December 26, 2009

On Singer and radical life extension

Russell Blackford is a guest blogger for Sentient Developments.


The current (December 2009) issue of The Journal of Medical Ethics contains my paper: "Moral pluralism versus the total view: why Singer is wrong about radical life extension." There, I critique an early 1990s paper by Peter Singer, which argues that we should not proceed to develop a hypothetical life-extension drug, based on a scenario where developing the drug would fail to achieve the greatest sum of universal happiness over time. I respond that this is the wrong test. If we ask, more simply, which policy would be more benevolent, we reach a different conclusion from Singer's: even given his questionable scenario, development of the drug should go ahead. A more pluralistic account of the nature of morality than used by Singer reaches a benevolent recommendation on life-extension technology.

My paper is intended not merely to offer a better solution to the conundrums raised in Singer's original piece, but also to suggest a methodology of much wider value in applied moral philosophy.

The scenario

Singer's argument employs an imaginary scenario in which life extension would not increase, and would actually reduce, the universal sum of happiness or welfare (henceforth, I will refer simply to "happiness") over time. Singer describes a scenario in which an anti-ageing, or life-extension, pill would more-or-less double human lifespans, but the level of happiness enjoyed in the second half of a typical individual's life would be lowered to some (relatively small) extent. He also stipulates that it would be necessary to ensure that fewer people came into existence over time if the life-extending pill were developed and used. Given this scenario, he thinks, we should not go ahead with developmental work on the hypothetical life-extension pill.

More specifically, Singer imagines a scenario in which those who take the drug experience no effect during their early decades of life. However, when they reach middle age, the drug retards further ageing so dramatically as to extend an average life span from about 75 years to about 150 years. During her additional years of life, an individual's health will not be restored to youthful levels, but it will be good enough for a very worthwhile quality of life (similar to the health of people in their sixties or seventies today). An individual may find that life has lost some of its experienced "freshness", and the combination of this (should it happen) with somewhat reduced health will make her additional years less happy than her first 70 or 80 years of life - but not greatly so.

Mark Walker has questioned this scenario elsewhere, suggesting that it is unrealistic to assume that the first 70 or 80 years would typically be happier than the second for those with what he calls "superlongevity". My own approach is more fundamental, as I conclude that Singer gives the wrong recommendation even if we accept all of his stipulated facts.

A further stipulation made by Singer is that resource limitations will require population controls, whether or not the drug becomes generally available, but they will need to be more severe if the pill is developed. Fortunately, Singer tells us, the pill will allow for an increase in average child-bearing age and a lower fertility rate. Nothing in his analysis depends on the exact ingredients of a population policy; rather, his essential point is that it will be necessary to devise an appropriate policy to ensure that only half as many people are born if the life-extension drug is available. I.e., he has in mind a scenario in which the total number of people who will be born and live out their lives over a large number of years will be half what it would have been without the drug. (The fairly simple calculations involved are discussed in my article; suffice to say that Singer is more or less right here, if we adopt all his basic assumptions.)

It is easy to demonstrate that, if we adopt all these assumptions - which Singer evidently regards as constituting a plausible scenario - the total sum of happiness, over a set period of time, is greater in a society without the life-extension pill than a society with the life extension pill. Moreover, the average society-wide happiness at any given moment is higher in the society without the life-extension pill. On the other hand, typical individuals of the future will have better lives in the society with the life-extension pill than in the society without it. This may seem paradoxical, but it is actually quite easy to demonstrate that it is true so long as we make some plausible assumptions. In that case, should we go ahead with developing the drug or not?

Recommendations: Singer's ... and mine

Singer argues that should not develop the drug; I disagree. But here the argument gets complex, and I cannot, in a relatively brief blog post, do justice to the complex issues that I needed a 7000-word article to tease out properly. I agree with Singer that we should take into account the interests of future generations, not just the interests of people who are alive now, but what follows from this?

It appears that Singer wants to maximise what we could call total future happiness-years (I hope the meaning of this is transparent: in any event, it involves multiplying the number of future people by the average number of years they live, and then by their average level of happiness across an entire life). He wants to do this at all costs, even if the people who come into existence have worse lives than the smaller number of longer-lived of people who would have come into existence under a different policy. I find that very implausible. Although Singer offers thought experiments to support his approach to the question, I find them unconvincing (my article explains why in some detail).

We should, I suggest, adopt the more benevolent policy, and we should not think of benevolence as a matter of maximising total happiness-years. In a situation such as the one that concerns us, the choice of the pro-pill and anti-pill versions of Singer's life-extension scenario, we should not try to maximise the overall number of happiness-years. We should try to produce the most fortunate lives.

Moral theories

It may be that utilitarians, such as Peter Singer, are inevitably pushed toward "total-view" thinking - which attempts to maximise the total amount of happiness in the universe - rather than toward a view that we should ensure the best possible lives for those people who will come to exist in the future. As a result utilitarians can, again paradoxically given the sympathies that underly their moral theory, make policy recommendations that are not the most benevolent available.

Unfortunately, all utilitarian theories developed to date contain paradoxes or involve counterintuitive implications. If, however, we take a more pluralistic approach to the sources of our morality, such difficulties vanish. I expect that a considerable diversity of values underpins our actual moral thinking. We care, for example, about the reduction of suffering, about the lives of others going well, and about people being able to live with a certain spontaneity. We value wilderness, art and culture, the quest for knowledge, the existence of complex, creative cultures … and many other things. To at least some extent, we value all these for themselves, not solely because of their further utilitarian effects.

We do not value the largest possible sum of happiness over time ... which can, in principle, be gained by multiplying the number of sentient beings (so long as they have lives that are at least worth living). What we value, rather, is that whatever actual lives come into being should go well. Other things being equal, we value the outcomes that would be chosen, among those possible, by a benevolent decision-maker, not by a decision-maker committed to total-view utilitarianism. As shown by the way Singer has set up his life-extension scenario, these two kinds of outcomes can diverge.


It is clear to me that I should vote to go ahead and develop the life-extension pill - and so, after reflection, should you, and so should Singer. No plausible values are violated by this action; quite the opposite. Far from feeling guilt or regret at having adversely affected another person, or having destroyed or damaged anything precious, an individual who votes to develop the life-extension pill has every reason to feel virtuous. She will have helped to create a world in which lives go better than (more and different) lives would otherwise have.

I am, however, conscious that some readers will find this very truncated version of the argument unconvincing. That may, of course, be because I am wrong! However, it may also be because the issues become quite complicated, and that I really do need the considerably greater length of the full article in The Journal of Medical Ethics to explain them properly. If you have library access, I suggest you look up the original article and the other works cited there.

Otherwise, I can only promise that I will return to similar issues in future writings, perhaps here, but certainly in my own blog over at Metamagician and the Hellfire Club.


Anonymous said...

If that's really the utilitarian view, then that would be like banning reliable cancer cures if there was evidence that cancer survivors were slightly more prone to depression. I don't need to know any of the "complicated issues" to know how insane that is.

Unknown said...

ya well with technology giving us all the option for eternal life very soon, i almost understand Singer's fears of overpopulation, but he is forgetting that technological advances shall also soon allow us to terraform mars, the moon (the moon of Mars lol), and possibly even venus ... then we will also engineer hardy forms of life which will be able to survive on the Jovian Moons etc ... isn't the ultimate goal of any species to increase it's numbers? god told Father Abraham to be fruitful and multiply ... but ya i hope they give me the option to suicide if i get tired of it after a couple hundred years

Bill Garthright said...

We've already increased lifespans greatly in the past century or so. If the later parts of our lives make us less happy than the early years, was that then a mistake?

The Earth is finite, so there's some level of overpopulation (the exact level isn't important for this argument). If we died quicker, we could have more people living for a shorter period of time. And if we tend to be happier as children than when elderly,... would that then make sense? I don't think so.

We've already achieved "radical life extension." Would any of us want to give that up?

Peter de Blanc said...

After reading your post, I still have no idea what you mean by "benevolence," other than that it's not utilitarianism.

Robert Wiblin said...

Longer life --> Greater human capital accumulation --> Greater wealth and technology --> Ability to support more people if we so chose.

DEEN said...

Does it really have to be that complicated? Maybe I'm not understanding the details well enough, but it seems to me that Singer's argument can be defeated by a relatively simple reductio ad absurdum:

Let's assume that there is a finite maximum population size that can be supported by the Earth's resources. That means we will eventually have a steady state society with a fixed population size. Population control ensures that only as many babies are allowed to be born as needed to replace those that die.

Now let's accept for the moment Singer's criterion of maximum total happiness, and apply it to this situation.

Suppose that happiness is more or less constant throughout your life. Then with a fixed population size, the total amount of happiness will be the same, whether it's a population of young people, old people, or very old people. It doesn't matter for the total happiness whether you have many sequential short lives, or fewer longer lives. In this situation, Singer's criterion therefore doesn't help one at all when deciding to develop the pill or not.

(If you instead assume that happiness increases with age (possibly due to increased knowledge and wisdom), Singer's argument will of course lead you to decide to develop the pill.)

However, if you assume (like Singer does) that happiness is higher in youth than in middle age, and the pill doesn't extend youth, then clearly not developing the pill gives greater total happiness. After all, a population of youthful short-lived people will be more happy than the same number of people that are long-lived but not youthful. As I understand it, this appears to be the gist of Singer's argument, and his reason to conclude not to develop the pill.

But if you would more generally apply this argument under thee assumption of a fixed population size, it would mean that you would maximize happiness by killing people shortly after their peak happiness years are over. This way more people can be born and live their peak years, resulting in a larger overall happiness.

Of course, that would be a monstrous policy. Therefore, Singer's criterion doesn't result in an acceptable morality and must be rejected as a useful moral criterion.

sethop said...

Hmmm. "Early 1990s" is a long time ago. Perhaps Singer's position on this has evolved? Has he made any response to your critique thus far?

c said...

I don't know Singer's work on this topic, but it seems to me that his position does not entail the view that we should logan's run everyone at 30, even granting his view that happiness declines during or after middle age.

Once a being comes into existence its continued existence is a matter of some particular concern and its preferences will be thwarted by any shortening of her or his life. So we can also dismiss the claim that Singer must be opposed to cancer treatments since these are therapies for already existing beings.

It does seem to me that Singer is right that whether we should want to develop or make available such technology depends crucially on the quality of life that would be extended. That is, such therapies cannot be presumed to be desirable or good.

Nevertheless it seems to me that death and generally confronting finitude is one of the deepest sources of anxiety and distress for human beings. Living our first 75 years without the looming limits of our existence might allow us to live more happily during this time and so might offset the happiness penalty of our later years.

But, I take it that Singer's basic point is correct--whether we should develop these technologies cannot be determined without considering the quality of life that is being extended.

DEEN said...

" seems to me that his position does not entail the view that we should logan's run everyone at 30"
I'm sure Singer doesn't actually advocate for that, no. But it is where the application of his criterion of the greatest sum of universal happiness over time appears to lead. At least in a society that has reached the maximum sustainable population, where a (happy) baby can only be born when another (miserable old) person dies. You have not actually addressed the argument and showed why it doesn't lead there.

In fact, in your reasoning, you're actually using a much different criterion than Singer uses: fulfilling the wishes of the existing living people. Which is a useful criterion, of course, but not one that supports Singer's argument, nor his conclusion, as I will show:

"Once a being comes into existence its continued existence is a matter of some particular concern and its preferences will be thwarted by any shortening of her or his life."

By this argument, as long as the individual wants to continue living, life should be extended by any means possible. Otherwise, we'd violate the preferences of the individual. Therefore, the pill should be developed.

"So we can also dismiss the claim that Singer must be opposed to cancer treatments since these are therapies for already existing beings."
Why wouldn't the life-extending pill benefit already existing beings as well?

"It does seem to me that Singer is right that whether we should want to develop or make available such technology depends crucially on the quality of life that would be extended."

Not by your argument above. Except maybe when quality of life has dropped below the threshold where the individual no longer wants to continue living. And in that case, the individual can simply choose not to take the pill. So the pill should still be developed.

"Living our first 75 years without the looming limits of our existence might allow us to live more happily during this time..."
Even more reason to develop the pill.

Seriously, all your arguments indicate that the pill should be developed.