In my view, nothing is stance-independently valuable. Things are only valuable because they are valued.
It is difficult to understand why awareness of moral evaluation changes with cultural changes and, above all, why we can also voluntarily educate ourselves to value things. This happens in the processes of moral conversion.
It seems to me that almost every view of what matters for welfare — hedonism, desire theories, preference views, and objective list theories — misses a lot of what we care about, projects concerns we don’t actually have onto us, or otherwise fails to care about them on our behalves as we (would actually) care about them. In one way or the other, they fall short in empathy.
Ways of caring are the ways by which things can seem good, bad, better or worse to someone. These at least include pleasure, unpleasantness, desires, moral intuitions, moral judgements, conscious preferences, conscious approval and disapproval, conscious goals, and potentially even the dispositions for these. I conceive of radical empathy as taking on all of everyone’s ways of caring, and caring exactly about what they (would actually) care about on their behalf.
Thanks to Ariel Simnegar, Lukas Gloor, Justis Mills, Tori, JackM and Vasco Grilo for helpful feedback. All errors are my own.
Feelings are kinds of subjective appearances: if something feels good to someone, then it seems good to them, in some way. Pleasure is therefore a way by which something can seem good to someone. Desire is another, and so is approval.
Many parents would sacrifice their own hedonic welfare or lives for the welfare of their children. A typical parent may prioritize their child’s welfare over their own pleasure and suffering even at the extremes like torture, based on their sense of moral obligation. This is a way someone can care.
There are many different ways by which things can seem good, bad, better or worse to someone. These include, possibly among others, pleasure (positive conscious affect), unpleasantness (negative conscious affect), desires, moral intuitions, moral judgements, conscious preferences, conscious approval and disapproval, conscious goals, and potentially even the dispositions for these. Ways of caring are the ways things can seem good, bad, better or worse to someone.
I will also use the terms preferences and attitudes generally for ways of caring, because they are standard terms in relevant literature.
I will assume ways of caring are themselves conscious, or also possibly dispositions to care consciously in such ways.[1] As an example for dispositions, someone who is disposed to enjoy — experience pleasure in — playing chess has a kind of dispositional preference or dispositional attitude in favour of playing chess, even if chess is entirely unfamiliar to them.
And doing something to someone in their sleep could be bad according to a dispositional preference of theirs, even if it would never make any difference to their conscious experiences.
Such dispositions seem worth counting to me, and they could ground the worseness of death in one way — deprivationist (Luper, 2021, Timmerman, 2016) — on the view in this piece, because death deprives us of the things we’re disposed to take positive attitudes towards. I will discuss that further in another piece.
No subset of ways of caring seems to me particularly worth singling out morally to the exclusion of others, except possibly all of the explicit and conscious ways of caring over all of the dispositional kinds.
For example, unpleasantness is a way for something to seem bad to someone and for them to call for less of it, but other ways of caring can fulfill that role, like desires, disapproval, goals and moral judgements. Similarly, pleasure is a way for something to seem good to someone and for them to call for more of it, but again other ways of caring can fulfill that role. Arguments for singling out unpleasantness and pleasure to the exclusion of the others aren’t convincing to me, and it seems chauvinistic and inconsiderate to me to exclude other ways of caring.
So, I’m not a hedonist. Hedonism misses a lot of ways we care about things.[2] And it risks entirely excluding potential moral patients without hedonistic states but who nevertheless care about things in other ways, like some potential artificial minds.
Nor am I a desire theorist, concerned only with desires and their satisfaction, if and because desires, as typically construed, don’t include all ways of caring. They exclude pleasure and unpleasantness, and perhaps approval and disapproval. Nor am I an objective list theorist, which would discount the things about which someone cares if they aren’t on some specified list of objective goods and bads.
Instead, I might describe myself as a preferentialist or subjectivist about what matters, so that what’s better is just what’s preferred, or what would be better according to our preferences, attitudes or ways of caring, in general. And again, I use fairly inclusive concepts for these.
This has important implications for what kinds of beings count morally. The ways nonhuman animals care count, including pleasure, suffering and conscious desire, so animals capable of them would count. Artificial minds capable of conscious attitudes like approval and disapproval would count, too, even if they’re incapable of pleasure and suffering in particular.
See also St. Jules, 2024 and Roelofs, 2022 (pdf) for more on ways of caring and moral patienthood, using different terminology.
On the other hand, I’m fairly inclined to count all and only the ways anyone cares about anything terminally, i.e. for its own sake, not also the instrumental ways they care about things. I may go even further and count only their underived (terminal) attitudes, the ones that are wired in, and not derived through reason, and so exclude derived terminal attitudes.[3] We don’t need to take on their mistaken or incomplete beliefs about consequences, or their erroneous deductions about what matters from what they do care about terminally (or in an underived way), and we may anticipate and count what they will care about in the future.[4] I expect that this justifies some paternalism towards children and nonhuman animals, at least, because they are especially prone to ignorance, mistaken beliefs and short-sightedness.[5]
While we often care about our own pleasure and unpleasantness, hedonism misses many of the ways we do care about things, so it falls short of my ideal of radical empathy. But this isn’t the only problem. Pleasure and unpleasantness are ways of caring and may matter to us, but it’s the things we take pleasure in or find unpleasant that pleasure and unpleasantness are about and necessarily matter to us. The view I want to defend is one according to which:
What could this look like?
If a mother and her child are separated and they’re distressed by this, what’s best for them — on account of this distress, at least — is not to drug or fool them until they feel better, it’s to reunite them. They care about each other, not (just) about feeling better.
Their distress is a response to or an evaluation of a mental state — perceptions and beliefs — representing the other’s absence, or failing to positively represent the other’s presence. That mental state is therefore about the other’s absence, and their distress is therefore an attitude about each other’s absence. The possibility for mental states to be about things is called intentionality (Jacob, 2023).
And the same, I would guess, applies when a cow and her calf are distressed by their separation.[6]
As another example, it seems to me that life without play is unfit for a puppy. Puppies are disposed to enjoy play. They take positive attitudes towards aspects of play. It isn’t the pleasure itself whose absence is worse, but the play, in which they would have taken pleasure.
Or, sometimes thinking about horrible suffering in the world makes me sad. I don't think you do me any favour by just taking away my sadness or disposition to respond with it. I don't want to stop feeling sad. What I want is for things to be better for others. My disposition to respond to others’ suffering with sadness is a kind of hedonic preference for things to be better for others. As far as I can tell, although I may be wrong about my own psychology,[7] I don’t have any attitude against my own sadness itself in this case.
Either way, that’s obviously not all that matters to me. And if you genuinely care about others, too, then you also care about things outside your own mental states, in particular what those mental states track, even if fallibly. You wouldn’t knowingly, voluntarily and permanently get into an experience machine (Nozick, 1974, pp.42–43, Nozick, 1989, pp.104–120, excerpts, Buscicchi, 2022, Wikipedia), like a dream or virtual reality that cuts you off from all connection to others and ever helping them, but in which you would still believe — falsely — that you’re interacting with or helping others. At least, you wouldn’t get in just because you know things will later seem better to you in the experience machine than not in it. Ahead of time, you know these appearances will be deceiving, and all of your attempts to help others will actually be fruitless.
Hedonic ways of caring matter, and we could define how much someone cares about something hedonically, in the moment, by the intensity of pleasure or unpleasantness they’re experiencing, or, perhaps, would experience, if their mental states — especially their perceptions and beliefs — accurately tracked what they’re about.
It’s also not just hedonism that’s guilty here of valuing things in ways that don’t match how we care about things. Many desire- and preference-based views will see my frustrated desire as something I’m worse off for having, all else equal, whether or not there is any part of me that takes a negative attitude towards the desire or its state of frustration. They would project onto me concerns that I don’t have. And to that I want respond:
These views — hedonism, most desire- and preference-based views — are not concerned with only what we (would) actually care about on our behalf. When measures of value stop tracking what we actually care about, it looks like Goodhart’s law: using these measures can result in neglecting things we actually care about in favour of things that don’t matter to us, or otherwise making bad tradeoffs on our behalves. That’s not very empathetic.
And then it’s especially perilous when we’re maximizing and risk picking the wrong measure (Karnofsky, 2022, Knutsson, 2021, St. Jules, 2024).
If something matters to you, then it matters to me on your behalf. I just want to take on what everyone actually cares about or would actually care about. This is how I understand radical empathy (Karnofsky, 2017, CEA et al., 2022), taken all the way. Radical empathy is not just a matter of counting anyone with a perspective worth counting, but also counting their perspectives.
The objects of our ways of caring are what we care about. The objects of our attitudes and preferences are what our attitudes and preferences are about, what they’re directed at. Object views, like us, are just concerned with what we care about. Or, more precisely, they only place value on and tie our moral reasons to what we care about, the objects of our attitudes and preferences.
Object views ground our moral reasons in ways that best match our psychologies. For example, following Schroeder (2007) and van Weelden (2019), if Ronnie wants to dance, he doesn’t necessarily also care about having the desire to dance or that dancing will be pleasant. He can just want to dance. We just care about what we care about, not necessarily also about each particular experience we have or each attitude we hold.
Of course, if Ronnie finds dancing pleasant, or finds the sight or thought of dancing pleasant, then that’s a way he cares about dancing. This pleasure is an attitude about dancing, or parts of the experience of dancing, or what dancing accomplishes, say. And again, it’s the dancing that will seem good to him, through his pleasure. He might not care about the pleasure itself.
Since the pleasantness of dancing is an attitude, caring about the pleasure itself is then a second-order attitude, i.e. an attitude about an attitude. But he can just have the first-order attitude, through finding dancing pleasant.
If Ronnie will care about his desire to dance or the pleasure in dancing, then the object view will count that. In general, if someone does (or would) care about some particular experience or attitude, then the object view will be concerned with that experience or attitude, too, on their behalf. There’s no need to separately bake it into the theory, like on other preference-based views and views that directly value pleasure and unpleasantness themselves,[9] and if we do, we’ll fail to match what they care about when they lack these higher-order attitudes.
And on hedonism, we don’t even place value in dancing itself like Ronnie does, only the pleasure (or reduced suffering) from dancing.
There are various arguments for impartiality, some used to defend impartiality in utilitarianism. At least some such arguments can also support the object view. If my particular position in the world isn’t special, then neither seem to be my particular attitudes. And if the fact that no one’s position is special would support weighing everyone’s interests impartially, then so too would the fact that no one’s particular attitudes are special support weighing everyone’s attitudes impartially.[10]
So, rather than projecting my own attitudes onto others, I should just take on their attitudes. To take on others’ attitudes means caring about what they care about on their behalf. This is exactly an object view.
Furthermore, the attitudes that will ever actually exist are special compared to those that will never exist, because ever actually existing is special. So I don’t count the attitudes that would never actually exist.[11]
Together, this gives me impartiality with respect to what would ever actually matter to anyone, i.e. what anyone would ever actually care about. This is an actualist object view. I will illustrate distinctively actualist reasoning in my next piece.
On one interpretation of the object view with my account of ways of caring, we can think of each way of caring as its own moral stance, broadly construed. For example, pleasure is a pro-attitude, an attitude that something is good, or to be promoted. What is pleasant is good according to the pleasure. Disapproval is a con-attitude, an attitude that something is bad, or to be avoided. We should therefore aim for what’s better according to the various such moral stances that would actually be held. This is not a matter of maximizing pleasure and other pro-attitudes or minimizing disapproval and other con-attitudes per se. Instead, it’s about promoting what we’d actually find pleasurable and preventing what we’d disapprove of, and so on.
Of course, many have positive attitudes towards pleasure itself, and for them, we should promote pleasure itself as those attitudes would have us do, all else equal. So for anyone who favours their own pleasure itself, we should also favour their pleasure on their behalf, all else equal. And on behalf of classical utilitarians or others who value pleasure in itself, including pleasure for others, we should promote pleasure in general, all else equal.
This is most naturally a subjectivist and therefore moral antirealist view, because its moral facts are stance-dependent. What’s good, bad, better or worse depends on the specific stances — or attitudes or preferences — that would actually be held.
By contrast, according to moral realism, there are stance-independent moral facts, i.e. moral facts that are independent of our attitudes or preferences. Most axiologies seem to me to be motivated by a question like “What is objectively valuable?” But this seems like the wrong question to me, when “objectively” is understood as “stance-independently”. In my view, nothing is stance-independently valuable. Things are only valuable because they are valued. Things only matter because they matter to someone, because someone cares about them. There’s nowhere else this ‘mattering’ could come from. Ethics is therefore inherently stance-dependent.
Rabinowicz and Österberg (1996) make similar arguments, and use universalized subjectivity to describe their motivating model of the object view.[12]
Still, the interpretation I outlined may be reinterpretable as compatible with stance-independent moral realism: it could be stance-independently true that it’s better to do better (impartially) according to the stances, attitudes or preferences that everyone would actually have.[13][14]
In the next piece, I illustrate how actualist object views can work with a case where I think they do best, contrasting with other views, and draw out some normative implications and implications for cause prioritization.
Some have proposed or defended views according to desires or agency need not involve consciousness to count morally (Kagan, 2019; Goldstein & Kirk-Giannini, 2023; Kammerer, 2022; Delon, 2023; Shulman & Long, 2024 (EA Forum)). On the other hand, I suspect desires are necessarily conscious to at least some minimal degree, under a highly graded account of consciousness (my comment here).
Even if we expanded our concept of hedonist states to include all the ways things (consciously) seem good or bad, this could still miss some ways things seem better or worse. For example, consider a preference-affecting utilitarian who just believes it’s better for a preference to be more satisfied than otherwise if it will exist anyway, but is indifferent to whether any preference comes to exist at all, for its own sake. Then, to such a person, it seems better for preferences to be more satisfied than otherwise, and this appearance is not derived from things seeming good or bad to them.
Someone may care about something in multiple ways. I’d dismiss the instrumental ways, and maybe the derived ways.
I (2024) write briefly on when we can be wrong about what’s best for ourselves:
1. ignoring or being misinformed about consequences or their likelihood,
2. discounting our future attitudes in a way that’s unfair to our future selves, and
3. neglecting or discounting some of the ways we care about things, e.g. ignoring our emotions, or our feelings about certain things.
Lemaire (2017) also writes:
The first type of case is that where one holds a mistaken instrumental belief or a mistaken belief that a certain object specifies a certain more general one. This is the case where I want to drink this glass of liquid because I am thirsty, but ignorant that it contains petrol. Are stringent actualists forced to concede that drinking petrol is good? Certainly not. The Stringent Critical Actualist can claim, as other actualists have, that an object is good for someone only if this individual has an underived or intrinsic desire or pro-attitude toward this object.[16] So, it is easy to deal with these cases insofar as the person in question has no underived desire to drink petrol, but rather an underived desire to drink something that will quench his or her thirst, and this excludes petrol.
This response is, however, insufficient when we consider the second type of case, the case that involves a deeply erroneous pro-attitude because it is far from obvious that such a desire is derived.[17] Let’s go back to the young man who desires to become a philosopher, although the desire relies on an erroneous, partial, or abstract but well-entrenched understanding of this career, and who would be disappointed if he were to fulfil his pro-attitude. Is the object of this pro-attitude good for the young man? According to Stringent Critical Actualism, it is, and I think that this is the correct response. It may even be an important good that presently contributes heavily to the well-being of his life. He thinks a lot about it, is excited by the prospect when he discusses philosophical topics, and he takes pleasure in viewing himself as a philosopher in the future. Hence, I believe that insofar as being a philosopher is a good for him from which he presently benefits, this good may demand some attention from us. On the other hand, Stringent Critical Actualism can appeal to temporal considerations: if it is true that he will be disappointed by such a career and will thus not desire it anymore if he were to become a philosopher, then being a philosopher will not be good for him at some point in the future. Hence, the balance of his present and probable future goods may justify actions intended to discourage him, maybe by indirectly allowing him to discover on his own what the job that he presently values so much would truly be like, or by offering alternative possibilities whose value might be more lasting.
It might also allow for some paternalism towards the average adult, e.g. through compulsory savings, compulsory health insurance, sin taxes or regulations on goods and services. However, the potential costs, harms and risks of paternalism, direct and indirect, should also be considered and weighed. I won’t take a position here on whether any such state paternalism would be all-things-considered justified for the average adult in practice.
Videos of cows-calf separations and reunions:
https://www.youtube.com/watch?v=OOXFm2-wty4
https://www.youtube.com/watch?v=wV92bw6Np24
Perhaps all unpleasantness is aversive, so induces a desire about and against the unpleasantness itself.
Just being concerned with what we care about doesn’t give a complete view. There are still, for example, potential tradeoffs between what we care about, based on how much we care about them, say.
Experientalist views, including hedonism, place value directly on experiences or components of experiences, like pleasure and unpleasantness.
Bykvist (2024) characterizes three views as follows, if interpreted as accounts of well-being:
We can distinguish the first-order wellbeing accounts neatly by what they would say about the (schematic) fact that you favour x and x obtains. According to the attitude-version, your favouring of x is good for you, because it is satisfied (x obtains). According to the object-version, x is good for you, because you favour x (assuming no other attitudes are directed at x). Finally, according to the satisfaction-version, the complex state of affairs that you favour x and x obtains is good for you, because it is an instance of a satisfaction (a combination of a favouring together with the obtaining of its object).
Garcia and Green Werkmäster (2018) negatively characterize object views — also called subjectivism — as follows:
Objects only have value in so far as they are targeted by attitudes, but neither attitudes nor attitude-satisfaction need have value.
Features are only value-makers in so far as they are the features for which objects are targeted by attitudes, but neither attitudes nor attitude-satisfaction need be value-makers.
By contrast, on attitude and satisfaction views, the attitudes or attitude-satisfaction themselves have value or are value-makers, either on their own on attitude views, or in conjunction with their objects, on satisfaction views.
Or, more precisely, I want to say neither my position nor attitudes are special just for being mine, and for many other specific reasons. They could be special for some reasons, e.g. the intensity of an attitude, like if I find something more unpleasant, it matters more than if I find it less unpleasant. In other words, I care more. If someone cares more about X than someone else cares about Y, then I should care more about X than Y on their behalf. Perhaps intensity is the only good reason to treat any (terminal, or underived) attitude differently from others.
Granted, my position and attitudes are in some sense special to me precisely because they’re mine, and these are my ethical views. Someone could reject either kind of impartiality on such grounds.
Furthermore, someone might think their own attitudes are special because they’re better informed or more rationally derived.
However, I think we hold our underived attitudes, like what we find pleasant, unpleasant, attractive or aversive, or that we care about our loved ones and their welfare, for reasons that are often “wired in” neurologically, or cognitively impenetrable, not based on reason. We just have them, and it’s from them we derive our other attitudes. I don’t think it makes sense to hold them to standards of how well-informed or rational they are, or to weigh some more than others on such grounds. For these reasons and others, I’m inclined to only count underived attitudes.
We can also have derived attitudes about the same things about which we have underived attitudes, of course.
The distinction between would-be actual existence and actual existence seems to me to be an epistemic one. We don’t always know which attitudes will actually exist before we act, and we don’t know how we’ll act before we act, but we can do our best to anticipate them, and judge based on our expectations of the attitudes that will be held, i.e. the ones that would actually exist.
Rabinowicz and Österberg (1996) distinguish two frameworks arriving at the satisfaction and object views, respectively, and their remarks about the satisfaction version apply quite generally to most utilitarian views and most axiologies, including experientialist views and attitude views:
To the satisfaction and the object interpretations of the preference-based conception of value correspond, we believe, two different ways of viewing utilitarianism: the spectator and the participant models.
According to the former, the utilitarian attitude is embodied in an impartial benevolent spectator, who evaluates the situation objectively and from the 'outside'. An ordinary person can approximate this attitude by detaching himself from his personal engagement in the situation. (...)
The participant model, on the other hand, puts forward as a utilitarian ideal an attitude of emotional participation in other people's projects: the situation is to be viewed from 'within', not just from my own perspective, but also from the others' points of view. The participant model assumes that, instead of distancing myself from my particular position in the world, I identify with other subjects: what it recommends is not a detached objectivity but a universalized subjectivity.
And:
the object interpretation presupposes a subjectivist (or 'projectivist') theory of value. Values are not part of the mind-independent world but something that we project upon the world, or — more precisely — upon the whole set of possible worlds. In this sense, our intrinsic value claims, while not world-bound in their range of application, constitute an expression of a particular world-bound perspective: the perspective determined by the preferences we actually have.
Thanks to Ariel Simnegar for pointing this out.
I personally have little sympathy for stance-independent moral realism, even framed this way. Why would this kind of moral fact hold independently of my stance towards it? Rather, it is just my stance that it’s better to do better according to everyone’s stances, attitudes, preferences or what they care about. And I am inclined towards impartiality with respect to what would actually be cared about. This is just a psychological fact about me.
Thanks for writing this, it's very interesting.
This sounds similar to Christine Korsgaard's (Kantian) view on value, where things only matter because they matter to sentient beings (people, to Kant). I think I was primed to notice this because I remember you had some great comments on my interview with her from four years ago.
Quoting her:
I guess "utilitarianism" above could be replaced with "hedonism" etc. and it would sort of match your writing that hedonism etc. is "guilty [...] of valuing things in ways that don’t match how we care about things". Anyway, she discusses this view in much greater detail in Fellow Creatures.
Fyi, the latter two of these links are broken.
Thanks Erich!
Yes, it is pretty close to Korsgaard! I think I actually had Korsgaard in mind and might have checked some of her shorter pieces while working on this, although "object views" ended up being what I actually wanted here. Also, there's this piece by Jeff Sebo explaining Korsgaard's constructivism.
Fixed! Thanks.
I’m not completely sure I would call your view constructivist, because of this comment by Sebo under the same piece.
Also, here’s a random thought, which I don’t necessarily think works/holds for your view, but I’m curious what you think. I think objective tends to mean, as Huemer puts it in Ethical Intuitionism, constitutively independent of the attitudes of observers specifically, rather than anyone’s attitudes or stances. For example, a preference utilitarian can think there is an objective moral fact that it is bad to, all else equal, do something to someone that they disprefer, even though the “badness” comes from their dispreference. It seems like that would be subjective only if the claim was that it was bad according to some observer. But I don’t know if that means your view accepts objective or stance-independent moral facts.
I'm personally not sympathetic to such a claim. What makes it objective? Rather, to me, it's just bad to the person who disprefers it (and possibly other individuals). They are an observer. They are observers of their own mental states and things in the world, and they have attitudes about them.
The view I describe in this piece could be made objective in the way you describe, though.