In “Contractualism and Utilitarianism” Scanlon introduces
what he calls “philosophical utilitarianism” (PU). PU is the view that “the only fundamental moral facts are facts
about individual well-being.” PU is
supposed to be answering a different question from the one answered by more
familiar versions of utilitarianism. It
is a view in what Scanlon calls “philosophical ethics,” which means that it is
supposed to explain “why anyone should care” about morality at all; to “make
clearer to us the nature of the reasons that morality does provide.” It’s supposed to do some other things too.
I don’t understand PU, nor do I understand how it could be
thought to do any of the things Scanlon wants a philosophical theory of
morality to do.
What does Scanlon mean when he says that facts about
individual well-being are the “only fundamental moral facts”? Maybe he has in mind some sort of
supervenience thesis. Once you fix all
the facts about well-being, you will have fixed all the moral facts; no change
in moral facts without a change in well-being facts. But that view is crazy. Consider a tiny world with inhabitants A, B and C. Let us fix all the facts about well-being as
follows: A gets 10 units of well-being,
B gets –10, and C gets none. Have we
fixed all the moral facts? No. We don’t know what actions are performed
there. It is consistent with this
distribution of well-being that C behaves rightly by causing A’s well-being, or
that C behaves wrongly by causing B’s misery, or that he does neither. Suppose we fix that C causes A’s
well-being. Now have we fixed the moral
facts? No. We don’t know what C’s alternatives were. Maybe he could have made A even
happier. So we also have to fix what C
could have done instead of causing A’s +10. Now there seem to be at least two other sorts of facts that help
determine the moral facts: who does
what, and what else could have been done instead. Is there some other more plausible way to understand PU, or a
view in the neighborhood that captures what he was getting at?
Supposing we’ve got a better interpretation of PU, how would
PU explain why we should care about morality, or make clearer the nature of the
reasons morality provides? PU itself
says nothing about what people should care about, nor about reasons, or
motivation, or any of the things Scanlon says philosophical ethics is concerned
about. Maybe there’s a sound argument
that takes PU as a premise and that has as a conclusion “everyone ought to care
about morality.” What would that
PU is supposed to be a competitor to intuitionism, which
Scanlon takes to be the conjunction of the following views: (i) “morality is concerned with certain
non-natural properties”; (ii) “we can identify occurrences of” them; (iii) “we
can recognize as self-evident certain general truths about” them; (iv) “they
cannot be further analyzed or explained in terms of other notions.” I can’t see how PU and intuitionism are
competitors at all. Intuitionism, as
defined by Scanlon, consists of a metaphysical claim, two epistemological
claims, and a claim about analysis (the analysis claim is perplexing in its own
way). PU is consistent with all of
Just to sum up, I have the following questions: What is PU? What question could PU plausibly be an answer to? How could PU and intuitionism be
24 Replies to “Philosophical Utilitarianism”
On your first question, about a more plausible way to understand PU, I am thinking that he could say that once one knows all the facts about the well-being consequences of the actions available to me, one knows all that is relevant about what I morally ought to do. The relevance of other factors is screened off by the facts about the upshot for well-being of my possible actions. I feel your other worries more immediately, however.
I don’t know what he does mean, but here’s what he might mean: All and only facts about how S’s X-ing would affect individual well-being constitute moral reasons for S to do X. This might explain why we should care about morality: because it concerns something we have reason to promote. And it makes clearer the nature of the reasons morality provides. They all concern the affects that our actions have on individual well-being. And it differs from intuitionism in being a form of ethical naturalism, reducing normative facts (facts about what we have reasons to do) to certain natural facts (facts about the causal relations that are acts have on individual well-being).
Here is my best guess.
PU: Philosophical Utilitarianism:
1. “A state-of-affairs, x, is intrinsically good to degree d” =df. x consists in some person being overall well-off to degree d; or x consists in some group of people where each person is overall well-off or badly-off to a degree such that, after we take their degree of well-being into account, d represents the aggregate level of well-being.
Behavioral Ethics Component:
2. Whether an action is morally obligatory, morally permissible, or morally forbidden is determined entirely by its consequences for aggregate well-being. All moral reasons for action are generated by the consequences that particular actions would have for the well-being of individuals.
Taken in conjunction with a theory of well-being, my (1), or some very similar thesis, might entail a theory about the fundamental bearers of intrinsic value. But it is stated as a thesis in metaethics: a thesis about how to analyze facts about intrinsic value. Maybe Scanlon construes analysis like this. Alternatively, one could restate (1) as a property-identity or property-constitution claim.
PI: Philosophical Intuitionism
1b. The open sentence, “A state-of-affairs, x, is intrinsically good to degree d,” cannot be defined or analyzed in any psychological or other naturalistic terms. Nonetheless, states-of-affairs are sometimes intrinsically good (bad). Their goodness (badness) is an indefinable, non-natural property, that is detected by rational insight and by no other means. (Perhaps it does not even systematically co-vary with any natural properties. A non-naturalist moral particularist could say so.)
Behavioral Ethical Component (?):
2b. Whether an action is morally obligatory, morally permissible, or morally forbidden is determined entirely by its consequences for total intrinsic value. All moral reasons for action concern the consequences that actions will have for the promotion of intrinsic value.
This is about the best I can do on Scanlon’s behalf. It at least makes the two views incompatible. Also, it seems to me that Mill could be interpreted as as proponent of PU, and that Moore could be interpreted as a proponent of PI.
On the down side, though, neither PU nor PI really seems to explain why we should care about morality. There is a gap in each case between our moral reasons for action and our all-things-considered reasons for action. In other words, in both cases, it remains unexplained why we should care about intrinsic value.
Perhaps the person who believes PU could adopt a neo-Humean theory of reasons (or of what we should care about) and then say that properly educated human persons desire to promote their own well-being and also take a sympathetic interest in the well-being of others. (Mill says some things along these lines in Utilitarianism). But I doubt that this move would produce reasons and motivations of the appropriate strengths.
I’d second David’s point: It’s facts about well-being plus causal facts — about alternative actions — that are the supervenience base.
What does Scanlon mean when he says that facts about individual well-being are the “only fundamental moral facts”? Maybe he has in mind some sort of supervenience thesis. Once you fix all the facts about well-being, you will have fixed all the moral facts.
I think it has to be right that F is the set of fundamental moral facts iff. there is no set of facts F’ that is justificatorily more basic than F. So in terms of moral justification, you hit bottom at the facts F about well-being. When asked about why it is good that well-being is improved, there’s no answer. I don’t see how this entails the supervenience thesis.
It does seem to entail that justification must be traced–in one way or another–ultimately to facts about well-being. I guess I have much less trouble seeing how PU serves as an explanation of why we should care about morality. It does seem to incorporate a reasonable answer to why we should be moral (which I’m guessing is behind the question, why care). Why act morally rather than, say, self-interestedly? Well, because we are all better off if all act morally than if none do. That is, it ties morality to one’s life going better in some way. It seems to me that any decent answer to “why be moral” has to do tie morality to well-being in some way. Of course, there are replies to this sort of answer–Hobbes’s Foole has a familiar one–but it is nonetheless a reasonable answer to the question.
OK, good ideas everyone. Thanks for the input. I have some worries. David suggests, and Robert seconds, that PU is the view that once you know the well-being consequences of all your alternatives, you know what you morally ought to do. The problem with this suggestion is that if it is right, PU is entailed by regular old utilitarianism. PU is just regular utilitarianism minus any specifics about exactly how the facts about the alternatives determine which one is obligatory. And if that’s so, then whatever explanatory work is supposed to be done by PU could be done by regular old utilitarianism. But PU is supposed to be an entirely different sort of view, answering a different question.
Doug, I have a similar worry about your view. If PU is the view you suggest, then regular old utilitarianism entails PU, if we suppose that moral rightness provides moral reasons. So there’s the same worry: what explanatory work is done by PU? One other thing: suppose I think that well-being is “non-natural”. Does that mean I can’t hold PU? I wouldn’t have thought that this would make a difference.
Jason, I’m pretty sure your (1) cannot be what Scanlon has in mind, because he explicitly says that philosophical accounts of morality are not accounts of the meanings of terms. And your (2) seems like David’s interpretation, so I have the same worry there.
Mike, is there a difference between the account you suggest and the view that well-being is the only intrinsic good?
I guess I’m baffled by your bafflement. PU looks pretty straightforward to me: It’s a metaethical view. Utilitarianism, by contrast, is a normative view. Those two views explain different things. The latter explains which acts are right and which wrong, for instance. The former tells us what we’re talking about when we talk about which acts are right and which wrong. And it is the job of the former to make sure that when it does tell us what we’re talking about, that it is something worth talking about –that is, that what we’re talking about matters. Why isn’t it as simple as that? Or am I just dumb as a bag of hammers? (I guess that’s an inclusive ‘or’.)
You write, “Doug, I have a similar worry about your view. If PU is the view you suggest, then regular old utilitarianism entails PU, if we suppose that moral rightness provides moral reasons.”
I don’t see that U entails PU even if we suppose that moral rightness provides moral reasons. PU entails that I have a moral reason to do y even if y is wrong provided that y has some positive effect on someone’s well-being. U does not say this only on the assumption that moral rightness provides moral reasons. Maybe you have some other assumption in mind: the right-making and wrong-making features of actions provide moral reasons. But, in any case, to say that U and some assumption entails PU is not the same as saying that U entails PU. So PU is distinct from U. As Robert might put it, PU is a metaethical view, a view about reasons for action, whereas U is a normative view, a view about the right-making and wrong-making features of actions.
Robert, the view David stated, that you said was PU, was this: “once one knows all the facts about the well-being consequences of the actions available to me, one knows all that is relevant about what I morally ought to do.” Is this really a view in metaethics? It doesn’t tell us *what we’re talking about* when we say an act is right or wrong. It tells us, in non-specific terms, what features of an act are its right-making features. It seems like a view in normative ethics. (Since I seem to be the only one who is baffled, it seems more likely that I’m the dumb one here.)
Doug, you might be right, I have to think more about your proposal. But it doesn’t seem right to say that, since PU is about reasons for action, it is a metaethical view. This view:
x is a reason to do A only if x is a fact about well-being
seems like a view in the normative ethics of behavior, not metaethics. Am I wrong about this? I can’t always tell whether a view is metaethical or normative.
I have trouble distinguishing normative ethics and metaethics myself. But I don’t think that it matters. U is a theory about the deontic statuses of actions, and PU is a theory about reasons for action (or, at least, a theory about a subset of reasons for action). Or to put it another way: U is a criterion of moral rightness, and PU is a criterion of moral reasons. Whether criteria of rightness and criterian of reasons are both subsumed under normative ethics or not, they seem to be criteria for different (although hopefully related) things.
. . . is there a difference between the account you suggest and the view that well-being is the only intrinsic good?
I’m not sure what it means to say that “well-being is the only intrinsic good”. I’d probably say–and this I’m pretty sure is a metaethical claim–that all fundamental moral facts are facts about well-being. Of course specific moral theories might vary wildly on what they claim constitutes one’s well-being. But to urge that it is well-being (however well-being is finally spelled out) that matters ultimately, I think, is a metaethical claim. It is a claim about what any adequate moral theory must treat as basic (viz., the well-being of anything that is morally considerable).
I would have said that PU is vaguer than ordinary util in that, for example, it leaves open the claim that distribution of well-being matters morally. PU limits what matters morally rather than offering an account of what is right an wrong. I guess one might say it purports to place constraints on the true moral theory without attempting to offer an account of what makes an act right or wrong.
You write, “I would have said that PU is vaguer than ordinary util in that, for example, it leaves open the claim that distribution of well-being matters morally.” But if well-being is all that fundamentally matters, then distribution of well-being would only matter insofar as it affects people’s well-being, right? Or are you thinking that facts about well-being include any facts that make essential reference to well-being? In that case, PU could include agent-centered restrictions against, say, an agent’s causing a reduction in one person’s well-being even for the sake of preventing five others from causing reductions in yet five others’ well-being. The fact that your x-ing would cause a reduction in someone’s well-being for the sake of preventing others from doing the same would be a fact about well-being.
Let me add one more thing. If you interpret “facts about well-being” broadly enough, then it seems that almost all plausible moral theories make facts about well-being to be the “only fundamental moral facts.” After all, most moral rules relate to people’s well-being in some way or another. Even in the case of so-called duties to self, abiding by these rules usually tends to promote or protect your well-being or the well-being of others. So I think that we have to interpret “facts about well-being” as facts about how our acts cause increases or diminutions in the well-being of individuals or in the total amount of well-being.
As I am thinking of it, the thought is that facts about well-being screen off from direct moral relevance all non-well-being facts. But how facts about the well-being facts matter is left open by the theory. I think some people refer to this thought as Welfarism. I find it rather controversial and would reject it myself–but of course much hangs on the account of well-being one thinks is correct.
Perhaps this is to say that if I held the view, I would resist treating “facts about well-being” as broadly as I think Doug is suggesting. I think the thought should be not just that facts about well-being are always among the things that matter in each case, but that they are the only thing that matters in all cases.
I don’t think sufficient attention’s been paid to Mike’s point about moral motivation, a point captured in his question “Why act morally rather than, say, self-interestedly?” The appeal to well-being seems, on its face at least, to provide a reasonable answer to this question.
What gets tricky here — if we understand PU as aimed at answering the motivational question — is the question of how PU can tell us why we _should_ care about morality.
In WWOTEO identifies “the problem of explaining moral motivation” as the problem of explaining “the reason-giving and motivating force of judgments of right and wrong” (147). He’s careful to note that his description of this problem is somewhat misleading because it makes it look as if it’s an issue “of understanding how people are motivated rather than an understanding of the reasons they have.” Scanlon’s explicit early on in the book that this focus on reasons instead of desire (motivation) is a change from the original discussion of contractualism in the article.
So I think that now he wouldn’t offer PU as an answer to the normative question (why should I care…), but as telling us why we do in fact care. Of course, he doesn’t think the right answer to this question appeals to well-being either, but rather to “the relation of mutual recognition” implicit in his contractualist formula.
I’m clear that, on your view, PU holds that only facts about well-being matter morally and that “facts about well-being screen off from direct moral relevance all non-well-being facts.” But I’m not clear on what does and doesn’t count as a well-being fact on your view. Until I understand that, I can’t understand what PU tells me to screen off. Now I want to limit well-being facts to facts about increases and diminutions in the well-being of individuals (or in the total amount of individual well-being) caused by the actions of agents. But, apparently, you have a broader interpretation, for you want to allow that a fact about how well-being is distributed counts as a well-being fact. Yet you don’t want to allow that just any fact that makes essential reference to well-being counts as a well-being fact. So what exactly does and doesn’t count as a well-being fact on your view?
You wrote “Yet you don’t want to allow that just any fact that makes essential reference to well-being counts as a well-being fact.”
I didn’t mean to say that. Previously you had offered a very broad interpretation of well-being facts, one according to which any plausible moral theory might accord with PU. You wrote, “After all, most moral rules relate to people’s well-being in some way or another. Even in the case of so-called duties to self, abiding by these rules usually tends to promote or protect your well-being or the well-being of others.”
I took you to be saying that all sensible moral theories would give a significant role to well-being. I was resisting the thought that this should be thought to be enough to satisfy PU.
I am thinking that if a view says that what makes an action right is that it evenly distributes well-being across the earth, makes the well-being of each person as low as possible, makes the highest moment of well-being on the planet as high as possible, etc. then it is in accord with PU. The true moral theory needs only add well-being facts to produce answers about what is right and wrong, I hear PU as claiming. Perhaps the thing to say is that the only empirical input that the theory allows to be relevant is facts about well-being–something like that. I admit that this is not precise, but it does not seem problematically vague to me.
This is interesting. Shame that I have missed the discussion but here’s few after thoughts.
First, I don’t think that Scanlon had in mind what David suggest, namely this:
“he could say that once one knows all the facts about the well-being consequences of the actions available to me, one knows all that is relevant about what I morally ought to do”
Consider this version of the complaint model that is almost a version of contractuliasm:
One ought to do the act that is such that all other actions would create bigger burdens to some individuals in terms of their individual well-being than this action creates to anyone.
This would satisfy David’s definition of PU but it certainly is not an utilitarian view in any sense.
I think Robert’s suggestion about the difference between normative ethics and metaethics is better, even though his view about metaethics is slightly different than that of most. In my view, Scanlon is characterising two levels of moral theories. So on the second level we have:
(i) Philosophical Utilitarianism
(ii) Philosophical Contractualism
(iii) Philosophical Intuitionism
and so on.
On the first level we have:
(i) deontic principles, i.e., principles with absolute constraints
(ii) Rossian prima facie principles
(iii) the complaint model
(iv) well-being maximising
and so on.
The first level is supposed to be a description of the correct moral deliberation that we ought to adopt to guide our actions.
The second level is supposed to be the justificatory level – an account of why some first-order theory correctly describes how we should morally deliberate.
One point of the paper is that, even though some of the first-order views go naturally with some of the second-order views, it still is possible to pick and mix views from the two levels.
So, take the rossian principles on the first level of theories. One could try to justify these on the ‘philosophical level’ by:
(i) utilitarianism – these principles maximise the aggregate well-being if generally adopted,
(ii) contractualism – these principles are such that other sets of principles would create bigger burdens for individuals,
(iii) intuitionism – these principle track the real right and wrong making normative facts in the world.
You could do similar combinations also from other first-order views.
Sorry for not responding sooner to these latest comments.
David, I don’t think PU is just a more vague version of utilitarianism. If it were, then it would be entailed by (regular, “first-order”) utilitarianism. And then Scanlon would be committed to the following inconsistent triad:
1. Utilitarianism entails PU.
2. PU and intuitionism are mutually inconsistent; at most one of them could be true.
3. Intuitionism and utilitarianism are consistent.
Doug, you may be right in the end that “philosophical theories of morality” are just theories about what sorts of states of affairs count as reasons for action. If so, Scanlon’s way of putting things in that paper is very misleading. And I don’t think intuitionism can plausibly be thought of as a substantive theory of reasons for action. I’m with you in not really caring whether these theories count as metaethical or not.
Mike, the reason I asked about intrinsic value is that your account of PU seemed to boil down to this: when asked why well-being is good, one can not give any further justification. This seems to be just the claim that well-being is intrinsically good. But if that’s what PU is, then PU is just a theory in axiology. It’s a “first-order” theory about value (or “what matters,” if you prefer).
Matt, I’m having a hard time seeing how PU, which seems to be a theory about what sorts of moral facts are most fundamental, could explain why people actually care about morality, which seems to require some sort of psychological explanation. Maybe between the earlier article and the book he has changed his view about what PU is?
Jussi, if your account of what’s going on is correct, my worry is that there is no room in your story for utilitarianism as generally conceived (by utilitarians anyway). It’s not a theory about how we should deliberate, nor is it a theory about why some “first-order” story about deliberation is true. It is a criterion of morally right action. I think you are right that there is some story about “levels” at work, where second-level views are supposed to explain, or justify, or motivate first-order views.
If so, then it seems the premise that must go is that Util entails PU (the thesis that “the only fundamental moral facts are facts about individual well-being.”). But I don’t see how Util could fail to entail that.
We’ve been there already, but – deliberation is something you do, i.e, an action, no? Utilitarianism is an account of which actions are right and which ones wrong as you say. Wouldn’t these two thesis imply directly that utilitarianism is therefore an account of which acts of deliberation are right and which ones wrong? If it is then it is an account of why some first-order stories about deliberation are true and some false.
You get the same conclusion if you look at non-deliberative actions. Accept that utilitarianism is an account of which of those actions are right and which ones wrong. If we think that right implies something about what we ought to do (do utilitarians deny this too?), then you get as an account of the correct deliberation – it is the one that leads to correct actions. That of course is a substantial question but I don’t see why utilitarians would not want to pursue it.
Ben, you say,
when asked why well-being is good, one can not give any further justification. This seems to be just the claim that well-being is intrinsically good
Well-being is not intrinsically valuable. But I’d be prepared to say that, according to PU, what is intrinsically valuable is what positively affects one’s well-being. So whatever well-being amounts to–whether it involves certain positive mental states or preference-satisfaction or considered-preference-satisfaction or the realization of some ideal values–intrinsic value should be understood in terms of individual well-being. And PU taken as a metaethical claim–taken as neutral over competing theories of well-being–makes the general claim that any adequate moral theory must make individual well-being justificatorily fundamental.
Comments are closed.