Of course, having established the importance of things like sentience and expected future utility, we’re now forced to confront yet another question: How should we think about situations where those factors are entirely absent, like when we consider the preferences of people who’ve recently died? Should those people’s wishes receive any moral consideration whatsoever? Based on everything we’ve been discussing so far, it might not seem that they should. After all, dead people are no longer sentient, and can’t receive any future utility. There’s no longer any person there capable of experiencing benefits or harms. So on what basis could their preferences be included in the utilitarian calculus at all? Well, if you believe in an eternal afterlife, you might not actually accept the premise of the question; you might simply say that dead people still do exist, just in a different form, so their preferences must be recognized just like everyone else’s. But assuming for the sake of argument that it’s possible for someone to completely cease to exist after they die, how might this affect their moral status? Peter Singer ponders the question:
Does skepticism about a life after death force one to conclude that what happens after you die cannot make a difference to how well your life has gone?
In thinking about this issue, I vacillate between two incompatible positions: that something can only matter to you if it has an impact on your awareness, that is, if you experience it in some way; and that what matters is that your preferences be satisfied, whether or not you know of it, and indeed whether or not you are alive at the time when they are satisfied. The former view, held by classical utilitarians like Jeremy Bentham, is more straightforward, and in some ways easier to defend, philosophically. But imagine the following situation. A year ago a colleague of yours in the university department in which you work was told that she had cancer, and could not expect to live more than a year or so. On hearing the news, she took leave without pay and spent the year writing a book that drew together ideas that she had been working on during the ten years you had known her. The task exhausted her, but now it is done. Close to death, she calls you to her home and presents you with a typescript. “This,” she tells you, “is what I want to be remembered by. Please find a publisher for it.” You congratulate your friend on finishing the work. She is weak and tired, but evidently satisfied just with having put it in your hands. You say your farewells. The next day you receive a phone call telling you that your colleague died in her sleep shortly after you left her house. You read her typescript. It is undoubtedly publishable, but not ground-breaking work. “What’s the point?” you think to yourself, “We don’t really need another book on these topics. She’s dead, and she’ll never know if her book appears anyway.” Instead of sending the typescript to a publisher, you drop it in a recycling bin.
Did you do something wrong? More specifically, did you wrong your colleague? Did you in some way make her life less good than it would have been if you had taken the book to a publisher, and it had appeared, gaining as much and as little attention as many other worthy but not ground-breaking academic works? If we answer that question affirmatively, then what we do after a person dies can make a difference to how well their life went.
I’m pretty much in the same boat as Singer here, just in the sense that I can understand both sides of the argument and see why either one might be defensible. On the one hand, my instinctive reaction to the scenario he describes is to feel appalled that anyone would be capable of betraying their colleague’s wishes in such a way. Obviously it would be wrong to just throw away the colleague’s typescript like that, right? But on the other hand, if we accept that morality obliges us to honor the wishes of the dead, even at our own expense, then that raises some tough questions. Where do we draw the line, exactly? Do the preferences of everyone who’s ever died still carry the same moral weight, even centuries later? Should we still give moral weight to the preferences of cavemen? Are we morally bound to fulfill the wishes of the ancient Egyptians, even if those wishes are excessively elaborate and demanding? To take the most extreme example, if there were some kind of apocalyptic event that killed everyone on Earth except for one person, would that person be obliged to spend her days carrying out the wishes of all those dead people, who would never be able to appreciate it, even if it meant she would have to forego her own preferences and sacrifice some significant measure of her own happiness in the process?
I can think of a few possible ways of approaching this issue. One way, of course, would be to just bite the bullet and say that dead people really don’t have any moral standing whatsoever, and that we have no inherent obligation to respect the preferences they held while they were alive about what should happen after their deaths. This wouldn’t necessarily mean that their wishes could be totally ignored, mind you, because there would still be other good reasons to respect them. For instance, maintaining a social norm of respecting dead people’s wishes can provide real value to those who are still living, because it helps reassure them that their own wishes won’t be totally ignored after they die, which can be a genuine comfort (especially for people who might otherwise feel particularly anxious about it). So accepting this premise, that respecting the wishes of the dead is just something we do for the benefit of the living, wouldn’t automatically mean that we’d have to just let people go around desecrating gravesites or whatever. But on the other hand, it would mean that our hypothetical apocalypse survivor wouldn’t have to feel obligated to respect the wishes of the dead in the same way that we do, because there’d be no one left for her actions to affect aside from herself. She wouldn’t have to worry about undermining social norms or making anyone feel anxious; whatever was best for her as an individual, that’s what would be morally best overall.
If these conclusions seem intuitive to you, you might find this Benthamian stance appealing (as I do myself much of the time). Still, it doesn’t seem to me that this approach provides the most satisfactory resolution for examples like Singer’s above, because in that scenario, the dying colleague never told anyone else about her typescript, so discarding it wouldn’t have the effect of undermining social norms or making anyone feel disturbed that their wishes might also be ignored after their death. From the Benthamian perspective, then, ignoring her wishes – or the wishes of any dead person – would be totally fine, as long as no one else was aware of them. But that can’t be right, can it?
Well, maybe. But there is an alternative approach that might be available to us here, which goes like this: Granted, in Singer’s example, it does seem unavoidably true that you wouldn’t be causing any post-mortem harm to your dead colleague by disregarding her wishes. But at the same time, it might be possible that by disregarding her wishes, you would be causing her life to have gone worse, as Singer puts it. See, when you originally made your promise to her to find a publisher for her typescript, you were responding to her preference that you make that promise to her. But the thing is, her preference in that moment wasn’t that you falsely promise her that you’d find a publisher – it was that you truly promise her that you’d find a publisher. So depending on whether you subsequently went on to actually deliver on your promise after she died, your act of making that promise to her could have constituted either a satisfaction of her preferences or a violation of them, in that moment, while she was still living. In other words, the moral goodness of your promise would be predicated on your actually keeping it later; so if you ultimately decided not to do so, you’d be retroactively changing the moral status of your earlier promise from good to bad. You’d be retroactively causing your colleague to have been wronged by you while she was still alive.
This idea of retroactive preference satisfaction is a weird one, so it’s probably worth unpacking a bit more here. To start with, let’s consider the basic premise that an expectation or statement about the future can be made true or false – right now, in the present moment – by whether the events it refers to actually end up happening in the future. Here’s Steven Luper:
Propositions are either true or false, and what makes a proposition true is an event or state of affairs that can be labelled its ‘truth-maker’. For example, the proposition I am typing is made true by my typing right now. In some cases, as when I presently assert I am typing, the truth-maker occurs at the same time as the assertion of the proposition. But the two are not always simultaneous. In some cases, as in I went kayaking last week, truth-makers antedate asserted propositions. In others, the assertions come before the truth-makers. For example, the sun will rise tomorrow is made true now by the sun’s rising tomorrow. If the sun does not rise tomorrow, then the sun will not rise tomorrow is true now. One proposition or the other concerning the sun’s rising is true now even though neither truth-maker has occurred yet.
Luper adds, “No mysterious sort of reverse causation is involved in a proposition’s being made true by states of affairs holding at times before or after the proposition is asserted.” But as he goes on to explain, the fact that such a thing is possible means that in certain cases, it may also be possible for “some desires [pertaining to future events to] be fulfilled retroactively.” He writes:
This happens, for example, when a desire is fulfilled by virtue of posthumous events. Suppose that I now want the sun to rise tomorrow. If the sun will indeed rise tomorrow, my desire is fulfilled now – I get what I want now. (Contrast the case in which I want to be watching the sun rise but it is midnight.) This is true regardless of whether I live to see it rise.
What this means for Singer’s scenario, then, is that when your colleague makes you promise her that you’ll find a publisher for her typescript, the preference she holds in that moment – that your promise to her be a true one – is either satisfied or violated in that moment; but whether it’s satisfied or violated depends on events that won’t happen until later on, after her death. If you go on to fulfill your promise and find a publisher, then her preference in that earlier moment will have been satisfied; but if you don’t, then it won’t have been. Either way though, in the moments immediately following her death, the mere fact that your satisfaction or violation of her preferences happened in the past shouldn’t cause you to just blow it off on the basis that “what’s done is done” – because in this case, what’s done isn’t actually yet done. Whatever decision you make now will determine whether her preferences were satisfied or violated while she was alive, and will thereby have a genuine effect on how well her life can be said to have gone for her. So to not care about whether you can cause her preferences to have been satisfied during her life, simply because “it’s all in the past,” would be like not caring whether you could literally reach back through time and cause or prevent some morally important event from happening in a more direct way. I mean, just imagine if you could (say) press a magical button that would send a signal back in time that would cause your most recently deceased loved one to have gotten in a terrible accident shortly before the end of their life, causing them to have spent the last few months of their life in severe pain. Would you consider it morally bad to press the button? Or would you consider it morally neutral, since they’re dead now and it can’t matter to them currently whether their past preferences were fulfilled or not? If you would in fact consider it immoral to press the button – as I think most of us would – then likewise, you should consider it immoral to break your promise to your colleague in Singer’s scenario, since that would cause her past preferences to have been violated. Granted, keeping your promise wouldn’t be an absolute imperative, any more than any other moral act would be; it would have to be weighed in the utilitarian calculus alongside everything else. So if ensuring that your promise to her was true would turn out to cause so much disutility that it would outweigh whatever utility she might have gotten from it while she was alive – like for instance, if getting her typescript published would require you to ruin your own life and/or the lives of others – then that would constitute a sufficient moral reason for you to break your promise. The point here is just that, in terms of the utilitarian calculus, any past preferences that you might be able to retroactively satisfy or violate should count for just as much as present or future preferences do. Remember our cardinal rule, that when you calculate the expected goodness of your action, you have to account for any and all preferences that are affected by that action? Well, I’ve been talking this whole time about how that mandate includes both preferences that exist and preferences that will exist; but now we may have to add one more category to the pile – namely, preferences that have existed in the past, which your current actions might be able to retroactively satisfy or violate, despite their no longer existing in the present. True, it’s rare to encounter situations where such preferences actually come into play – but those situations can and do exist (assuming you buy Luper’s explanation), so ultimately you do have to account for them.
Of course, that’s not quite the end of the story; there are a few complicating factors to consider here. For one thing, it might not seem immediately obvious how this approach could be generalized to apply to cases in which no one has explicitly promised the dying person that they’ll fulfill their wishes after their death. If there’s never the possibility of keeping such a promise and thereby satisfying the person’s living preference that the promise be true, how can there be any retroactive benefit in fulfilling their wishes after their death? But I don’t think this is actually that big of an issue, because even if a dying person never makes anyone promise them that their wishes will be respected after they die, that person will still have a desire and an expectation that this will be the case – and they’ll still have a preference that this expectation be a true one rather than a futile one that never gets fulfilled. So respecting their wishes after their death can still retroactively satisfy that preference they held while they were alive, and can thereby cause their life to have gone better for them. In that sense, it’s just like satisfying the preferences of a still-living person; you don’t have to have ever explicitly promised someone that you’d respect their wishes in order for your respecting their wishes to be good for them, nor for your ignoring their wishes to be bad for them. Your actions can be better or worse for someone even if you’ve never met them before in your life; simply the fact that they want their wishes to be respected is enough to make it good for them if those wishes are respected and bad for them if they’re ignored. And the same applies to retroactive preference fulfillment for people who’ve died.
Now, like I said before, there are limits to all this. Retroactively-fulfillable preferences have to be weighed in the utility calculus alongside every other preference – so if someone wanted something done after their death which would be too much of an imposition on those who were still living, then fulfilling that desire wouldn’t necessarily be morally obligatory. If it would be unreasonable, for instance, for someone to expect their community to build them a giant golden monument while they were still alive, then it would be just as unreasonable for them to expect such a thing after their death (sorry, pharaohs). Likewise, in our earlier example of a sole post-apocalypse survivor, if it would be unreasonable for humanity to expect a single person to fulfill all of their preferences now, while everyone was still alive, then it would be just as unreasonable to expect a single person to shoulder the burden of retroactively fulfilling all of humanity’s post-mortem preferences after the apocalypse. As a general rule, if a dying person’s expectations just include basic things like “not desecrating their gravesite or violating their dead body,” then those expectations are simple enough that they can and should be respected; but if their expectations would demand significantly more than that from those still living – like if a father demanded that his children never be allowed to date anyone after his death, or if a wife demanded that her husband risk his life to scatter her ashes in the middle of a hurricane, or something like that – then the net expected utility of such expectations would be so obviously negative that the survivors wouldn’t have to feel morally obligated to fulfill them. In short, a post-mortem request would only be justifiable if (all else being equal) the retroactive preference satisfaction that the dying person would derive from having their request fulfilled would outweigh the negative preference satisfaction that the survivors would have to incur in order to fulfill the request.
Having said all that, though, we still aren’t quite done yet; this idea that we have to account for dead people’s wishes raises the broader question of what the larger-scale implications are here. Are we obligated to account in our moral calculus for all the post-mortem wishes of everyone who’s ever lived? I’m actually inclined to think so, personally; I don’t think that desecrating ancient gravesites is any more acceptable than desecrating the gravesites of people who’ve died more recently. But as dramatic as it sounds to say that we have to acknowledge the post-mortem preferences of everyone who’s ever lived, this might not actually be as big a deal as it you might think. (In fact, it may be largely a moot point.) After all, whatever most of our forebears’ object-level preferences might have been about what should happen after they died, they would also have had a meta-preference that those preferences would ultimately turn out to have been noble ones – i.e. that their wishes would turn out to be morally good for those still living, rather than misguided and harmful for them. So if it turned out that what they thought they wanted would actually be bad overall (like if their wish to have everyone follow their religion after they died ended up being harmful because their religion turned out to be wrong), then their true extrapolated preference would be that their misguided object-level preference be ignored, and that future generations just do whatever was most moral instead. It seems safe to assume that the vast majority of our forebears would have held such a preference – that what they would have wanted more than anything after their death was simply for things to go as well as possible for the survivors – so although there might have been a few of them whose post-mortem wishes really were purely selfish, such exceptions would have been vastly outweighed by the meta-preferences of the benevolent majority who just wanted what was best for the world after they died. Ultimately then, what that means for us here in the present is that, for the most part, the best way to retroactively maximize our forebears’ preference satisfaction is to just do what we would have considered to be globally best anyway.