Tuesday, November 20, 2012

Ethics as What's Worth Caring About

Ethical theories can be seen as attempts to track what is worth caring about.  This thought may naturally seem to support consequentialism, and especially utilitarianism.  After all, the utilitarian can simply say, "I care about people!  I want everyone to be as well-off as possible."  And that seems a pretty attractive goal!  There seems no doubt that people's welfare (and, more broadly, the welfare of sentient beings) is worth caring about.  Is anything else comparably important, or worth caring about?

Take promises.  People sometimes criticize utilitarianism on the grounds that it affords no intrinsic significance to promise-keeping, so utilitarians may be expected to break promises (at least when it's sufficiently clear that it really would be for the best).  But does it really make sense to care about promises more than people?  That sounds terribly perverse!  Promises can be a useful tool for coordination, and hence serving our collective interests.  But when promise-keeping and human welfare diverge, surely it's the latter that really matters.

Or consider objections based on "justice" and "fairness".  Of course, I'm opposed to injustice as I understand it, which generally involves people suffering harms for no good reason.  But if you think of justice and fairness as something opposed to the general welfare -- e.g., giving intrinsic weight to equality, or disproportionately weighting the interests of the worst-off -- then this seems harder to justify in terms of what's worth caring about.  "I care about justice and equality rather than people" sounds like a kind of weird moral fetishism.  Even prioritarianism: "I care about people, and the worst-off most of all" just seems kind of perverse -- doesn't it make more sense to care about all people equally (at least if they're not people you have any special relationship to)?

Various attempts at deontological distinctions -- doing/allowing, intended/foreseen, harming/failing to benefit, etc. -- seem similarly unmotivated when looked at through the lens of what's worth caring about.  "I care about not myself performing such-and-such act types" sounds like a narcissistic concern for one's own moral purity, not an accurate perception of what really matters.

One might reasonably expand consequentialism beyond narrowly welfarist concerns: "I care about people, but also cultural flourishing, natural beauty, and philosophical understanding," sounds pretty reasonable to me.  Likewise, putting an agent-relative twist on one's welfarism: "I care about people, but my wife and kids most of all."  No problem.  But the sorts of thoroughly non-consequentialist concerns discussed above just don't seem much worth caring about.  Do you disagree?

I guess deontologists would want to say that they, too, "care about people", but just have a different understand of what this involves.  They may say that caring about people in the ethically appropriate sense entails respecting their rights to be left alone, rather than promoting their welfare.  But I have trouble understanding why anyone would care about rights independently of welfare.  After all, a person's welfare just is what we should want insofar as we care about them.  Whereas rights are either generally reliable rules or else constitute objectionable status quo bias.  While the former are at least worth attending to, neither are plausibly of fundamental importance.

So, that's how things look from my (admittedly thoroughly consequentialist) perspective.  What do others think?  Can you offer a more sympathetic and compelling story about what deontologists care about?  Or would you instead object to framing ethical debate in terms of what's worth caring about?


  1. I don't know if it really ends up being a more viable or sympathetic story, and it certainly would not be one compelling to a consequentialist, but I suspect a lot of deontologists would think that this equivocates on what is meant by 'caring for people'; consequentialism usually defines 'welfare of people' so abstractly that it's quite consistent with not doing anything about this or that person, or even this or that group of people, or their welfare. Consequentialisms usually care about populations, and subordinate the real individual interests of people to abstract features of the populations of which they are a part; Kantians usually take themselves as caring about individual people, so that any general welfare considerations have to meet the precondition of taking respecting each actual, individual person as what matters as itself mattering non-negotiably. This, in fact, is more or less the Kantian line on consequences already: consequences like welfare are merely means to humanity-in-each-individual-person taken as an end in itself.

  2. ""I care about not myself performing such-and-such act types" sounds like a narcissistic concern for one's own moral purity, not an accurate perception of what really matters."

    This is (so often) the way this particular objection is put. Yet, the objectors (in the relevant papers) deny that an agent's motivating thought is self-reflexive in this way. One may be conscious of one's refusal to be the agent who has to do something, but one is (normally) refusing on the basis of some other motivation, and it is the moral status of that motivation which the utilitarian must attack. Yet, this actual, first-order motivation can easily be the very care and concern for other persons that you allege "leads naturally" to consequentialism. And that, as I understand it, is the problem.

    "Utilitarians will, or course, dispute his right to refuse, but the point is that the agent's affirmation 'not through me' does not, in such a case, express a motivation of the suspect, 'self-indulgent', kind. In itself, it does not represent any motivation at all, and the motivations which can lie behind it include some which are, for various reasons, suspect and some which are not. The reflexivity of the utterance does not represent in itself any suspect motive, but only the self-consciousness of the refusal, however the refusal is motivated." (Williams 1981 ch 3.)

  3. "Consequentialisms usually care about populations, and subordinate the real individual interests of people to abstract features of the populations of which they are a part."

    Brandon, can you provide a concrete example of one of these "abstract features" to which consequentialists (supposedly) subordinate the "real individual interests of people"? As far as I can see, the only features that consequentialists regard as having moral importance are precisely these "real individual interests"; the ones that subordinate the interests of people to "abstract features" (such as "justice", "fairness", "dignity", and so on) are the deontologists! (Of course, these abstract features might be morally significant. But Richard's point is that it's hard to see how the significance of these features could be understood as deriving from a basic concern for sentient beings.)

    1. I'm not sure what a concrete example of an abstract feature would be. But perhaps I can clarify. A non-consequentialist is simply not going to look at most consequentialist discussions and say, "Oh, obviously, these people take people to be the really important thing," however consequentialists interpret what they are doing, because non-consequentialists will interpret 'people are what matter' as 'individual people are what matter'; anyone with deontological leanings is going to say, "Oh, clearly these ethicists regard optimal patterns of preference satisfaction (or whatever) in an overall population as what really matters, to such an extent that they'll sacrifice real, live people if that's what's required to get the optimal patterns."

      Kantians will contrast this with themselves. Kantians use abstractions freely, but the categorical imperative requires that in everything you do you take real, live individuals as what matters: every single autonomous individual capable of acting according to moral law must be treated as an end in himself or herself, period -- no comparison to the population at large is required. A Kantian would deny that all preferences matter that much, but would insist that there are certain things due to each and every single person, because each actual person is what matters.

      In other words, when a consequentialist says, "I think people matter," most people would interpret this as saying aggregate-welfare-of-people-matters. But when a Kantian says, "I think people matter," the natural interpretation of this is each-and-every-single-person-matters. And I suspect most Kantians would be inclined to deny that aggregate welfare is anything other than an abstract feature of a population -- it's what really matters for consequentialists, and while it has to do with people indirectly, people are treated as a means to it, and not as ends in themselves.

    2. Note, though, that my work on the separateness of persons argues that that interpretation of consequentialism is mistaken.

      See also my previous discussion of "collectivism", and how such charges really only make sense regarding Rousseauian views that (in stark contrast to utilitarianism!) see "the group" as something over and above a plurality of "real live individuals".

    3. Well, I don't think it universally is mistaken (and I'm not convinced your collectivism argument works, since one doesn't have to treat the group as something over and above a plurality of real live individuals to be only concerned with individuals abstractly in their relation to the rest of the group), but I agree that there are consequentialisms that can make a vigorous sort of response here, and that the point would need refinement. If we're talking about people mattering, however, the sort of mattering that concerns "distinct intrinsic desires for each person's welfare," is fairly weak and for a Kantian, for instance, not particularly morally relevant; what matters most is not what one desires but what actually takes precedence. And the Kantian, of course, is going to say that if you look at who really makes people take precedence, it's the Kantian, who takes each actual person as an end in himself or herself, thus establishing non-negotiables for how we deal with each person precisely because they are persons, and not the consequentialist, who is usually willing to countenance trade-offs that at least in principle leave nothing non-negotiable for any particular individual person, precisely because achieving an optimum of some sort takes precedence over any particular person.

      Like I said, I don't think this will be compelling to a consequentialist; but I think it's important not to underestimate just how differently a consequentialist and a deontologist will already have to understand what it means to treat people as morally considerable, or as mattering, because precisely what divides them is already what it means for something to matter (or at least really matter in a way robust enough for moral purposes) in the first place.

    4. Of course, to intrinsically desire something just is to treat it as an end in itself, rather than instrumental to any further end. It's simply confused (I argue) to think that countenancing trade-offs between distinct final ends somehow makes them into instruments rather than final ends. It's misleading to say that "achieving an optimum" takes precedence over particular people (as that invites confusions over the general and particular characterizations of utilitarian ends; it is still the particular that we are actually aiming at). What's actually going on is that the interests of other particular people may clash with the interests of this particular person.

      But I take your point (also noted in the main post) that deontologists will have very different ideas about what's involved in caring about people (morally). I'm just dubious about whether those ideas really make sense. And I think that their standard objections to utilitarian conceptions of concern demonstrably fail.

    5. If you're allowing trade-offs, you are clearly treating it as a conditional end, not as an end in itself in anything like the Kantian sense; to be an end in itself just means that actually-taking-it-as-the-primary-end-of-action is not conditional on other things, and thus not subject to any trade-off. Saying otherwise would be to say that moral law consists of imperatives that are simultaneously categorical and hypothetical.

      Your response reminds me of the old joke that a utilitarian is a man who is a friend to everyone and therefore will reluctantly stab any one of them in the back if it makes the rest happy. Yes, utilitarians deal with particular people, and derive their principles from considering particular people, but this is not what we're talking about, if we're talking about people mattering morally (the Kantian would say); what we're talking about is the actual role given to particular people in actual cases of caring-about and taking-things-as-mattering. And the way in which particular people are morally important in actual cases is in consequentialist terms determined almost entirely by their relation to the rest of the population, i.e., how they fit into a population pattern, and not by the simple fact that they are people, as in the Kantian approach. While the joke about utilitarians is just a joke, friendship is actually a good point to raise, because if anything is a paradigmatic case of taking someone to matter, it's friendship; but which looks at first glance more like real friendship: treating someone as an end in himself or herself, and working for their welfare because you always owe it to them (in particular) in some way to do some basic kinds of good to them regardless of the circumstances; or working for someone's welfare, conditional on other people's interests not taking precedence? And is not the latter something that most people would see as treating one's friends as mattering (in anything beyond very weak senses like 'being considered in some way at all') only to the extent that they have the right role in an abstract scheme of interests, i.e., subordinating real, live people to abstract ends?

      Obviously this last is not probative. But I think it does show the problems with simply taking it to be obvious to any non-consequentialist that consequentialism takes people to matter in any substantive way; the mere fact of dealing with people somehow and wanting good for them somehow is not sufficient. A Kantian can at least say, "For any given person, you have non-negotiable duties to that particular person simply because he or she is a person." That seems a pretty straightforward way of taking people to matter, substantively, something a lot like friendship. (And in fact it's an old Kantian view that one way to understand Kantianism is to understand it as requiring that we treat everyone as possibly a friend; which, while not involving all the positive requirements of friendship, does mean treating no one in a way inconsistent with taking them to matter in the way a friend matters.) What does a consequentialist really have to set against that, if someone asks how a consequentialist intends to treat her, herself, as mattering because she is a person? That he won't sell her up the river as long as it's not required for making lots of other people happy.

    6. "What does a consequentialist really have to set against that, if someone asks how a consequentialist intends to treat her, herself, as mattering because she is a person? That he won't sell her up the river as long as it's not required for making lots of other people happy."

      That's a rather one-sided account! :-) Let's add that the consequentialist will always do whatever he can to positively help her, so long as it doesn't impose greater (aggregate) costs on others. It's true that we each must surrender any claim to "inviolability", but that's a small price to pay for all to have done so, such that our interests will still count even when this requires lying to the murderer at the door, etc. It'd clearly be rational, behind the veil of ignorance, to give up such "rights"; their actual effect is just to privilege the powerful at the expense of the more vulnerable, who could be more greatly helped if only one were more willing to impose (smaller) costs on others.

      "subordinating real, live people to abstract ends"

      This is getting repetitive. Again, there's nothing "abstract" about the utilitarian's ends. They are subordinating real, live people to other real, live people (insofar as smaller costs to the former enable greater benefits for the latter). Caring about multiple concrete persons at once doesn't turn these people into abstracta. Being willing to make sensible trade-offs between them doesn't turn people into abstracta. It's misleading rhetoric, and people who want to think clearly shouldn't indulge in it.

  4. "Ethical theories can be seen as attempts to track what is worth caring about." It doesn't seem feasible to answer the question of what is worth caring about without first answering why anything is worth caring about.

    If we must start with the bare question of what, there do seem compelling alternatives for "ultimate values": obtaining the most comprehensively true view of reality; attaining the most harmonious form of civilization.

  5. >Even prioritarianism: "I care about people, and the worst-off most of all" just seems kind of perverse -- doesn't it make more sense to care about all people equally

    I find this mystifying. Claiming about caring about all people equally sounds perverse—at least on examination against intuition—in that nobody says they want to improve the lives of the rich when at the (even minor) expense of the poor. My guess is that "caring about all people equally" is only more cognitively fluent. (See my "Cognitive disfluency: Simpler isn't always better" — http://tinyurl.com/3e9fqcs )

  6. Utilitarianism doesn't care about "people", it cares about a subset of people best suited to producing good consequences. This is because, shockingly, consequentialism cares about consequences, not people.

    Bernard Williams gets it right we he observes that utilitarianism leads to a situation where one part of the population must serve the other. Perhaps utilitarians can avoid this is by saying that we ought to all help other animals. Unfortunately, they surely must accept that we ought to kill all other animals (which alone seems good enough of a reason to reject the plainly false claim of "caring about all").

    By the way, the smugness in this post is almost unbearable: "Why do you hate people non-utilitarians? Is it because you're irrational or are you just a bad person?". This attitude is part of why utilitarians are so unbearable.

    1. "consequentialism cares about consequences, not people"

      That's a very strange thing to say. The actual view is, of course, that we should care about the consequences that an act has for people. (Any old act has some or other "consequences", after all. The question is which consequences count as good or bad, and the way to determine that is to see how well or poorly people are affected.)

      "Utilitarianism doesn't care about "people", it cares about a subset of people best suited to producing good consequences."

      You don't seem to understand the view at all. It fundamentally "cares" about all people equally, and one implication of this is that some people may have greater instrumental value than others, insofar as they are able to help other individual people, who also fundamentally matter.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)