who live for the next 75 years.[7] Singer says we should advocate the two-crop option. His reasoning is that the average utility per unit of time (AU/T) is higher in the two-crop option, and so the total utility at the end of 150 years will be higher in the two-crop scenario. The reason for the lower AU/T in the one-crop option, according to Singer, lies in the fact that the superlongevitists would not enjoy the same level of health and ‘fresh experiences’, and this will impact negatively on their level of utility. So, the opportunity cost here, says Singer, is that to allow persons to take the life extension pill is to decrease total happiness because the superlongevitists will have a lower AU/T.
One point of clarification perhaps worth emphasizing is that Singer is not advocating the two-crop option because there are more people. As a utilitarian he is concerned with the aggregate maximum of happiness, not an aggregate head count. And it should be clear that Singer could quite consistently claim that, if persons were to live to be 150 years old then they might experience much greater total happiness in their lives than those who live to be 75, so long as the total happiness is less than double what it would be for an average 75 year-old.
One assumption in Singer’s argument is what we may think of as the ‘ecological premise’: we are at (or close to) the carrying capacity of the earth in terms of the number of individuals that can exist at any one time.[8] The finite nature of our planet’s resources means that the human population must be ‘capped’. It is perhaps apparent how central this idea is for his argument. Imagine for a moment that there is no limit to population expansion. Under this assumption then, other things being equal, the additional years of the superlongevitists would only add to total happiness, and so should be enthusiastically welcomed. For example, if the opportunity for galactic colonization opened up the possibility of a great or indefinite population expansion [Bostrom 2003] then we might have good reason to reject Singer’s argument. Indeed, on this assumption, so long as any life is minimally satisfactory—it adds rather than decreases total happiness—and other things remaining equal, then utilitarians ought to enthusiastically embrace prolongevitism.
Interestingly, Singer says nothing in support of the ecological premise; instead, he assumes this premise as part of his thought experiment [133]. Rather than attempting to address the difficult question of the carrying capacity of the earth, I suggest that we grant this, at least for the sake of the argument.[9]
Another premise in Singer’s argument worth commenting on is his understanding of temporal impartiality: we ought to give equal moral weight to future persons as we do actual persons. This is, of course, a controversial premise: some find absurd the idea that we should give the same moral consideration—or indeed any moral consideration—to possible future persons [Parfit 1984; Heyd 1992]. Those that hold the person-affecting view—the view that only actual persons ought to enter into our moral considerations—will have an easy time dismissing Singer’s argument. For the opportunity cost that Singer invokes must be expressed in terms of possible persons. Think again about the two-crop option. The second crop invokes the idea of creating a new crop of persons in 75 years. But at the time of the decision between the one and two-crop options, these persons do not exist—they are mere possibilia—so on the person-affecting view, their happiness cannot enter into our moral considerations. Given the person-affecting view, the one-crop option is the best, for the choices in our moral calculus must resolve to either 6 billion persons living 150 years, or the same 6 billion living only 75 years. So, if we agree with Singer that someone living 150 years will likely have more total happiness in a lifetime than those that live to 75, and we adopt the person-affecting understanding, then prolongevitism follows.[10]
Should we accept Singer’s temporal impartiality? He offers us a couple of reasons to do so. One has to do with the thought that if we accept the person-affecting view then we should have no problem leaving radioactive containers that will leak as soon as the last person alive at present dies. The thought is that we ought to give some consideration to those that might be affected by this radioactive waste, even if they are not yet born [142]. The other example is that, so long as we agree not to have any further children there would be nothing inherently wrong with using up every last resource on this planet [144].[11] Singer suggests that this would be to ignore the claim of all future generations. Again, rather than try to resolve this question here, I suggest we accept, for the sake of the argument, Singer’s temporal impartiality view and see what follows.
The final premise we need to consider is the ‘lower AU/T premise’—the thought that the superlongevitists will have a lower AU/T (LAU/T). As we noted above briefly, Singer offers two reasons in support. One stems from the idea that there is a certain ‘freshness’ associated with youth. On this, Singer quotes Hans Jonas:
…the wisdom [of] the harsh dispensation [is] that it grants us the eternally renewed promise of freshness, immediacy, and eagerness of youth… There is no substitute for this in the greater accumulation of prolonged experience: it can never recapture the unique privilege of seeing the world for the first time with new eyes…The ever renewed beginning, which is only to be had at the price of ever repeated ending, may be mankind’s hope, its safeguard against lapsing into boredom and routine, its chance of retaining the spontaneity of life…[136].
His second reason invokes the health of the proposed superlongevitists. Recall that Singer’s thought experiment asks us to imagine persons whose ‘extra life’—the additional 70 to 80 years afforded by the life extension pill—will be ‘good enough’ but ‘not quite as good as the average over a normal life span today’. From this point it seems a short step to his apologist’s conclusion:
For health is certainly a major factor in our welfare, and if the average level of those living longer lives is not quite as good as the average level of those living shorter lives, then we should expect that, other things being equal, the average quality of life of those living longer lives will not be as good as the average quality of those living shorter lives [139].
Since the average quality of life of those living shorter lives is higher, it seems we ought to prefer a world in which people live shorter lives. So both ‘freshness of youth’ and ‘average level of health’ are said to support Singer’s LAU/T premise.
Of course, for Singer, any value of ‘the freshness of youth’ and the value of health is ultimately to be resolved to their contribution to utility. For example, if (as bizarre as it may seem) in the land of masochists people were happier when their health was failing them, other things being equal, utilitarians should promote ill health. In other words, freshness and health are merely instrumentally valuable: they are valuable to the extent that they promote utility.[12] Thus, Singer’s argument requires that it be the case that a lack of freshness and health will negatively impact utility. Indeed, Singer requires a stronger premise than this, because of the claim of ‘other things being equal’. So, in order for us to accept his premise we require evidence that other things are indeed equal; or, if other things are not equal, the effects of lack on freshness and lower health on utility must be of such a magnitude that they outweigh any contrary factors. For otherwise, prolongevitists can grant that superlongevitists may lack freshness and health, and that freshness and health are major determinates of utility, but they are outweighed by other positive gains in utility.
2. Empirical Support for LAU/T
We have just seen how critical the point that freshness and health are over-riding determinates of utility is for Singer’s argument, hence, it is somewhat surprising that Singer provides so little support for this claim. Before scuttling the development of the anti-aging pill, surely we should ask whether there is empirical evidence for LAU/T. One obvious place to look is contemporary research on happiness. I do not know of any research that investigates ‘freshness’ as an influence on happiness, but there is ample evidence that, other things being equal, one’s level of health does affect one’s level of happiness [Gerdtham and Johannesson 2001 and references therein]. Since health indices tend to decline as we age, it seems that on this point Singer is on solid ground. However, this relationship between health and happiness does not address the rider: ‘other things being equal’. When we look at empirical research done on the relation between age and happiness it turns out, somewhat surprisingly perhaps, that it is U-shaped as in Fig. 1:
Happy
Unhappy
Young Middle Aged Elderly
Fig. 1.
This U-shaped curve is well documented in psychological and economic literature [Clark and Oswald 1994; Deaton and Paxton 1994; Oswald 1997; Easterlin 2001; Frey and Stutzer 2000; Gerdtham and Johannesson 2001; Blanchflower and Oswald 2003]. The youngest and oldest segments of the population are the happiest, while utility reaches its nadir sometime in ‘middle age’: the late 30s or early 40s. So even granting that a loss of ‘freshness’ and declining health might take away from an individual’s happiness, there is at least one other major determinate of happiness, and this factor significantly outweighs any negative contribution caused by declining freshness and health.[13] What this means is that the seemingly reasonableness of Singer’s argument appears to run afoul of empirical research. Clearly this is not to deny that health and freshness are major determinates on an individual’s level of happiness, only that they cannot be overwhelming determinates: on average elderly people are happier than those who are middle-aged.
How decisive is this evidence from the social sciences? This is certainly not the place to review this literature and its methodology in detail, so I will confine myself to two brief points. First, social scientists explicitly made the connection to the utilitarian tradition in their discussion and measurements of happiness, so this provides at least some reason to think that utilitarians ought to give some weight to this literature [Kahneman et al. 1997; Blanchflower and Oswald 2003; Veenhoven 2004]. Second, it would be wrong to imagine that these social scientists have a naive view of the philosophical and scientific difficulties in measuring utility, since even a cursory acquaintance with this research tradition reveals a sophisticated awareness of methodological questions [Warr 1990; Watson and Clark 1991; Konow, and Earley 2003; Ferrer-i-Carbonell 2001; Gerdtham and Johannesson 2001]. Even if one is not satisfied with the methodology of this research, this in itself does not show that Singer is correct, only that we need to study the problem more. In any event, we can be satisfied here with the modest conclusion: the extant empirical research does not support Singer’s view that freshness and health will depress utility in superlongevitists.
It might be thought that what follows from this research is the more radical conclusion: that we should die young—what we might think of as the ‘Logan’s Run’ conclusion. (Logan’s Run is a 1960’s science-fiction book that was made into a campy movie in the 1970s. In the book, everyone was executed upon reaching age 21). This sort of conclusion might seem to follow since it would allow us to avoid having people in their middle years, their 30s and 40s, since this population tends to have LAU/T. However, it could plausibly be objected that such a society might not be sustainable. Furthermore, we would have to take into account how the institution of this policy might affect the happiness of the population. That is, just because the young are relatively happy now, it does not follow that they would continue to be so in the future where they know that they will die at 21. Obviously we need to account for how this policy might affect utility—it may well have a dampening or ‘chilling’ effect on utility.