Friday, May 01, 2009

Giving People (Moral) Reasons

I have received a comment from a member of the studio audience that gives me an opportunity to review some of the elements of desire utilitarianism.

Where I wrote that a creature that only has desires that are fulfilled by having slaves and has no desire that would (could) be thwarted by such a desire was an entirely imaginary creature, Kevin wrote:

[T]his should not be difficult to imagine. Prior to the Civil War, such a view was quite commonplace.

Saying that a view is commonplace is not the same as saying that a view is true. It may have well been the case that southern slave owners before the civil war believed that slavery only fulfilled desires and thwarted none. However, this would not make it true. (Nor, by the way, would I even hold that they believed such a thing.)

DU seems a theory that is compatible with just about any moral vision one could have. If my desire is fulfilled by owning slaves, then DU can be used to justify that (even if in a utilitarian fashion which says that more desires are fulfilled by ownership of slaves than not).

DU does not say, "X is (morally) justified if and only if X fulfills the most and strongest of my desires." That would be a theory of practical 'ought', not a theory of moral 'ought'. DU says, "X is (morally) justified if a person with good desires – a person with those desires that people generally have reason to promote – would do X (in the case of moral obligation) or could do X (in the case of moral permission)."

Since people have many and strong reason to cause in their neighbors an aversion to depriving others of their liberty, it follows that a person with good desires would be averse to depriving others of liberty, which means a person with good desires would have an aversion to holding slaves or supporting an institution of slavery, which means that slavery is immoral.

You don't offer any persuasive reason for DU to take an anti-slavery stance except for a very strange assumption that you know the slaveowner's desires better than he.

First, the only type of 'persuasive reason' I am interested in here is whether or not a proposition is true. DU does not base moral claims on the desires of the slaveowner. It may well be the case that (a) slavery is wrong (people generally have many and strong reasons to promote an overall slavery, and (b) a particular slaveowner has no reason to care about this. Such a slave owner will go ahead and do that which is wrong. We cannot 'persuade' him not to do so. However, our inability to persuade him is no proof that slavery is, in fact, not wrong.

One of the biggest problems with moral theory is the realization that giving a person reason to do x means finding a reason that would be persuasive to that particular person. I am not sure any of the reasons you are giving would be persuasive to the slave-owner (especially if he were living in a country that was slave-owning).

There is an equivocation here built into the phrase "giving a person a reason to do X". The equivocation is between correcting a false belief that the agent might have whereby doing X actually fulfills the most and strongest of the agent’s desires, but he does not know it.

DU holds that there is more than one way to "give a person a reason to do X". One is rational persuasion that certain propositions are true. Morality does not use this method.

Namely, let's say that you want to persuade me to do X.

Let us also assume that I will only do that act that will fulfill the most and strongest of my desires, given my beliefs – including altruistic desires. This is not egoism. This is consistent with the claim that a person's desires can be desires for the well-being of other people.

Primarily, there are three types of persuasion that concern us here.

Persuasion Type 1: Practical 'ought'. You can 'give me a reason' by pointing out that a belief of mine is false. I do not want to do X, but this is because I do not realize that X will result in Y, and Y will, in fact, fulfill the most and strongest of my desires. By correcting my belief you give me a reason to do X.

Unfortunately, it might be the case that what fulfills the most and strongest of my desires is something seriously harmful to others. I simply might enjoy giving people pain, in which case "giving me a reason to do X" would mean "helping me to realize that I can inflict more and greater pain by doing X." You have given me a reason, but it is not the type of reason typically associated with morality.

Persuasion Type 2: Threats Another way you can give me a reason to do X is by threatening me. "If you do not do X, then I will make you suffer." You have now given me a reason to do X. It is easier here to look at cases in which you might want to give me reason NOT to do something. For example, you wanted to give me a reason NOT to rape your child. You cannot persuade me that I have no desire to rape your child, but you can still threaten to thwart as many of my desires as you can if you discover that I have raped your child.

Even in this case, your threat is only good insofar as (1) you catch me, and (2) you have the power to carry through on your threat. In addition, these types of threats are no different than threats that you might face from people who might want you to turn over your wallet, help in the roundup and extermination of Jews, or work in the cotton fields without pay.

Persuasion Type 2': Religion. Religion answers two of the three problems associated with threats – the problem of catching the perpetrator and of having the power to punish the perpetrator.

There is a significant difference between saying, "If I discover that you raped my daughter I will seek to thwart as many of your strongest desires as I can," and "If my invisible friend who knows everything and can condemn you to an eternity of flaming torture with a thought catches you raping my daughter, He will make you suffer."

The problem with this method is that these omniscient and omnipotent friends do not exist. The commandments that these entities hand down are not the declarations of a perfectly just being. They are the inventions of the people who invented the God to begin with. The inventors of Gods are neither perfectly wise nor perfectly good. They will give these Gods moral views that are mistaken in some cases. In other cases, they will use God himself as a threat – "Serve God or suffer for eternity" is simply another way of saying, "Serve me or suffer for eternity."

Persuasion Type 3: Altering desires There is still another way in which you can give me a reason to do X. This does not involve giving me a true belief, or threatening me. Instead, it involves altering my desires.

If you were to give me an aversion to rape itself, then you have given me a reason not to commit rape. If you give me an aversion to stealing then you give me a reason not to steal, and if you give me a love of freedom then you give me a reason not to take the freedom of others. If you give me an aversion to lying then I have a reason to tell the truth, and if you give me a desire to help others then you have given me a reason to be charitable.

However, you cannot change a person's desires by reasoning with him. Persuading a person to adopt an aversion to lying or a love of freedom by reasoning with him is no different than persuading a person to love the taste of broccoli or to hate chocolate cake by reasoning with them. You can persuade them that broccoli is good for them and chocolate cake is not, but this is not the same as persuading them to like broccoli and to hate chocolate cake.

The tools for "giving people reasons" in this sense are praise and condemnation. They also include reward and punishment in the behavior modification sense – that which comes to be associated with something that is desired becomes, over time, to be desired for its own sake.

One of the drawbacks of this method is that it has no immediate effect. If somebody is intent on doing something evil (e.g., killing a hostage), then only the first two forms of persuasion are applicable in the short run. If none of them are useful, then evil will be done. Yet, it is still the case that what the agent does is evil if it is still the case that people generally have many and strong reasons to use type 3 reasoning to prevent people from killing hostages (or from taking hostages to begin with).

Even though moral persuasion does not involve reason, we can still use reason to decide the value of using these tools. The analogy that I have used in the past is that one cannot use reason to change a flat tire. Yet, there are objectively true and false claims that one can make about how to use the tools available to change a flat tire.

I have already gone overly long with this post so I will make one more quick statement and save the rest for later.

Moral statements are a combination of Level 1 persuasion and Level 3 persuasion. They are a type of Level 1 persuasion in that they say, "People generally have many and strong reasons to apply Level 3 persuasion to prevent people from performing that type of action." They are, themselves, a type of Level 3 persuasion because they are, in themselves, acts of praise and condemnation. Thus, they are (or can be) objectively true or false. At the same time, they contain an emotive component that is a part of their meaning that is functional (it aims to mold the desires of others through praise and condemnation) without being literally true or false.


Kevin Currie-Knight said...


The biggest reason I find DU infuriating to follow is because it sets up quite a tautology - and a tautology that effectively insulates you from any contrary argument.

The tautology you set up is that "one should act in a way accordant with, and promoting of, good desires." This statement is not only quite vacuous ("Good acts are those that promote good desires,") but also very, very question begging: if the criteria for a good act is that which furthers good desires, then what is the criteria for what consitutes good desires? If you can't answer this, the theory is effectively empty.

So, you write that you are not suggesting that DU tells us to act in accord with our strongest or the societally most common desires, but only the desires that are good. You further write that it is quite possible for someone to think that a desire of theirs is good but be wrong about this.

When I add these up, I see that you have insulated yourself from any possible criticism of your theory. If I were to say, for instance, that DU, followed consistently, leads to (or can lead to) an act that is patently immoral, than you can quickly say that the desire in question only APPEARS good but is really bad. If I say that the desirer feels it to be good (or the society at large feels it to be good) you can quickly say, "but they are wrong. I know what a good desire is, but they clearly do not."

This is a bit like the "no true Scotsman" fallacy, where it is said that real Scots don't x, and if anyone points out a Scot that does x, the retort is that they are not a REAL Scot. But in this case, "real Scot" is actually "good desire." As long as you are the one who gets to decide what is and is not a good desire (I have yet to hear you tell how else this can be deciphered), your theory is effectively unfalsifiable (which is quite a bad thing).

Thus, if all DU has to offer is to say, "Good acts are those which further good desires," then you really don't have much. It is a tautologous idea that is quite question-begging, and utterly unfalsifiable in the sense of the "no true Scotsman" fallacy.

Luke said...

Suggested topic: Desire utilitarianism and moral luck.

Eneasz said...

Hi Kevin. No one gets to "decide" what is a good desire and what is a bad desire. The point is that we must discover what is good and what is bad via experimentation, the same way we discover everything else about the natural world. No one decided what the speed of light is, it had to be discovered. Likewise we (humans) are constantly setting up new experiments (societies) to try different rules and observe what their effects will be. Admitedly it's not phrased like this, someone says "This is right, and will make us all better off!" and then when implemented the world sees if he was right or not. But that is how progress is made. Before the collapse of the Soviet Union it wasn't established that communism tends to thwart desires and make people worse off. Now we know.

See also Who gets to decide

Kevin Currie-Knight said...


How do you propose that we test for what is good or bad? It seems to me like trying to test for what is beautiful - it becomes a matter of opinion. How would we recognize what is good (good is not an empirical quality)?

Andy said...


There is no tautology. In other posts, he defines good as 'that which is able to fulfill the relevant desires.' For moral good, all desires are relevant and morality evaluates desires, not actions.

Therefore, to say an act is morally good is to say a person with good desires would have performed it. A desire is good means that if everyone were to have this desire, it would cause more fulfillment of other desires than if everyone didn't have this desire.

We can think of the desire to rape. Surely, the desire to rape thwarts the desires of everyone else because they are in danger of being raped. The more rapists there are, the more dangerous of a situation everyone else is in. If everyone instead had an aversion to rape, no one is in danger and everyone is much better off. (By much better off, I mean more desires are fulfilled)

So, there is no tautology.

After reading up on moral luck, I too second Luke's request for a post on DU's relationship to moral luck.

Kevin Currie-Knight said...


You haven't added anything to the equation in your post. IT is still just as circular. We are still saying that a good act is that which fulfills good desires. You also say that a "an act is morally good is to say a person with good desires would have performed it." A "desire is good means that if everyone were to have this desire, it would cause more fulfillment of other desires than if everyone didn't have this desire."

So, to sum up: An act is good if it fulfills the type of desires we call good because if acted on, they would fulfill other desires."

Then you write of rape as an obvious case. Rape thwarts other desires. Therefore it is a bad act.

But what if there were more rapists than rapees in the world? Wouldn't that, in a utilitarian sense, mean that tearing down of rape laws would fulfill more desirs than it would thwart?

Obviously, I am not saying that rape is morally just, but am saying that Fyfe's theory, vacuous as it is, lacks the ability to condem it without some good rhetoric. A case can be made, under his theory, that rape is a bad act, or that rape could be a good act. A theory that malleable is suspect.

In order for Fyfe's DU to convince me, he'd have show that "the ability to fulfill other desires" is the exact same thing as "good." As the case of rape shows, I am not sure that calculation is right (as rape would still seem to be bad even if the desires of rapists were far more satisfied than the desires of rapees.) There seems more to it than that.

Andy said...


It's not that rape is bad because it thwarts other desires. It's that the desire to rape is bad because it thwarts other desires. As I said before, we are a lot better off if there are no rapists, therefore, the desire to rape is a bad desire.

This has nothing to do with how much the rapist enjoys raping his victim. The point is, the act of rape will hurt the other person and cause others to be in danger. I am not evaluating the act of rape though but the desire to rape. This is not an act-based calculus but a desire-based one.

So, if no one had the desire to rape, we would all be better off. Our other desires would be fulfilled. It doesn't make any sense to object, "Well what if there were a lot of people with the desire to rape?" or "What if the rapist enjoys rape more than the victim hates it?" I've already stated that if no one likes rape, we would be better off. I'm not sure what your objections have to do with challenging this statement.

It's not that the aversion to rape fulfills good desires. Just that it tends to fulfill all other desires. Our desire for safety, our desire to not be raped, our desire that our friends and family are safe, etc. I am not saying these desires listed are good or bad, they can be amoral desires. The objects of evaluation still have value and hence need to be included when considering the desires fulfilled and thwarted by the existence of another desire.

Eneasz said...

Kevin - Andy hit it on the head. See also Alonzo's post 1000 SadistsHow do you propose that we test for what is good or bad? It seems to me like trying to test for what is beautiful - it becomes a matter of opinion. How would we recognize what is good (good is not an empirical quality)?

Actually I would argue that good IS an empirical quality: that which tends to fulfill other desires. Sometimes it's relatively easy to test - it was discovered that killing, stealing, raping, assault, and a host of other things are not good millenia ago. Sometimes it's difficult and results in unneccessary misery and/or death. It took the collapse of a major country and the misery of millions to firmly establish that communism is worse than capitalism in tending to fulfill desires.

But it's certainly not impossible.