Choose pain-free utilitarianism

Some of my friends are hedonic utilitarians, or close human approximations (people whose excuse to talk excitedly in bars and on the internet is sometimes hedonic utilitarianism). I am a preference utilitarian, so I would like to talk excitedly on the internet about how they are wrong.

Robert Wiblin sums up a big motivation for hedonic utilitarianism:

“I am hedonic rather than a preference utilitarian because if I were aware of a being that wanted things but had no experiences I would not care about it as its welfare could not be affected”

Something like this seems a common reason. What makes a thing good or bad if not someone experiencing it as good or bad? And how can you consciously experience something as good or bad if it’s not some variation on pleasure and pain? If your wanting chocolate isn’t backed by being pleased by chocolate, why would I want you to have chocolate more than I would want any old unconscious chocolate-getting mechanism to have chocolate? Pleasure and pain are the distinctive qualia that connect normativeness to consciousness and make it all worthwhile.

This must be wrong. Pain at least can have no such importance, as empirically it can be decomposed into a sensation and a desire to not have the sensation. This is demonstrated by the medical condition pain asymbolia and by the effects of morphine for example. In both cases people say that they can still feel the sensation of the pain they had, but they no longer care about it.

To say that the sensation of pain is inherently bad then is no different than to say that the sensation of seeing the color red is inherently bad.  The leftover contender for making pain bad is the preference not to have pain. You may still care only about the sensation of having or fulfilling a preference, and not about preferences that are fulfilled outside of knowledge. The feeling of preferring could still be that sought after sensation inherently imbued with goodness or badness. It must be some variation on preferences though; hedonism’s values are built of them.

22 responses to “Choose pain-free utilitarianism

  1. I see the sensation of pain as (normally) affecting the internal hedonic state, and that internal hedonic state is what’s important. That the connection between pain and a change in hedonic state can be severed doesn’t lead to preferences, or anything else, becoming more important than pleasure.

    To put it another way, I think there’s not just pain and desire not to have pain — I think there’s also a sensation of suffering that usually accompanies pain, and that that is the important part. (I think people with pain asymbolia are just getting location/intensity/type data about their pain, which is enough to make them feel the sensation of pain, but not the suffering it usually leads to).

  2. Didn’t I say that I cared about preferences, but only preferences about experiences?

  3. This would be a good topic for Felicifia – most of us there are hedonists (by coincidence, as far as I can tell).

    To me preferences are either a different way of describing happiness/suffering (note, if you’re going to call pain something which people don’t find in its own right unpleasant then you’re just using the term in a different way to most hedonists), in which case I’ll call it hedonism for history’s sake, or they’re an ill-defined mishmash of ideas that includes but probably isn’t limited to,

    1) Being happier when (i) something that isn’t true becomes true.
    2) Thinking you’d be happier when (i)
    3) Being happier when (ii) you start to believe that something is true that you don’t currently.
    4) Thinking you’d be happier when (ii).
    5) Acting in such a way as to make something likely to become true that currently isn’t.

    All these things also seem to rely on a slightly nebulous ontology/epistemology about truth, too. So in order to show greater value to PU than HU, you first need to show why and how it differs.

    Similarly, I would say that ’empirically it can be decomposed into a sensation and a desire to not have the sensation’ is question begging, at least to those of us who think ‘desire not to have the sensation’ amounts to ‘consider the sensation unpleasant’.

    • I’m not sure it matters which thing the word ‘pain’ refers to. The point stands that what is usually referred to as pain can be had with or without the inclination to avoid it.

      5 is what I understand preference utilitarianism as. It needn’t have anything to do with happiness, except that people try to get happiness to some small extent, and tend to be happy when they get things they want.

      Is ‘considering the sensation unpleasant’ another sensation?

  4. I don’t know how far we can get debating this – ultimately it’s a matter of arbitrary axioms.

    To me the only things that are morally valuable or not are experiences. It follow that satisfying a preference for something to ‘happen’/’not happen’ in the real world, without altering anyone’s experiences, is neutral.

    If you think experiences aren’t all that matter (as you presumably do Katja), then you won’t agree!

    • You have other axioms, such as consistency, logic, and some trust in your intuitions, so I can attack the consistency of this axiom with your intuitions and other beliefs. For instance if you were a hedonist, and you came to agree that pleasure and pain are compound phenomenon with no unique features it will probably disagree with some intuition that your moral ‘axioms’ should be simpler.

  5. Both preference utilitarianism and hedonism tend to be horribly under-specified. What theory of preferences is the preference utilitarianism combined with? What theory of happiness or pleasure is the hedonism combined with?

    Because of that the pain point is interesting and important as it highlights the need for further investigation about *which* qualia are the bearers of value.

    However, taken as a argument against hedonism It is a false dilemma. A hedonist has several options when it comes to morphine pain and pain asymbolia:

    (1) Normal pain is bad, morphine pain isn’t. A pain experience includes (at least) two different types of qualia; a “badness qualia” and a sensory qualia. The former is present only in normal cases of pain and explains why these are bad. The latter is akin to a redness qualia regarding value and is present in both cases.

    (2) Both normal pain and morphine pain is bad. The bad quality is also present in cases of pain asymbolia and morphine treatment. However, since the motivational system of the patient is altered by the drug/condition she neither tries to avoid the pain nor says that she wants to avoid it (although it is bad).

    (3) Normal pain is bad, morphine pain isn’t. A sensation (its qualia) are bad. This is the sensation of having your pain-preference frustrated. It is present in cases of normal pain but not in cases of morphine pain.

    (4) Normal pain is bad, morphine pain isn’t. Having your pain preference consciously frustrated is bad (your alternative)

    Of these only (4) looks similar to (a version) of preference utilitarianism and they are all far from a pure preference theory according to which preference satisfcation simpliciter is the good.

    (Of course, strictly speaking, depending on your theory of pleasure/happiness even a pure preference theory of the good can be counted as hedonism if it is combined with a preference theory of pleasure/happiness.)

    • 1) How do you tell if something is a qualia or not? Most people don’t mention a separate feeling of pain, but associate their aversion with what is present in both cases.

      2) What makes you think pain is bad and pleasure is good rather than vice versa if you can’t trust your aversion to what is bad?

      3) I agree this is plausible – I just said it should include preferences.

      4) I wouldn’t even say it had to be conscious – perhaps just having the preference need be conscious for me to care

      Again, I just said preferences should be an aspect, not that a pure preference theory must result.

  6. I am a Ulitity Monster and I say that you are barking up the wrong tree.

    More seriously, preference utilitarianism seems at first like a good measure of calculating the perceived utility benefits that certain actions provide. However, it seems that that when interpersonal calculations of liberty in non finite goods come in to play, preference utilitarianism falls flat on its face at precisely the same point that all the other derivatives of utilitarianism do. Regardless of whether hedonism or preferential mechanisms are used to calculate the utility of ourselves or others, the same conclusions seems to be arrived at in all but the most straw man like scenarios.

    It seems to me that if utiltarianism is going to have any credibility in multi-party exchanges (and therefore be useful as a social philosophy) it is necessary for someone to start making mash-up albums combining utility with non zero-sumness, and preference theories with something that equates to more than pettiness.

  7. I don’t understand Utility Monster.

  8. Pablo Stafforini

    Let me add just one brief comment to Jesper’s excellent reply. It is true that subjects afflicted by pain asymbolia do not care about the sensations they experience. But these people also describe these sensations as “horrible”. What we have here seems best described as a phenomenon of dissociation between apprehension of the badness of an experience and motivation by its apprehended badness , which falls into the second category of Jesper’s typology. By contrast, the effect morphine and other opioids have on pain seems to be different. In this case, the phenomenal experience itself has changed, in such a way that its badness is no longer present. So these cases should be grouped into Jesper’s first category.

  9. Economists have a good instrumental reasons to prefer preference utilitarian as a way to promote win-win deals. More here.

  10. ‘5 is what I understand preference utilitarianism as.’

    But if you don’t combine 5 with any of the others, or anything similar sounding, it doesn’t necessarily entail consciousness. The sea erodes cliffs – does that mean the sea prefers the cliffs not to be there?

    But as soon as you tie it to consciousness it becomes very hard to distinguish from something sensation-based – you’re back to the definition problem

    To Jesper’s list, I would add option

    5) Withhold judgement of the specific circumstances until you have some evidence of how they feel. Once you have the idea of a ‘badness-qualia’, the question of where it exists becomes an empirical one, not evaluative.

    • It depends how you tie consciousness to 5. You could care about the preference being held consciously, or about consciousness of the outcome, or about the outcome affecting inherently value-laden qualia.

      I don’t think I understand this new 5 – could you elaborate?

  11. Another problem for any view of what constitutes preference satisfaction that can entail non-mental events is that it requires some pretty weird metaphysics.

    For eg that if I hope that somewhere in the cosmos there’s a nebula in the shape of Jebus’s face, it would mean that to a PU the world had got slightly better if such a nebula suddenly formed on the far side of the universe, and slightly worse if it dissipated.

    For an HU the only relevant facts in the case are whether I feel better or worse thinking about the possibility of Our Lord’s cosmic visage :)

    • Is whether the world is good or bad to do with metaphysics?

      • Not sure what you mean by the question. I am not advocating the use of obscure philosophical notions to differentiate between the values of worlds, if that’s what you think – that’s what I think you’re doing.

        For me, the only criterion on which we can differentiate between the quality of different worlds is the mental states they contain. PU either looks like a different way of phrasing the same thing, or has to somehow link the mental states of the world with physical states in some obscure nonphysical way.

  12. Nathan Helm-Burger

    @Utility Monster: I don’t understand what you mean by “mash-up albums”, but I think what you are saying is that Utilitarianism is not a workable social policy-generating theory because of the difficulties in applying it to real world situations. If that’s what you’re saying, I agree, but would qualify this with a “so far”. I think that better computing resources and better understandings of physical manifestations of qualia (e.g. firing patterns in the central nervous system), mean that Utilitarianism may at some point in the not too distant future become quite workable.

    @Sasha: Perhaps I’m not understanding Preference Utilitarianism, but I would expect that the PU would only think the world (i.e. collective mental experiences of living beings) had gotten better if you KNEW that Jebus’s face had materialized. Which, supplies you with a qualia of “goodness”. But here I’m using ‘experience’ to describe preference utilitarianism, so perhaps I’m not getting it yet.

    Now, here’s what I’m thinking about Preference Utilitarianism. It makes sense to think of our preferences being the most important part of “goodness” happening to a person, because our preferences so powerfully define good for us.
    For example, a masochist who gets maximal sexual pleasure from sexual experiences which are painful and humiliating. This is an example of getting mixed signals of “goodness” qualia and “badness” qualia, but the “goodness qualia outweighing the “badness”. It’s considered a mental disorder, because it’s clear that if the “badness” qualia could be removed from the equation, the “goodness” qualia would be that much better. However, in practice, this is difficult to separate out for the person.
    When trying to intuit what “goodness” and “badness” are to us, and apply this to utilitarian computations, it seems that much of our experiences are more powerfully determined by our preferences about them than by some instinctive (?) reaction like “pain” qualia. So, I think that, in applying Utilitarianism in the real world with current technology (i.e. unable to empirically verify “badness” and “goodness” qualia) it makes sense to take a PU approach. Asking someone for their preferences about a situation will give you their most powerful estimate of what their qualia will balance out to (assuming honesty). In contrast, a shallow interpretation of a hedonist utilitarianism could mislead someone into judging “goodness” and “badness” from qualia that are easier to measure empirically such as pain, or even further removed indicators such as social expressions of pleasure or pain (e.g. saying “ow”). Thus, although I think that the fundamental measure of “goodness” must be that “goodness” qualia, PU might work better currently in a practically applied utilitarianism.

    I guess this is kinda like saying that Newton’s laws weren’t quite right, but they are more useful for programming a robot to play pool than general relativity.

  13. By the way Katja, there’s a decent piece on the subject by John Broome at

    You can prob skip sections 1-7 in principle, although since he defines some of his terms en route that might make it difficult to follow.

    I don’t think it settles the question, but it’s the most rigorous response to it I’ve seen written.

    • The key point in the article is the standard one that there’s no canonical way to compare and aggregate the preferences of different individuals (or preferences of competing subsystems or temporal selves of a human, or potential preferences we might turn out to have).

      However, while it is a standard point, Katja has not addressed it..


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s