Assault on Utilitarianism: Pedophiles and Utility Monsters

by Aaron Wilson, Guest Blogger on November 16, 2011

Due to feedback on my last post I think it may be beneficial to include one more idea to the list of what I called “the cornerstones of utilitarianism”.
These three cornerstones are:

Premise 1: We all intrinsically value happiness.
Premise 2: Happiness is the only intrinsic value.
Premise 3: We ought to do what will promote intrinsic value.

Taken together, these three claims lead to the utilitarian conclusion that happiness is what makes a thing good and we ought to promote happiness. This post will be a critique of Premise three. Bentham and Mill, the founders of utilitarianism, take premise three as being obviously true. At worst this is an unfounded assumption and at best it is implied that premise three is intuitively obvious or self-evident. For now we will take the more generous interpretation and later we will show that other methods of proving premise three is true also fail.

Since pleasure is intrinsically valued premise three implies that we are to believe that upon realizing we intrsinically value pleasure that it will become intuitively obvious that all pleasure is good. As Bentham explains this, “Now, pleasure is in itself a good: nay, even setting aside immunity from pain, the only good: pain is in itself an evil; and, indeed, without exception, the only evil…” (Bentham 102). It is not at all obvious to me that the pleasure a pedophile gets while lusting after children is in and of itself a good thing, even if he does so without the parent or child knowing. I seek not only to show that pleasure is not intuitively and obviously good, but using a method called “reduction ad absurdum” I think we can show that such utilitarian claims are intuitively and obviously false.

Reductio ad absurdum is a method of disproving something by showing that its inevitable consequences would be absurd. Nothing shows utilitarianism to be more absurd than Robert Nozick’s example of a utility monster. Nozick argues:

“Utilitarian theory is embarrassed by the possibility of utility monsters who get enormously greater gains in utility from any sacrifice of others and these others lose. For, unacceptably, the theory seems to require that we all be sacrificed in the monsters maw, in order to increase total utility.” (Nozick 41)

It is entirely possible that one of more of these utility monsters exist. If we are feeling poetic we could think of the utility monster or utility monsters as Satan or even a whole host of demons. Bentham unknowingly explains why the existence of a utility monster makes his theory seem absurd when he writes, “It is… not from the principle of utility that the most abominable pleasure which the vilest of malefactors ever reaped from his crime would be to be reprobated…” (Bentham 9). Bentham actually believes that the vilest of malefactors always cause more harm to others than the pleasure they receive, but his theory implies that if Satan or his demons received greater pleasure than the suffering we receive from the destruction of our lives then we ought to sacrifice ourselves to the devil. Utilitarian theory makes it very clear we ought to maximize happiness and it is equally clear that the “most abomidable pleasure which the vilest of malefactors ever reaped” makes its way into that utilitarian calculation. I daresay that if your idea of being moral can include sacrificing saints for the delight of demons then your definition is wrong.

Something else to think about:
Other animals clearly experience pleasure and suffering. Mill might argue that these animals experience lower quality pleasures than humans, yet even with that stipulation their pleasure would still correspond to some level of value. If utilitarianism is true, are hungry humans actually utility monsters? Keep in mind that it is physically possible for farms to give animals a high quality of life as well as it being possible to kill animals in ways which do not involve much suffering.

PDF

Tagged as: Fallacy, good, happiness, hedonism, intrinsic, Jeremy Bentham, John Stuart Mill, Morality, Nozick, pleasure, predophile, Robert Nozick, utilitarianism

Mike Norman

“Premise 3: We ought to do what will promote intrinsic value.”

and

“This post will be a critique of Premise three. Bentham and Mill, the
founders of utilitarianism, take premise three as being obviously true.
At worse [sic] this is an unfounded assumption and at best it is implied that
premise three is intuitively obvious or self-evident.”

I may be able to do one better. Premise 3 seems to be a tautology. “Intrinsic values” are those things which ought, intrinsically, be promoted. I think, however, that I agree with you that “happiness” cannot be this intrinsic value. Two caveats are in order about that agreement, however. The first is that those quotes around “happiness” above are scare quotes. This “happiness,” the one which you seem to be using in your argument, is the short-sighted one that seems so similar to the one to which you, especially, seem to run any time you try to make a point about “objective morality.” In the pedophile example you cite, for example, you cite only the pains and pleasures of the two actors, as if they are not enmeshed in a community of persons who will experience various degrees of pleasure and pain as a result of these acts. Further, you omit the pleasure of a regularly ordered society in which people feel safe to let their children run about and play, the pleasures afforded by being safely let out to run about and play, and the pleasures granted by the mere assurances that justice will be done. I’m not here defending utilitarianism, but rather pointing out that, in addition to the legitimate move of reductio ad absurdum, there is also a fallacious way to make that move; One can introduce the absurdity by means of an inappropriate reduction. Your simplified happiness cannot actually capture the essence of the utilitarian’s happiness, even if utilitarianism is actually ultimately flawed.

The second caveat regards your society of demons. Now, this point is related, because no society of demons could actually exist. Society, civilization, is founded on a mutual promise of marked reduction in capricious harm. No demons that “value” capricious harm could then come together in a society. One might as well posit a society of wolverines. However, if we even assume that that kind of malice could be a value, we would then have conceded the entire point! Yes, in that case, it would be good for them to do what they value! Keep in mind that we’re talking about values in the real sense, here, and not the limited straw-man sense. If it really ultimately furthered demonic society to engage in capricious mischief, then we would have to concede that it is right for demons to engage in capricious mischief! But this is the whole point about “intrinsic” value; You can’t just arbitrarily change it to whatever straw-man version suits your rhetorical purposes.

And now it gets truly weird:

“I daresay that if your idea of being moral can include sacrificing
saints for the delight of demons then your definition is wrong.”

and

“If utilitarianism is true, are hungry human’s [sic] actually utility monsters?”

A mere two sentences separate those two quotes! Let me run this down in syllogistic form.

P1: If a moral system validates the sacrifice of innocents to monsters, then it is wrong.
P2: Utilitarianism validates the sacrifice of innocents to monsters.
C: Therefore, utilitarianism is wrong.

However, I suspect that the force of your question about hungry humans was rhetorical, namely that you presume that the answer to the question is that we are not utility monsters. This leads us to the following problematic syllogism:

P1: If a moral system validates the sacrifice of innocents to monsters, then it is wrong.
P2: Whatever Aaron Wilson’s moral system is, it validates the sacrifice of innocents to monsters.
C: Therefore, whatever Aaron Wilson’s moral system is, it is wrong.

By my reading, you seem to be damned whether you embrace utilitarianism or not.
- Anonymous
  
  “This “happiness,” the one which you seem to be using in your argument, is the short-sighted one”
  Actually the happiness I am talking about factors in the long term consequences.
  
  “ In the pedophile example you cite, for example, you cite only the pains and pleasures of the two actors, as if they are not enmeshed in a community of persons who will experience various degrees of pleasure and pain as a result of these acts. ”
  Actually the pedophile I referenced only observes 1 actor. He is lusting without any other parties knowledge, so those parties cannot be effected.
  
  “Now, this point is related, because no society of demons could actually exist. Society, civilization, is founded on a mutual promise of marked reduction in capricious harm”The demons can form a society where they do not harm each other, this does not imply they will choose not to harm other species (such as humans).
  
  “However, I suspect that the force of your question about hungry humans was rhetorical, namely that you presume that the answer to the question is that we are not utility monsters.”
  I was just posing a thought experiment. The question about whether vegetarianism is justified is completely detached from the argument I was making.
  
  “P1: If a moral system validates the sacrifice of innocents to monsters, then it is wrong.
  
  P2: Utilitarianism validates the sacrifice of innocents to monsters.
  C: Therefore, utilitarianism is wrong.”
  
  My real argument is just highlighting the notion that “it is clear that not all pleasure is good”. In other words “The most abomidable pleasure of the vilest malefactor” is not good. Maximizing pleasure or happiness is clearly not the only grounds upon which we deem a thing to be moral or not. If it was, we would be lead to the absurd notion that it is possible that morality include acts like sacrificing humanity for the sadistic pleasure of demons.
  - Mike Norman
    
    “Actually the happiness I am talking about factors in the long term consequences.”
    
    I’ll play along: No, it doesn’t.
    
    See how easy it is to play the denial game, rather than the refutation game? I’d be willing to bet that as soon as you name one long-term consequence of the act of lusting after children, your argument will lose the entirety of its force. I might actually be tempted to admit that this is a matter of faith for me, in that I think our moral intuitions have some usefulness because they are evolved and therefore useful to living beings. I think that you lack this faith because you seem to think that moral principles are merely logically derived or merely revealed. What an irony, eh?! At any rate, some evidence that you’re considering long-term consequences within a social context, or some defense that shows how those consequences are already in what you’ve written, would be appreciated.
    
    “Actually the pedophile I referenced only observes 1 actor.”
    
    Even with respect to that actor, which, by the way, I find to be a truly stupid objection, cultivating that pleasure would have adverse consequences in terms of social isolation. Will nobody really be able to observe a tumescent avuncular type at a playground and not be disturbed? Therefore, even on utilitarian grounds, it can’t be “good.” I’m no longer really surprised that you can’t see this.
    
    “The demons can form a society where they do not harm each other…”
    
    This is a part of the reason that I have trouble with this smarmy “purely logical” mode of reasoning. I should say, by way of minimal politeness and precision, that this is not unique to you, you are not the originator of it, nor are you to be especially condemned for it. Logical reasoning can only ever be the beginning, can only ever be a preliminary. The only things it can ever do are (1) rule out flat contradictions and (2) reveal things that actually were a matter of definition. In this case, for example, it can only show that there is nothing in the definition of a demon that would logically rule them out from harming every class of sentient beings for pleasure, except for themselves. The moment you start to mentally examine a “real” demon, an actual supposition of what they must be like, you see how absurd that really is. How would that appetite be checked when they deal amongst themselves? Would they really gain no pleasure from harming each other? Would that powerful instinct that derives pleasure from harm really be so susceptible to abeyance that they could form a society? We’d have to look at substantial aspects of demons, really as stand-ins for persons in this analogy, to reason about them effectively. When we do, however, we see how absurd it is to rely on this “logical” idea that they could be absolutely ravenously hungry for the infliction of harm in one context and yet have it checked completely in every other context.
    
    “My real argument is just highlighting the notion that ‘it is clear that not all pleasure is good’.”
    
    For better or for worse, we cannot always prescribe “just” what our arguments do or do not do. If, in the course of “just highlighting,” we drag some other inconvenient entailment in, then we are in real trouble. The force of the argument which we use to “highlight” our point must stand or fall with the criteria applied to it, no matter how we find out that the argument fails on those criteria. For example, if my argument that bread pudding is delicious somehow hinged on my assertion that there is a chainsaw-wielding clown standing behind you, and that clown is shown not to be there, then my argument about bread pudding has come unhinged. There’s no use saying, at this point, that my argument was really about bread pudding and not clowns. Pandora’s box is not so easily nor effectively shut.
    
    Your particular arguments about the goodness of pleasures lead to a particular absurdity, one which I called out. What you actually wrote was:
    
    “I daresay that if your idea of being moral can include sacrificing
    saints for the delight of demons then your definition is wrong.”
    
    and
    
    “If utilitarianism is true, are hungry human’s [sic] actually utility monsters?”
    
    Did my unpacking of these statements into syllogisms do them an injustice? If so, please point to it. If not, please explain how the force that your statements have about the correctness of utilitarianism can remain even after they have failed to meet the logical criterion of non-contradiction. And, yes, we are reasoning logically, but precisely in order to rule out a flat contradiction.
    
    I should say that I am pleased that your cornerstones moved a bit. I flatter myself that I had a part in that, and I hope it turns out for the better. It still seems to me that Premise 3 renders Premise 1 superfluous, or vise versa, but maybe I will be pleasantly surprised.
    
    Edit: If you could reply from the web page rather than in email, I think it would fix some of the formatting issues that make your comments so difficult to read. Thanks!
Kevin

Concerning pedophiles, I don’t think this is a good objection. We are concerned with the consequences of having a person go free with this sort of desire. If the hypothetical concerned a person in jail for life, I doubt many would have much reservations against his particular lust. I, for one, would not.

Concerning the utility monster, I think that the hypothetical needs to be clearly outlined. I can think of hypothetical scenarios in which the monster gets more pleasure by eating us than we lose by being eaten, but it is justified by utilitarianism to kill such a monster and others where it might be our duty for sacrifice (but this is only normatively warranted if we consider the monster’s utility function into our decision making process). We might find some of these answers “unintuitive” but we are considering hypothetical scenarios that have no basis in reality, so that is what we would expect. If you asked a random group of people whether it could ever be ethical to walk up to a random person and punch them, I would guess most people would say no. However, if our desires regarding pain were reversed, such an action may be as common and benign as a high five. It may not be intuitive that “hurting” people could be ethically warranted, but that is because our intuition is based on our experiences and those are tied to this reality, not hypothetical realities.
- Anonymous
  
  The utility monster doesn’t have to gain the pleasure from “eating”, it is just any act of destruction. It is possible that satan exists, a being which revels in the most vile of actions. The pleasure he gets from being vile under utilitarianism is “good”. If that pleasure outweighs the suffering of his victims, then maximizing his pleasure = maximizing total pleasure. Therefore the utilitarian must argue that we foster the pleasures of this vilest of malefactors if it turns out he exists.
  - Kevin
    
    Right, if that is the outcome that optimizes well-being, then to someone who cares about the well-being of sentient creatures equally, this is the most preferred course of action. Where is the objection? It’s only unintuitive because it is so far out of touch from reality that it is unimaginable. This conflicts with our value of fairness and loss aversion, so the ratio of pleasure of the utility monster to the others displeasure would need to ridiculously high if this is to be on a long term basis. Such a ratio is unheard of, which is why we can’t fully understand/appreciate the result. We have no real point of comparison.
    
    Its almost like the example of what’s better, a speck in the eye of N people where N is equal to Graham’s number, or one person being tortured continuously for 50 years. It is tempting to say that the torture would be worse. Presumably utilitarianism would favor the torture as causing the least amount of suffering.. The paradox results simply because we can’t even comprehend the consequences of the former option. This is not an objection to utilitarianism, it is simply outlining our limitations in understanding the hypothetical even thought the hypothetical is clearly defined.
    - Anonymous
      
      “Right, if that is the outcome that optimizes well-being, then to someone who cares about the well-being of sentient creatures equally, this is the most preferred course of action. Where is the objection?”
      
      So you are arguing that sacrificing victims to demons for their sadistic pleasure could be moral? The objection is that such a conclusion is absurd.
      
      “ It’s only unintuitive because it is so far out of touch from reality that it is unimaginable.”
      It is quite easy to imagine. If thousands of demons existed and all wanted one man to die, it is quite easy to imagine their pleasure is greater than his suffering. If these demons don’t suffer from diminishing marginal returns (or if there is just a whole bunch of them), that leads us to sacrificing the entire human race. I don’t find sacrificing the entire human race, or even 1 person for the sadistic pleasure of demons to be a morally acceptable conclusion. Any theory which does call this moral is clearly defining “moral” incorrectly.
      - Kevin
        
        You basically said that you don’t like the answer, but that is not the same as saying that the answer is incorrect. You were on the right track when you suggested that we might be utility monsters. Instead of demons, why not have a tribe, and instead of humans, make the victim a whale (whom would be just as conscious, self-aware, and have the propensity to feel pain compared to us as we would be to the utility monsters). Make the tribe unable to get this good (in this case, staving off hunger/feeling full would be the good) from any other means. Is it acceptable to cause pain and misery to the whale for the pleasure of the humans? We don’t find this to be objectionable or absurd, yet when we are not the beneficiaries, we suddenly object; why is that? Does morality have some sort of anthropogenic component that makes homo sapiens special and anything suggesting the contrary mean that you are clearly defining “morality” incorrectly?
        
        Regarding being able to imagine the scenarios, I’m not talking about understanding the parameters of the question. With regards to dust speck vs. torture question, I am fully aware what the hypothetical is proposing. The question posed is very simple, which is preferable, N people get a speck of dust in their eye or one person gets tortured continuously for 50 years. The problem lies in when I try to imagine what it would actually represent. I fail to realize what N specks of dust in people’s eyes would add up to. When it comes to the utility monster, I can’t empathize with a creature that gets N times (or whatever the ratio is) as much pleasure compared to me. Because of this, I can’t appreciate the magnitude of the contribution that the sacrifice is making.
      - Anonymous
        
        Even if you come to the conclusion “humans can eat animals because our pleasure is more important than theirs” it doesn’t imply that “all pleasure is good”. To deal with this utility monster example you must be willing to admit “sacrficing the entire human race for the “sadistic pleasure of demons” is a morally praiseworthy act.
        
        While related, there is a significant difference between humans eating animals, and humans sacrificing themselves to the vilest of malefactors and calling it good.
      - Kevin
        
        While I agree that not all pleasure is “good” in a
        utilitarian sense, since “good” requires a cost-benefit analysis to determine
        whether the pleasure is off-set by displeasure or not; it does not form an
        objection against a utility monster since the hypothetical stipulates that the
        pleasure is not off-set by displeasure.
        Eliminating certain pleasures from the utilitarian calculus would be
        arbitrary since that is how we determine what is good in the first place.
        
        “While related, there is a significant difference between
        humans eating animals, and humans sacrificing themselves to the vilest of
        malefactors and calling it good.”
        
        And from the whale’s perspective, how are we not the vilest
        of malefactors? From their perspective,
        we have done them no benefit and have only caused them harm. This is exactly what the relation between the
        utility monsters and we are supposed to be like. I fail to see the difference. Is it significant because it is humans doing
        the sacrificing? Does morality have an anthropomorphic
        component to it?
        
        “To deal with this utility monster example you must be
        willing to admit “sacrificing the entire human race for the “sadistic
        pleasure of demons” is a morally praiseworthy act.”
        
        Given certain parameters, it is. For example, if they can get this pleasure
        from another source with less negative consequences, then they are morally
        obligated to do so. The human equivalent
        would be to say that whaling is morally wrong except for those who only have
        that option for their food source. You
        need to consider the opportunity costs with each example, but you could
        construct a hypothetical in which the compassionate thing to do is sacrifice
        yourself.
- guest009
  
  You don’t have to think of the utility monster as eating us. In fact, you can make the utility monster more damning by offering that mankind be sacrificed solely for the monsters “sadistic pleasure”. It implies we ought to be sacrificed even for the “most abomidable pleasure from the vilest of malefactors”.
teamawesomepanda

Basically what you
have done is taken examples, such as the paedophile and utility monster, to
show that the basic formula of Utilitarianism, which most people find fairly
intuitive or non absurd leads to counter intuitive or absurd conclusions. This
does not demonstrate that utilitarianism is false.

It seems absurd to me that there are 10 sextillion stars in
the universe when I can only see a few hundred in the night sky. But my belief
that it is absurd doesn’t mean that it is wrong. I think it is absurd that homosexual sex is morally wrong.
Have I refuted Divine Command Theory?

“I daresay that if your idea of being moral can include
sacrificing saints for the delight of demons then your definition is wrong.”"

I think if your idea of being moral can include raping small
children being a morally praiseworthy act because god commanded you to (as
William Lane Craig seems to think) then your definition is wrong. Have I refuted
Divine Command Theory?
4iner2

For the record, I don’t know of any utilitarian who sees a problem with the conclusion of the utility monster argument. I don’t think you’re going to be able to convince many utilitarians to abandon their position just by claiming this is an instance of reductio ad absurdium.

As for your pedophile argument, we can either conclude that

a. utilitarianism is false because it leads to an absurd conclusion.
b. there’s nothing wrong with what the pedophile does.
c. we need a better definition of pleasure (for the purpose of deciding what to maximize in utilitarianism.)
d. the act is still wrong even if we accept utilitarianism because we really ought to maximize preferences, including those of the parent, child, etc, not pleasure.

I would go with c, although I’m not sure what the best definition is. Maybe something along the lines of “preferable by an isolated and perfectly rational actor”, although this begs the question of how something can have a rational justification for its preferably.

Also, if you’re interested in talking directly to a group of utilitarians, there’s a forum I recently discovered called Felicifia.
Blogging Account

Why must utilitarianism take happiness and misery to be different amounts of the same thing? Why can’t they be separate variables?
Stephen Juan

The utility monster examples are straw men. Mill defines “happiness” as living as close as one can to a virtuous life. Pedophile acts conflict with virtuous acts no matter how “virtuous” is defined or by whom.
Nadav Kravitz

Utilitarianism becomes tricky in trade-off situations, in which one has to quantify the gains and losses in order to calculate properly how to achieve net gain. To account for moral monsters, we could possibly make utility a function of quality of life Q, in which 0 is death, 1 is minimally acceptable etc. Then we set utility U equal to (for instance: this is more or less arbitrary) U=A*log(Q) where A is a parameter taking into account various things. A utility monster would have a relatively very large A, while regular people would have a relatively small A.

However, note that we are never forced to kill anyone for the monster, as that would result in infinitely negative loss, at only finite benefit to the monster. One could imagine a being that gets infinite gain from the smallest increase in quality of life, but then we run into the trouble of dealing with infinities (If the monster gets infinite gain from eating one person, what is the net utility after he has eaten one person? infinite gain minus infinite loss equals…?). The utility monster only emphasizes the shortcomings of the theory in practical situations where cost-benefit analysis is difficult, but it does not invalidate the theory.

Previous post: Podcast 50: Mike Jones

Next post: A Dialog between Nick Bohl and Theo Warner

Assault on Utilitarianism: Pedophiles and Utility Monsters

Pages

Random Quote

Newsletter

What is Tuesday Afternoon?