Thursday, March 18, 2021

Three Dogmas of Utilitarianism

I think that something very close to utilitarianism is the right moral theory, and most of the standard objections are bunk.  But here are what I take to be three genuine flaws in "orthodox" utilitarianism. (Two can be fixed from within utilitarianism.  One pushes us to accept a slightly different consequentialist view that is no longer strictly speaking utilitarian.)

(1) Confusing value with what's valuable.

Consider Norcross' "Act Relevance" principle (MBD, p.4):

Intrinsic value provides intrinsic reasons for action. That one outcome contains more intrinsic goodness than another is, or at least provides, a reason to act in such a way that the former rather than the latter occurs.

I think this is subtly mistaken.  It gives the right extensional verdicts (about how much reason there is for various acts), but for the wrong reason.  The moral reasons stem not from the value-facts themselves, but rather from the value-makers -- the underlying features in virtue of which the outcome contains more value.

This is really important for both normative- and meta-ethical reasons.  Normatively: it's important that our reasons to help individuals stem from those concrete individuals themselves, and not just from abstract value-facts. Otherwise, you end up vulnerable to the sorts of "anti-theory" objections raised by Stocker, Williams, etc.  But those objections can be entirely de-fanged by getting this distinction right, at no cost to our core consequentialist commitments.  Or so I argue in 'The Right Wrong-Makers' (summarized here).

Meta-ethically, it makes clear how value-facts could be non-natural even though what we ought to respond to (what gives us reasons) are various natural features of a situation: people's pain and suffering, for example.  Values mark how important various natural features are, and how strongly we should respond to them.  So if one outcome contains more value than another, that indicates that we have more reason to prefer the former, but it's the underlying value-makers that provide the reason.  See my 'Why Care About Non-Natural Reasons?' for more detail.

(2) Neglecting Fittingness

Orthodox (e.g. "global") consequentialists insist that the value facts exhaust the normative facts.  So any normative assessment of character, virtue, or blameworthiness must ultimately reduce to an assessment of whether the trait or attitude in question serves to promote value.  But this is just clearly wrong, as I argue in 'Consequentialism: Core and Expansion' (in progress):

Suppose that an evil demon will torture everyone for eternity unless you come to maliciously desire others' suffering for its own sake.  How should we assess such a desire?  In the circumstances, it'd obviously be a good (utility-promoting) thing were you to acquire the malicious desire.  If you had a magic pill that would instill in you the needed desire, you should certainly go ahead and swallow it. That much is clear. Even so, there is surely some sense in which it's bad, or at least misguided (in light of the consequentialist's value theory), to maliciously desire that others suffer. It is unfitting, even vicious, to view harms in such a positive light.  Granted, preventing eternal torture is more important than avoiding personal vice, so a good person would take the pill anyway.  But a clear-eyed view of the situation requires us to recognize that they would be sacrificing something--their virtue--in doing so.


Similar remarks apply to praise- and blame-worthiness. Prominent consequentialists, from Sidgwick to Norcross, have insisted that "blameworthy" can only mean expedient to blame. But of course that isn't what it means.  We know from other judgment-sensitive attitudes that rational warrant can be distinguished from expediency. A belief is credible when it is well-supported by epistemic reasons, not practical ones. Similarly for our reactive attitudes: we can distinguish the expediency of blaming someone from the question of whether they truly merit negative reactive attitudes (perhaps for demonstrating ill will--counting some for less than one, in violation of utilitarian principle).

This is not in any way to abandon consequentialism on a normative level, but just to recognize that purist orthodoxy is unnecessarily conceptually blinkered. There are other concepts we can use, other thoughts worth expressing, and other questions deserving of our inquiry, besides those to which consequentialists have traditionally limited themselves.  We may even offer distinctively consequentialist answers to these new questions.  For example, consequentialists may push back against the common assumption that there is anything vicious, disrespectful, or ill-willed about pushing someone in front of a trolley, if the act was done from entirely beneficent motivations.

(3) Treating all interests as innocent

Finally, as noted in my previous post, it's a mistake to count all interests equally.  We should count innocent interests equally, but others may be liable to discounting.  So this calls for a genuine rejection of utilitarianism per se: though the resulting consequentialist view remains a very close cousin.

(One could depart further by additionally rejecting impartiality, but as I've previously suggested, I think one could reasonably go either way on that question.)

Any fellow-travellers inclined to join me in consequentialist heterodoxy?  Or, for non-consequentialists, what "orthodoxies" of your preferred moral tradition do you reject, and why?


  1. Here to register solidarity with "heterodox consequentialism," and especially with point (3). A bit of a shameless plug: I've been working on the view that we should give priority to the interests of the innocent - which I call, "Luck Utilitarianism" - and argue that it allows the consequentialist to provide a better account than they have thus far of the ethics of self-defense. Excited to learn that other card-carrying consequentialists have found the idea compelling.

    1. Sounds good! Feel free to share a link to any papers you have on this if/when they're ready for broader consumption...

  2. "3" seems problematic to me - I can hear Alfred Doolittle explaining he is one of the undeserving poor. And how far retrospectively do we have to look to assess someone's innocence viz arguments one should save the person with shortest life expectancy given their previous bad luck.

    1. There are interesting questions about how best to develop the details of the view. But I don't take the mere existence of such questions to constitute an objection. (That is, I don't yet see any reason to doubt that the questions could be given plausible answers.)

    2. A few quick and sketchy points on this question in case they're helpful:

      A view like the one Richard mentions in (3) might differ from a "Desert-Adjusted Utilitarian" view. For example, many “desert”-based views depend on total, global assessment of character traits. There's less (or perhaps even negative) moral value to promoting the interests of the vicious rather than the virtuous. A view based on responsibility, might differ by focusing on the responsibility for particular acts.

      To illustrate, suppose Moe culpably and recklessly pushes Curly and Larry down a hill, caring not at all about what happened to him. In the process, Moe trips from the force he had to exert. The three of them tumble down. Moe and Curly each have a very painful broken arm. Larry only suffers a minor sprain to his wrist. Now, you have one pain killer that will relieve all of one of the individual's pain. Moe is responsible for the fact that he's worse-off than he otherwise would have been. Curly is not responsible. This comparative fact about responsibility may give us reason to discount the moral value Moe of well-being. Larry, on the other hand, stands to benefit to a lesser degree than Curly since Larry has a less intense injury. Therefore, even though they're both innocent in this case, giving the pain killer to Curly will produce more well-being. On this “Responsibility-Sensitive Utilitarianism,” we ought to give the pain killer to Curly.

      Now, Curly may be equally as vicious as Moe once all character traits are accounted for. If so, the Desert-Adjusted Utilitarian wouldn't agree with the view just mentioned. It would permit giving the pain killer to either Curly or Moe. So part of the answer to the Doolittle case might just be that not all ways of making utility sensitive to responsibility will involve maximally global assessments. Responsibility for one's less-than-optimal prospects in a particular forced-choice scenario may be all that's necessary for discounting in that forced-choice scenario. This still makes for interesting results in a range of cases (Self-Defense cases and so on). Moreover, on this view Doolittle (or perhaps someone with whom it's easier to sympathize) will be equally responsible for his plight, perhaps, as any other innocent party with a longer life span. If both fail to be responsible for whatever plight they face, neither of their interests will be discounted. They're thus relevantly like Curly and Larry. So deviations from the impartiality of utilitarianism only occur when one is responsible to some degree. (Note if no one is responsible for anything because determinism or whatever else, then the view will just be extensionally equivalent to utilitarianism.)


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.)

Note: only a member of this blog may post a comment.