Version: 1.0.0
Last updated: 2021-09-19
In 2015 Samuel Scheffler gave three lectures at the Oxford Uehiro Centre for Practical Ethics, presenting draft material from his book: Why Worry about Future Generations?
Scheffler thinks that population ethics is a bit overrated by Derek Parfit & friends. He argues that a focus on impartial axiology “underestimates and misrepresents our concern” for future generations.
In his closing lecture, titled “Conservatism, Temporal Bias, and Future Generations”, Scheffler suggests that we can, do and should mostly ground our concern for the long-term future in a “sane conservative disposition” (my emphasis):
We have reasons of at least four different kinds for caring about the fate of future generations. Reasons of love, reasons of interest, reasons of value and reasons of reciprocity. All of these reasons depend in one way or another on our existing values and attachments and on our associated disposition to preserve and sustain the things that we value.
[...]
The concern for the future of humanity flows naturally from a conservative concern for the things that we value. Now, it's our very attachment to the status quo that propels our concerns into the future. Without such attachments, it's not clear how much reason we would have to care about humanity’s survival.
[...]
There's no consensus among those who hope to find a satisfactory population axiology about which one is the best candidate. But even if there were such a consensus, what claim is the preferred axiology supposed to have on our motives? Why is it thought either that we do, or that we should, care which population outcomes are judged superior by the lights of this or that axiological principle?
My suspicion is not that the proposed rationale for population axiology overestimates the extent of our concern for our successors, but rather that it underestimates and misrepresents our concern.
It underestimates it because it neglects the variety of reasons we have for concerning ourselves with the fate of future generations.
And it misrepresents it because what those reasons support is not a generic concern for the welfare interests of future people, but rather a more specific desire rooted in the values we affirm in our daily lives: that the chain of generations should be extended into the indefinite future, and that our successors should be able to live under conditions conducive to their flourishing.
Scheffler's closing remarks:
Once we free ourselves from the thought that the basis for any concern about the future of humanity must lie in an [impartial] principle of beneficence [...] we can see that we have reasons of a number of different kinds, all rooted in our actual attachments as flesh and blood human beings, for wanting future generations to survive and to flourish. In so far as these reasons depend on our existing values and attachments and on our conservatism about value, they depart from moral and temporal neutralism.
Yet it is to these very departures rather than to any form of neutralist beneficence that we must look in order to identify our strongest and deepest reasons for caring about the fate of our successors. Or so I have been trying to show. At the very least, I hope to have persuaded you, that there is an alternative to thinking about problems of future generations in exclusively or primarily beneficence-based terms, or indeed in exclusively moral terms of any kind.
If we broaden our horizons, we may find that we have even more reasons than we realize to worry about the fate of future generations.
In the appendices to this link post, I include:
- Scheffler's summaries of the three lectures (~450 words).
- My highlights from Scheffler's third lecture (~3500 words).
- My highlights from Hilary Greaves' critical review of Scheffler's subsequent book (~1,100 words).
Some of the key claims
In Appendix (2), I highlight Scheffler arguing for (or sometimes just asserting) the following claims:
2.1 Our existing attachments justify concern about future generations. Impartial axiology may be a source of moral justification, but it is not our only source of justification, moral or otherwise.
2.2 Utilitarianism cares about maximising value in general. The conservative disposition shares this concern, but moderates it with a commitment to grant special care to particular things that already exist.
2.3 We can believe something is valuable without actually valuing it.
2.4 Valuing involves attachment, attachment requires acquaintance, and non-existence makes the relevant form of acquaintance impossible.
2.5 Particular things that we value are sources of distinctive reasons for action and distinctive patterns of emotional engagement.
2.5 Conservatism is compatible with change, the creation of new things.
2.6 The conservative disposition to sustain and preserve the things that we value underwrites our concern for the future of humanity. It is not a backward-looking impulse that competes with or inhibits this concern.
2.7 Some forms of status quo bias are not irrational.
2.8 We value through time. We should be cautious about assuming that every manifestation of temporal bias in our valuing attitudes is irrational.
2.9 Our concern for the future of humanity and for the flourishing of future generations depends on a conservative disposition that applies directly only to presently existing and past bearers of value.
2.10 We should be careful about adopting the pose of “impartial concern” and “temporal neutrality” too rigidly. Both principles seem attractive at first blush, but for Scheffler they have lost their allure.
N.B. I have only listened carefully to the third lecture. I have not read the book. There's probably a bunch more interesting stuff in there if you care to look.
How to listen to or read the lectures
The Uehiro website has audio of all three lectures. You can get them in your podcast app by searching "uehiro lectures".
I couldn't find transcripts. If you prefer reading, I guess the book is your best bet. You can read it for free here with your academic library membership, or here without.
About Samuel Scheffler
Samuel Scheffler is a Professor of Philosophy at NYU and the author of six books: The Rejection of Consequentialism (1982, rev. ed. 1994), Human Morality (1992), Boundaries and Allegiances (2001), Equality and Tradition (2010), Death and the Afterlife (Ed. Niko Kolodny, 2013), and Why Worry about Future Generations? (2018).
He edited Consequentialism and its Critics (1988), an essay collection with contributions from Parfit, Foot, Nagel, Nozick, Railton, Rawls, Scanlon, Sen and Williams.
Hilary Greaves wrote a critical review of the book—my highlights are in appendix 3, and some commentary is below.
Leopold Aschenbrenner has a blog post exploring some of these themes, titled Burkean Longtermism. He says he was not aware of the Scheffler lectures when he wrote the post.
Joseph Carlsmith discusses related themes in his blog posts titled Alienation and meta-ethics (or: is it possible you should maximize helium?) and Against neutrality about creating happy lives.
There are several long-running philosophical debates lurking in the background here. Some that are salient to me:
Things I haven't read yet, but which I hope to look at soon:
I'm interested in metaethics and moral psychology, but I don't feel on top of the relevant literature, and I have certainly not reached stable views.
If you are interested in these themes, I would love to hear from you, whether in the comment section here, on Twitter, or in conversation.
I'm particularly keen to hear from people who think they might be able to quickly “put me out of my misery” by pointing out a basic confusion, missing concept, or key person or thing it seems like I ought to read. If you think you're seeing rookie errors or gaps in my thinking, you are probably correct.
Conservative in theory, or just in practice?
My nose, such as it is, makes me wonder if the principles of impartial concern and temporal neutrality are underrated by most people, but overrated by some moral philosophers. And—relatedly—if conservative thought might bit a bit underrated by several figures in the Benthamite tradition.
Quite a lot of people—especially philosopher-psychologists—make claims along these lines. A contemporary example is Johnathan Haidt on Durkheimian utilitarianism. But what, exactly, is the disagreement here?
Often, the dispute seems “merely” practical—something like:
Some flavour of consequentialism is right in theory, but in practice we need to think differently due to our regrettable hardware constraints (aka “human nature”).
In these cases, it often seems like people are talking past each other, and are in greater agreement than they realise. C.f. Multi-level utilitarianism.
Sometimes, though, the dispute seems deeper. For example, Bernard Williams—channeling Nietzsche—wants to push back against a “scientistic” trend in moral philosophy, and against philosophers who exhibit “a Platonic contempt for the human and the contingent in the face of the universal”. According to Williams, such philosophers presume that:
...if there were an absolute conception of the world, a representation of it which was maximally independent of perspective, [then] that would be better than more perspectival or locally conditioned representations of the world.
And, relatedly:
...that offering an absolute conception is the real thing, what really matters in the direction of intellectual authority.
Williams thinks of philosophy as fundamentally about “making sense of being human”, so the metaphysical moral realists' attempt to represent “the world as it is anyway”—to construct a theory of value abstracted from any human perspective—strikes him as misguided.
It's all practice?
A related objection comes from the American pragmatists. I've only read a tiny bit of William James and Richard Rorty, but my impression is that for them:
P1. The “theory vs practice” distinction does not make sense. It's all practice—it's all about what is adaptive.
P2. Justification is about what other people will let you get away with.
P3. Moral truths are contingent, not timeless.
On this picture, it seems like our existing attachments are, fundamentally, the only option for justifying a concern for future generations. Our attachments change over time, but there's no external standard by which we can judge the changes as good or bad.
Hilary Greaves on Scheffler
I've long had a hard time pinning down the distinction between normative and non-normative reasons. It seems like Greaves and Scheffler probably have a crux or two in this area. This leads Greaves to dismiss as “descriptive” an account which Scheffler clearly intends to be normative:
He is not pointing out (as the standard account does) that the existence and fate of future generations matters “from the point of view of the universe”, and drawing the conclusion that they should therefore also matter to us. Rather, he is pointing out that they do already matter to us.
The two philosophers seem to have a different idea of what value theory is aiming for: for Greaves, it seems to be about seeking independent, universal, science-like truths; for Scheffler it seems more contingent, more rooted in where we happen to find ourselves. This gets us back to Euthyphro: do we value X because X is good, or is X good because we value X? And also back to Williams, who thinks that metaphysical moral realists want to be able to tell a story of improvement, when really the story is just one of change.
Greaves has a hard time understanding the nature of Scheffler's “reasons of interest”:
On one very natural reading, reasons of interest are reasons of prudence, rather than of morality. We have self-interested (that is, prudential) reasons to want to live valuable lives, and thus the existence of future generations is instrumentally important to us in prudential terms. On this prudential reading, paying any significant attention to reasons of interest would be somewhat grotesque. There in the balance hangs an astronomical amount of human welfare: the happiness of quadrillions of future persons, and even the question of whether any of them gets a chance to experience the joys, loves, projects that life has offered us. But never mind them – I (rightly) feel better about my own life, as I cycle to work, as I hang out the laundry, if I (correctly) believe that future generations will exist and/or that my activities are benefitting them, and therein lies a significant part of my reason to do anything about it.
Daniel Dennett recommends using a “surely alarm” when we read philosophy. Personally, I also like to run a “superlative alarm”. In these passages, it's triggered by the word “grotesque”.
I find myself tempted to argue that the prudential reading is not as grotesque as it might seem. Or at least, insofar as it is grotesque, it is at least a familiar—human, all-too-human—form of grotesqueness, that we should perhaps hesitate to condemn.
The “huge stakes vs comparatively tiny personal concerns” dilemma often comes up for consequentialists. How could Mr Parfit spend so much time and money on photography, when he could have bought more malaria nets? Or Mr Singer pay for private nursing for his mother, for that matter?
The answer, of course, is that they are both humans. They both claim some right to privilege the people and pursuits they happen to find themselves attached to, over things that—impartially considered—matter much more. Their philosophical beliefs about what is impartially valuable do not determine what they actually value, and they find a way to live with that.
The question of what we humans ought to do seems importantly different from the question of what would be ideal, according to an impartial axiology. We find ourselves somewhere, and—with the conservatives—we don't have to wish we were nowhere. A conservative disposition seems compatible with a deep and serious concern for future generations—just not a totally self-effacing one.
Appendix 1. Scheffler's summaries of the three lectures
Lecture 1: Temporal Parochialism and Its Discontents
Most of us who live in contemporary liberal societies lack a rich set of evaluative resources for thinking about the human beings who will come after us. We do not possess a highly developed set of ideas about the value of human continuity, or about the values we hope will be realized in the future, or about the values and norms that should inform our own activities insofar as they affect future generations or depend on the expectation that there will be future generations. Yet we are hardly indifferent to the fate of our successors, and it is not uncommon for issues like climate change that implicate our attitudes toward the future to generate passionate interest and intense controversy. Much of the philosophical literature dealing with future generations focuses on issues of moral responsibility and approaches these issues from a broadly utilitarian perspective, devoting special attention to the puzzles of “population ethics”. In this lecture, I explain why I take a different approach. Rather than focusing exclusively on issues of moral responsibility, I want to consider the broader question of how future generations feature in or are related to our practical and evaluative thought as a whole. My aim is to explore the evaluative commitments that may be latent in our existing attitudes and may help to enrich our thinking about the significance that future generations have for us.
Lecture 2: Reasons to Worry
In this lecture I argue that, quite apart from considerations of beneficence, we have reasons of at least four different kinds to try to ensure the survival and flourishing of our successors: reasons of love, reasons of interest, reasons of value, and reasons of reciprocity.
Lecture 3: Conservatism, Temporal Bias, and Future Generations
The reasons discussed in the previous lecture all depend in one way or another on our existing values and attachments and our conservative disposition to preserve and sustain the things that we value. The idea that our reasons for caring about the fate of future generations depend on an essentially conservative disposition may seem surprising or even paradoxical. In this lecture, I explore this conservative disposition further, explaining why it strongly supports a concern for the survival and flourishing of our successors, and comparing it to the form of conservatism defended by G.A. Cohen. I consider the question whether this kind of conservatism involves a form of irrational temporal bias and how it fits within the context of the more general relations between our attitudes toward time and our attitudes toward value.
Appendix 2. My highlights from Scheffler's third lecture
2.1 Our existing attachments justify concern about future generations. Impartial axiology may be a source of moral justification, but it is not our only source of justification, moral or otherwise.
According to the alternative perspective that I've been defending in these lectures [...] we have a variety of reasons, all rooted in our existing attachments to humanity, and to valued forms of human activity and endeavor to care about the capacity of future generations to survive and to flourish.
From this perspective, it's a mistake to think that our reasons for caring about the fate of future generations are hostage to our ability to construct a satisfactory population axiology, a complete theory of the relative goodness of total states of the world. The contrast between these two views is both normative and motivational.
Normatively there is on the one side, a moral imperative to implement a general principle of beneficence. On the other side, there was a set of compelling reasons—whether moral or non-moral—to secure the ability of our successors to survive under conditions conducive to their flourishing.
Motivationally there is on the one side a generalized concern for the welfare interests of all people, including all future people. On the other side, there's a conservative disposition to sustain the humanity we love and the existing values we now cherish.
It will already be clear that I find the second perspective, more persuasive, both normatively and motivationally. As a normative matter I find the reasons that it highlights for concerning ourselves with the fate of future generations, more compelling than those suggested by the beneficence-based approach.
And at the motivational level, the fact that it grounds our concern for future generations in a conservative disposition to sustain our existing attachments, puts that concern on a more secure footing and integrates it into a unified stance we may take toward the diachronic dimension of our values.
2.2 Utilitarianism cares about maximising value in general. The conservative disposition shares this concern, but moderates it with a commitment to grant special care to particular things that already exist.
The conservative disposition I've been discussing is not a form of political conservatism. It's a disposition to preserve or sustain the things that we value and both the things that we value in the steps necessary to preserve them. #todo what??? mistranascirption?
[...]
In his wonderful essay, defending what he calls small C conservatism, [Jerry] Cohen advocates "a bias in favor of existing value", by which he means that we should regret the destruction of particular valuable things as such, even when it would lead to their replacement by things of greater value.
He thinks that "everyone who is sane" has this bias to some degree and that it is quote "rational and right, that they should".
For Cohen, the crucial distinction is between value in the abstract and the particular things that have value. Or, alternatively, between the value that things bear and the bearers of such value.
The conservatism that he defends holds that particular things that have value take priority over value itself in at least two related senses.
First particular valuable things do not matter or count simply because of the amount of value that they bear or that resides in them.
Second, we have at least some defeasible reason to preserve particular, valuable things as such, even if by sacrificing them, we could produce more value overall.
[...]
Quoting Cohen:
To seek to maximize value is to see nothing wrong in the destruction of valuable things as long as there's no reduction in the total amount of value as a result. Unlike the conservative, the utilitarian is indifferent between adding to what we have now got at no cost, something that has 5 units of value and adding something worth 10 units of value at the expense of destroying something worth 5.
If the utilitarian is willing to sacrifice a particular valuable thing, whenever it can be replaced by another particular valuable thing with even slightly more value, then the original item is being valued solely in proportion to the value that it bears. And to say that is just to say that the utilitarian, unlike the conservative does not value the bearers of value independently of the value that they bring.
Cohen's insight is best appreciated if we focus not on Cohen's category of particular value as a type of value or indeed on any other category of value, but rather on what it is for a person to value a given thing.
2.3 We can believe something is valuable without actually valuing it.
I regard the distinction between something's having value and one's valuing it as significant.
Valuing something in my view involves a complex syndrome of attitudes and dispositions, including a belief that the thing is valuable, a susceptibility to experience a variety of context dependent emotions concerning the thing, and a disposition to treat considerations pertaining to the thing as providing one with reasons for action in relevant contexts. Here, I'm using thing in a broad sense that encompasses any object of our valuing attitudes.
It's possible to regard something as valuable (or in Cohen's terms as possessing particular value) without actually valuing it oneself in this sense. Indeed, most of us regard many things as valuable that we ourselves do not value.
Valuing something involves more than just believing that it's valuable. It involves a kind of attachment to or investment in or engagement with that thing. This sort of attachment or investment or engagement is constituted both by emotional vulnerability and by a disposition to see oneself as having reasons for action, with respect to the valued item that one does not have with respect to other comparably valuable items of the same kind.
If I value my relationship with you, for example, then I will typically be vulnerable to feelings of distress. If you are harmed. And I will see myself as having reasons for acting in your behalf in relevant deliberative contexts that I do not have for acting in behalf of other equally valuable people.
So, for example, if my, if I value my friendship with you, then I'm justified in thinking: I have reasons to act in your behalf that other people do not have. And that I do not have with regard to people who are not my friends. And if I value an antique rug that has been in my family for generations, then I'm justified in thinking that I have reasons to care for it or preserve it that other people do not have. And that I do not have with regard to other antique rugs. This does not mean that I have no reason to do anything at all on behalf of people who are not my friends or indeed that I never have reasons to help preserve other antique rugs or other people's family heirlooms. It means only that by virtue of valuing particular valuable things, we have reasons for action that go beyond the reasons that we and others may have solely in virtue of the intrinsic value of those.
These points about the relation between valuing and reasons for action are relevant to Cohen's defensive of conservatism, because in general, we cannot value things that do not exist and have never existed in the way we value existing things. Valuing involves attachment, attachment requires acquaintance, and non-existence makes the relevant form of acquaintance impossible.
So for example, one cannot value the friendships one has not yet formed in the way that one values one's existing friendships. One cannot value the projects one will someday develop in the way one values the projects one already has. One can not value the children one has not yet conceived in the way that one values ones existing children, nor can one now value the great works of art that artists will produce in the future in the way one values those great works that already exist. One can of course attach value to one's prospects and plans before they have borne fruit and to one's hopes and dreams before they have been fulfilled. But in these cases, the prospects and plans and hopes and dreams already exist, one cannot in the same way, attach value to the plans one has not yet made or the dreams one does not yet have.
2.5 Particular things that we value are sources of distinctive reasons for action and distinctive patterns of emotional engagement.
If this point is correct, then it's possible to identify a conservative attitude more or less along the lines suggested by Cohen's discussion, that goes beyond a temporarily neutral assignment of priority to the bearers of value over the value that they bear.
This form of conservatism includes, in addition, a bias in favor of certain particular bearers of value that already exist. The bias derives from the fact that we can form value-based attachments to existing things in a way that we cannot form such attachments to things that do not yet exist. And things that we value are sources of distinctive reasons for action and distinctive patterns of emotional engagement.
The content of these reasons and the contours of these patterns of engagement will vary depending on the type of thing that's in question. But in most cases, the reasons will include reasons to care for and preserve the things that we value. And the emotions will include vulnerability to feelings of distress if those things are harmed or damaged or destroyed. In so far as this sort of bias is built into our valuing attitudes, every valuer must, to that extent, possess the conservative disposition. This vindicates Cohen's assertion that everyone who is sane has something of this disposition.
2.5 Conservatism is compatible with change, the creation of new things.
At the same time, it's important not to exaggerate or misinterpret the normative significance of this disposition. Although we have special reasons for action pertaining to items that we already value, these reasons will not always be the strongest reasons we have in any given case. They may be outweighed by sufficiently strong reasons of other kinds. Furthermore, there will be many cases in which we can create new items of value without neglecting the reasons we have to care for the items we already value.
So conservatism is not incompatible with creativity. This is important because even if all sane people have something of the conservative disposition, all sane people also have something of the creative disposition.
This disposition is not limited to artists or to others who are colloquially described as creative people. It reveals itself in the impulse to make, to build, to invent, to change, to improve, to reform, to renew, to innovate and of course, to procreate. It reveals itself even in the impulse to act because each act is a novel intervention in the world, each act contributes something new to the course of human history.
In that sense, the conservative disposition to sustain and preserve the things that we value is itself a creative disposition. To be sure there are times when all it requires of us is that we refrain from performing actions that would harm or destroy those things. But there are also times when we must take affirmative steps, often requiring great imagination and tenacity, if we're to succeed in sustaining and preserving the things that we value—after all conservators are not people whose job is simply to do nothing. The conservative disposition and the creative disposition are not incompatible then not only because there are cases in which one of them applies and the other doesn't, but also because there's a sense in which the conservative disposition properly understood is itself a creative disposition.
2.6 The conservative disposition to sustain and preserve the things that we value underwrites our concern for the future of humanity. It is not a backward-looking impulse that competes with or inhibits this concern.
This brings us back to the role of the conservative disposition in supporting our concern for the survival and flourishing of future generations. I've emphasized the extent to which our reasons to care about the fate of our successors are rooted in our value-based attachments to humanity and to the many different forms of human activity and endeavor that we cherish. Far from being a backward looking impulse that competes with, or inhibits a concern for the future of humanity, our conservative disposition to sustain and preserve the things that we value itself underwrites that concern. Nor does the fact that our concern for future generations depends on this conservative disposition, mean that it's incompatible with our creative impulses. To see this one has only to reflect on the creativity and imagination that will be required to overcome the challenges to human survival and secure the prospects of a decent future for our successors.
Moreover, since human beings are in essentially creative species, whose history has a history of change, experimentation and innovation, and who are always developing new modes of living and new dimensions of value, a concern to ensure the future of humanity is itself a concern to sustain the open-ended and unpredictable course of human creative activity.
As applied to the future of humanity, in other words, the conservative disposition is a disposition to ensure that human creativity and innovation will continue to flourish. Despite the ways in which the conservative disposition supports rather than competes with a concern for the future, skeptics may deny that the disposition is rational. Insofar as it gives existing valuable things priority over valuable things that do not yet exist, it may be said to amount to an irrational form of status quo bias. Although I'm sure there is such a thing as irrational status quo bias, I have a difficult time seeing that there's anything irrational about the conservative disposition as I've described it. That's because I have a difficult time seeing what the alternative to it might be.
2.8 We value through time. We should be cautious about assuming that every manifestation of temporal bias in our valuing attitudes is irrational.
The conservative disposition reflects the fact that our value-based attachments can only be directed at what is or has been actual. We could not have a temporarily neutral disposition to form attachments to things that do not yet exist in the same way that we do to existing things. What would it mean to be just as attached to our future friends or to the children we will one day have, or to the great paintings that will someday be produced or the great novels that will someday be written as we are to our actual friends or children, or to the great paintings in novels that have already been produced boost. When it comes to attachment, temporal neutrality is not an option.
[...]
An alternate normative suggestion might be that attachment is always irrational. We should strive to realize an ideal of detachment and to free ourselves as far as possible from all of our attachments. Whatever may be said for or against such an ideal of detachment, however, it does not support the idea that our bias toward existing attachments in particular is irrational.
Instead what's alleged to be irrational is attachment itself rather than the temporal sensitivity of our disposition to form attachments. It's true that if all attachments are irrational, then it follows trivially that a temporally sensitive pattern of attachments is irrational. But if all attachments are rational, then it also follows trivially that a temporally neutral pattern of attachments would be irrational.
The ideal of detachment does not show that there's anything irrational about temporal sensitivity, per se.
[...]
In general, the interactions between our values and our attitudes toward time are complex. And we should be cautious about assuming that every manifestation of temporal bias in our valuing attitudes must be irrational. Indeed, to the extent that the very term bias suggests irrationality or lack of justification, it's, undiscriminating use to refer to all forms of temporal preferences is unfortunate.
Our values and desires are shaped by our self-understanding as temporarily extended creatures and by our experience of temporality. We would not have the values we have if we did not understand the temporal dimension of our lives and the ways that we do. And the direction of influence also runs the other way: the values that we form serve in turn to shape our attitudes toward time. We would not have the temporal attitudes that we have, if we did not have the values that we do.
We need to try to understand these reciprocal influences and not to assume that every manifestation of temporal bias in our valuing attitudes is irrational. As with studies of rational judgment and decision-making in other areas, the trick is to navigate between the complacent assumption that our ordinary thinking must be in good order and the revisionist application of oversimplified models that lack any authority over our actual practices and tendencies of thought.
2.9 Our concern for the future of humanity and for the flourishing of future generations depends on a conservative disposition that applies directly only to presently existing and past bearers of value.
I've tried to illustrate these broad themes by showing how a conservative disposition to sustain existing bearers of value, which some might take to involve a form of irrational status quo bias, is built into our valuing attitudes and cooperates rather than competes with a concern for the future.
The conservative disposition strongly supports a concern for the survival of humanity and the flourishing of future generations. To state my view in a way that is only superficially paradoxical: our concern for the future of humanity and for the flourishing of future generations, depends on a conservative disposition that applies directly only to presently existing and past bearers of value.
2.10 We should be careful about adopting the pose of “impartial concern” and “temporal neutrality” too rigidly. Both principles seem attractive at first blush, but for Scheffler they have lost their allure.
And although I've argued that our bias toward the future is limited, I'm skeptical of the neutralist claim that rationality requires us to eliminate or overcome it entirely. I believe this is one of those cases in which when confronted with the complexity of our actual thought, we should be wary of the prescriptive application of a simplified model of rationality that would classify any recalcitrant attitudes as being normatively deficient.
There is of course a well-known parallel between the view that temporal neutrality is the default rational stance, departures from which stand in need of special justification, and the view that impartial beneficence is the default moral stance, departures from which stand in need of special justification.
The first view treats temporal neutralism as presumptively authoritative, and is suspicious of any tendency people may have to be more concerned about what happens at some times than at others. The second view treats an equal concern for the welfare of all people as the presumptively authoritative moral position, and is suspicious of any tendency people may have to attach special value to their relationships with particular people or to be specially concerned about what happens to some people rather than to others. I reject both of these views. I'm comfortable with the thought that our temporal attitudes are complex and that we lack any single master attitude toward time that we are not uniformly biased toward the past or the future or uniformly neutral.
I'm also comfortable with the thought that we have strong value-laden attachments to particular people and projects and relationships, and that these attachments are sources of differential reasons for action and differential forms of emotional vulnerability. This bears on the contrast that I mentioned toward the beginning of this lecture, between two different ways of thinking about questions concerning future generations. As applied to those questions, a combination of temporal and moral neutralism leads more or less directly to the quest for a principle of beneficence that would solve the puzzles of population.
At first glance, such a principal might seem to be the perfect antidote to the kind of temporal parochialism that I discussed in lecture one. I've tried to make clear throughout these lectures, I'm convinced that this solution is illusory and that once one focuses on the rich variety of human values and attachments and on the complexity of our actual attitudes toward time, it begins to lose its charms.
It's tempting to think that once that happens, our reasons for concerning ourselves with the fate of future generations simply drain away. The beneficence based literature tacitly, though no doubt unwittingly, encourages this thought. In these lectures however, I've tried to show the reverse is true.
Appendix 3. My highlights from Hilary Greaves' critical review of Scheffler's book
Scheffler finds (what I have called) the standard account wanting, largely on motivational grounds. This broadly utilitarian account, he says, misleadingly suggests that the reasons to care about future generations are entirely reasons of morality and (further) of moral duty (25-6). Further, on that account, moral motivation is generally based on sympathy (35). In contrast, Scheffler suggests, there are at least four types of reasons to care about future generations that do not have these features. These are reasons of interest, reasons of love, reasons of valuation and reasons of reciprocity. These four categories of additional reasons, he suggests, will together constitute an alternative, distinctively non-utilitarian approach (25) to the question.
[...]
How does Scheffler’s four-part account relate to the standard [utilitarian] story? One contrast is that while the standard story is straightforwardly normative, Scheffler’s account is in one clear sense descriptive. He is not pointing out (as the standard account does) that the existence and fate of future generations matters “from the point of view of the universe”, and drawing the conclusion that they should therefore also matter to us. Rather, he is rather pointing out that they do already matter to us.
A second possible contrast is that while the reasons involved in the standard story are clearly ones of morality, Scheffler’s reasons are apparently supposed not to be (except perhaps reasons of reciprocity). This issue looms largest in the context of reasons of interest. On one very natural reading, reasons of interest are reasons of prudence, rather than of morality. We have self-interested (that is, prudential) reasons to want to live valuable lives, and thus the existence of future generations is instrumentally important to us in prudential terms. On this prudential reading, paying any significant attention to reasons of interest would be somewhat grotesque. There in the balance hangs an astronomical amount of human welfare: the happiness of quadrillions of future persons, and even the question of whether any of them gets a chance to experience the joys, loves, projects that life has offered us. But never mind them – I (rightly) feel better about my own life, as I cycle to work, as I hang out the laundry, if I (correctly) believe that future generations will exist and/or that my activities are benefitting them, and therein lies a significant part of my reason to do anything about it.
Scheffler is of course aware of the prudential reading of “reasons of interest”. Perhaps because of its grotesque aspect, he goes to some length to disavow that reading (53 ff.), framing it as a common misreading of his previous book “Death and the Afterlife” (Samuel Scheffler, Death and the Afterlife [New York: Oxford University Press, 2013]). Instead, he writes, the intended account is that “[w]e have an interest in [the] survival [of future generations] in part because they matter to us; they do not matter to us solely because we have an interest in their survival” (57). I find this reply a bit obscure. Insofar as I can understand it, however, it removes any prospect for reasons of interest to provide any independent grounds for caring about future generations, over and above either those that are involved in reasons of love and valuation, or those that are involved in the standard story. (On the prudential reading, whatever the demerits of the reasons in question, they are at least independent.)
[...]
Overall, Scheffler’s discussion is interesting and helpful, but I fear that its distance from the standard, broadly utilitarian, account has been oversold. Recall what were supposed to be the main two worries with that account: that it made reasons to care exclusively matters of morality and moral duty, and that it adduced only sympathy as a possible source of motivation. On reflection, we can see that these worries are overstated in the first place.
The utilitarian account does involve exclusively moral reasons to care about future generations. It does not follow, though, that the utilitarian account is exclusively about moral duty. What it is more fundamentally about is axiology (a notion towards which Scheffler seems to voice some antipathy (chapter 4)): it centrally involves observations about how enormously much better the history of the world will be if future generations exist and flourish. Coupled with the simple fact (itself fairly widely agreed) that we all have pro tanto moral reason to promote the good, these observations are enough to establish that we have moral reasons, and potentially very powerful ones, to take steps to protect and benefit future generations. It is an optional further step to conclude from that that we have moral obligations to take those steps, and (even if we do take that step) it is an optional tactical decision to choose to emphasise the obligations rather than merely the reasons. Indeed, many utilitarians themselves are not particularly interested in the notion of moral obligation, or (as in the case of scalar utilitarianism) explicitly eschew any such notion.
The reasons offered by Scheffler’s own account seem to fall into three categories (I am not sure which reasons fall into which categories, and one or more categories might be empty). First, some of them might be prudential reasons. This would make them genuinely independent of moral considerations, and perhaps motivationally useful in the case of people who don’t feel a strong pull towards moral concerns; but Scheffler himself vehemently denies this reading. Second, they might not be reasons to care at all: they might simply be observations that we do care, coupled with some more concrete remarks about the nature of our caring (as seems to be the case for “reasons” of love and of valuation). Third, some of them might be moral reasons, although not necessarily matters of obligation. This is perhaps the most charitable reading, but in that case the reasons under discussion are of a piece with those involved in the standard, broadly utilitarian account.
The picture that seems to emerge is: Future generations matter. Because we recognise that and because we are somewhat decent creatures, they also matter to us: we love humanity, we value the things whose continuation would make the lives of future people good (as they have made our lives good), and we take particular satisfaction in activities that protect the existence and interests of future people. Reasons of interest, love and valuation are a real part of the story, but, on this reading, they all stem from the fundamental, broadly utilitarian, evaluative facts. They do not, after all, constitute a distinctively non-utilitarian approach. They need not invite a reading in terms of moral obligation, but neither need the utilitarian account.
3) "Practical issues" with utilitarianism vs. "ontological" concerns with value
I can make sense of the notion of something like "a community of rational agents" or "sentient beings", and I can see why I value principles coming from this notion; but I'm not sure what a POVU can mean. This is not an issue about abstraction per se. (I’m sorry, this is gonna be even more confusing than the previous comments, but I believe this very discussion is entangled in too many things, not just my thoughts.)
First, you have some issues concerning decision theory: I don’t know what sort of agent, preferences and judgments figure in the POVU; also, if the universe is infinite, the POVU may result in nihilistic infinite ethics. There are many proposals to avoid these obstacles, though.
I think the overall issue is that, even if you can make sense of POVU, it’s underspecified – and then you have to choose a more “normal” POV to make sense of it (the “abstract communities” I quoted above).
To see how this is different from “practical concerns”, take Singer’s mom example: I can totally understand that he spends more resources on his mother than on starving kids. On the other hand, I could also understand if he acted as a hardcore utilitarian. I'd find it a bit alien, but still rational and certainly not plain wrong; the same if you told me that someone else, in a different society far away from here, 500 years into the past or the future, had let their elders die to save strangers.
Now let’s do some sci-fi: I'd act very differently if you told me that a society had built a Super AI, the God Emoji, to turn their cosmic endowment into something like the "minimal hedonic unit" - see this SMBC strip. Or, to draw from another SMBC strip, if a society had decided to vanish from the Earth to get into a hedonic simulation. I think this would be a tragedy and a waste. (And that Aaron should declare SMBC comics hors concours for the EA Forum creative prize.) However, I'm not sure the world in My little pony: Friendship is optimal, or the hedonist aliens in Three worlds collide, would be equally a waste - even though I don't want any of that for our descendants.
But I don't think even these examples picture something like "the POV of the universe"; I think they try to capture a conception of what the POV of sentient life, or the POV of all rational beings, could be… But these notions are more “parochial” than philosophers usually admit - they still focus on a community of beings doing the evaluation. If that’s the case, though, you could think about some hard constraints on your population axiology – concerning the “minimal status” of the members of the community I (or any other agent in our decision problem) want to belong to. In some sense, the sci-fi examples above are "wrong" to me: I can be in no "community" with the "pleasure structures" of the God Emoji; and I don't think the "community" I'd form with the hedonist aliens would be optimal.
Maybe I'm being biased… but it's hard for me to avoid something like that when I think about what policies and values I'd want for the longterm future (I guess that’s why we would need some sort of Long Reflexion). I want our descendants to be very different from me, even in ways I'd find strange, just like Aristotle would likely find my values strange… and yet I think of myself (and them) as sharing a path with him, and I believe he could see it this way, too. So I believe Scheffler has a point here: it’s still me doing a good deal of the valuing. I think it's way less conservative than what he thinks, though.