Monday, January 28, 2008

The Sum of All Reasons

In a comment to a different post, I received a different question about desire utilitarianism. I get this question in a number of different forms, and would like to take this opportunity to address one of those forms.

…my question is: when you say, "Furthermore, it argues against individualist subjectivism that it is all desires, not just those of the agent, that must be taken into consideration," Why MUST all desires be taken into account? What is the justification for that?

Okay, let us assume that I have given you a long column of numbers. I ask you for the sum of all of those numbers. Let us say that you decide only to add up those numbers that ended in a ‘7’, or only the first 15 numbers. The answer you give is not the right answer. It is a simple matter of fact that the sum of a column of numbers considers every number in the column, not a subset of those numbers.

Similarly, if I were to ask you for the center of population of the United States, you cannot answer that question by looking only at the people who live in New York State. You can, perhaps, give a reasonable approximation if you considered only the female population of the United States, but that is still only an approximation. It is still the case that the right answer is the answer that you would get by including the relative location of every person.

And, finally, if I were to ask you to determine the actual acceleration of a body in space and time, this is equal to the vector sum of all of the forces acting on it. If you were to arbitrarily select some subset of those forces, you will not get the right answer, unless (by chance) the forces you exclude exactly cancel each other out.

Various sets of relationships between states of affairs and desires exist. There are, for example, relationships between objects of evaluation and my desires alone. When we speak of these relationships, we speak in terms of personal preferences. For example, having sex with Jenny might be the act that fulfills the more and stronger of my own desires alone.

However, much of what we understand as morality simply cannot be reduced to statements about relationships between objects of evaluation and the desires of the agent. For example, there is no valid inference from, “I want to have sex with Jenny,” or even from, “Having sex with Jenny would fulfill the more and stronger of my desires,” to “Jenny has an obligation to have sex with me.” The concept of an ‘obligation’ has a number of elements that cannot be derived from personal preferences.

There are also relationships between objects of evaluation and some subset of desires. For example, it may well have been the case that the institution of slavery before the civil war fulfilled the more and the stronger of the desires of the slave owners. However, even here it is impossible to explain how the fact that an institution fulfills the desires of the slave owners that the slaves therefore had an obligation or a duty to serve as slaves. If the inference were valid, then how does it work?

In desire utilitarianism, value statements describe relationships between states of affairs and desires. They are descriptive in that, for true value statements, those relationships truly exist. They are prescriptive in that desires are reasons for action, so a description of the relationship between an object of evaluation and certain desires is also a prescription for those people who have the desires in question. If A is such as to fulfill desire D1, then it follows axiomatically that people with desire D1 have a ‘reason for action’ to bring about A.

From this it follows that the only way to prescribe universally – the only type of universal prescription that exists – is one that considers all desires. Excluding any desires means that one prescriptions do not apply to those people whose desires are not included – it means that one’s prescriptions are not universal.

Now, I do argue that a right action is not the action that fulfills the more and stronger desires that exist. A right action is the action that a person with good desires would perform. Of course, the concept of a good desire is a concept that excludes bad desires. Does it then follow that moral prescriptions do not apply to people with bad desires?

No, it does not. A “good desire” is a desire that tends to fulfill other desires. “Good desires” are desires that people generally, as a matter of fact, have the most and strongest reasons to encourage in others, and that others have reason to encourage in the agent. It does not allow any desire to be included or excluded on the basis of intrinsic merit, because there is no such thing as intrinsic merit. It only allows desires to be evaluated in the only way that evaluations are possible – in terms of relationships between the object of evaluation (in this case, desires) and all other desires.

So, moral claims prescribe desires for everybody. They are those desires that it makes sense for people to encourage everybody to have, or desires that it makes sense for people generally to encourage nobody to have.

Now, we have these ‘desires that it makes sense for everybody to have’. They exist, and they are real.

I do not make any claims about these desires other than that which can be objectively proved. I am not saying, ‘These desires exist, plus they have some sort of intrinsic merit that makes them worthy of being pursued or promoted.” I am only saying, “These desires exist, plus, to the degree that they tend to fulfill or thwart other desires, then to that degree those with the desires being fulfilled/thwarted have reason to encourage or discourage the adoption of these desires.”

There is a tendency to assert that I need to add something more to this – a tendency to ask, “Why should I consider all other desires?” However, “should” questions in the form, “Why should I do X?” are questions that ask, “What reasons for action exist for doing X?” A person answers a “should” question by identifying the reasons for action that exist.

However, I do not see any need to add something more to this. If a desire exists that tends to fulfill the most and the strongest of desires, and desires are the only reasons that exist, then what else do I need to say to suggest that people have reason to promote this desire? If the most and the strongest desires that exist are fulfilled by this desire, then the most and the strongest reasons for action that exist prescribe promoting this desire.

“Should” questions are questions that ask for reasons for action. The question, “Why should I do X?” simply asks, “What reasons for action exist for doing X, for not doing X, and on which side of the scale do the more and the stronger reasons exist?”

If you answer a “should” question with a reason for action that does not exist, then the answer is false. For example, somebody might believe that there is intrinsic value in doing that which is natural, or that which pleases God. They may think, for example, that there is a ‘reason against action’ to engage in homosexual acts in that the unnaturalness of the act itself generates a reason not to perform it, or that God’s displeasure generates such a reason. As a result, one might be tempted to conclude that one should not engage in homosexual acts.

However, all claims that ‘reasons for action’ can be found in intrinsic merit or in what pleases or displeases God are false. When people answer a “should” statement with a statement that refers to these types of entities – or any entities other than desires – their claims are false. It is possible that reasons for action that do exist – real reasons for action – might still recommend the same act, but certainly not for the same reasons.

Since “should” questions are questions that ask us to provide reasons for action, and desires are the only reasons for action that exist, “should” questions require that we answer by describing the object of evaluation to desires. These claims are capable of being true.

It is as much of a mistake to deny a reason for action that exists, as it is to refer to a reason for action that does not exist. If an agent asks, “Should I do X?”, and there is a reason for action for doing X, but the person who answers the question ignores this reason (and thus concludes that the agent should not do X), then that person has given a wrong answer. Referring to reasons for action that do not exist, and ignoring reasons for action that do exist, both lead to wrong answers in answering “should” questions.

When people make reference to reasons for action that do not exist, they often cause people to do that which they should not do, or convince people not to do that which they should do. False beliefs about the reasons for action that exist lead to all sorts of mistakes. The greater the mistake – the more weight one gives to reasons for action that do not exist, the more we have people who are not doing what they should be doing, or doing what they should not be doing.

There simply are facts about relationships between desires, such that some desires tend to fulfill or thwart other desires. We can pay attention to those facts or not. But, do we have reason to pay attention to those facts? We have the best possible reasons – the only reasons for action that are real, and the more and stronger reasons to pay attention to the best and worst desires.

8 comments:

Anonymous said...

Question: Consider two agents, A and B. Each has 10 equal desires. Each currently has 5 fully satisfied desires, and 5 unsatisfied desires. You as agent C, and desire utilitarian, have the ability to fulfill 2 of agent A's desires, at a cost of 1 of agent B's desires. You may continue to trade at a two for one ratio as much as you would like. What if any is the desire utilitarian argument against minimizing agent B's expensive desires in favor of agent A's cheap ones?

Anonymous said...

Alonzo,

What am I missing here?
A right action is NOT the action that fulfills the more and stronger desires that exist.
A right action IS the action a person with good desires would perform.
Good desires are desires tend to (that lead to actions creating states of affairs) that fulfill other desires.
So if I have good desires why wouldn’t I desire to perform the action that would fulfill the more and stronger desires that exist?
And wouldn’t that then be the right action?.

Alonzo Fyfe said...

Atheist Observer

Desires are persistent entities - they continue to exist through a range of choices.

Assume the following set of choices with the following options:

Choice 1a vs 1b
Choice 2a vs 2b
Choice 3a vs 3b
Choice 4a vs 4b
Choice 5a vs 5b

In each case, the A option is the option that will fulfill the more and stronger desires.

There are two possible desires.

D1 will motivate an agent to do 1a, 2a, 3a 4b, and 5a.
D2 will motivate an agent to do 1b, 2b, 3b, 4a, and 5b.

Desire D1 is a desire that tends to fulfill the more and stronger desires. However, Choice 4, the agent with desire D1 will not do the act that fulfills the more and stronger desires. For choice 4, 4b is the right action, even if it does not produce the best consequences.

This handles the large set of counter-examples to act utilitarianism. These counter examples occur highly unlikely (impossible?) situations where it just so happens that the act that produces the most utility seems wrong. The case of a doctor who can cut up a healthy patient and distribute his organs among five sick patients, or executing an innocent man in order to appease a mob. These are all "choice 4" situations. Notice that they involve highly rare and extremely contrived situations.

The only type of person who will always choose the act-utilitarian best act is a person that has only one desire - a desire to fulfill other desires. He cannot be averse to pain, prefer the taste of chocolate over vanilla, like action-adventure movies. If he has any desire other than the desire that other desires be fulfilled, then he will face "choice 4" situations where the act-utilitarian best act is not the act that fulfills the most and strongest of his desires.

Martin Freedman said...

kmeson

I am not sure if Alonzo answered you when he answered Atheist Observer so here is my response to your question.

You as agent C, and desire utilitarian, have the ability to fulfill 2 of agent A's desires, at a cost of 1 of agent B's desires.
First do not confuse fulfilling of desires with evaluation of these desires, everyone is attempting to do the former and DU is a solution about the latter. Now a good desire is a desire that fulfills or tends to fulfill other desires. This is the means to evaluate any desire, whatever the desire actually is. DU is about what people should do (desire) rather than what they do do.

In your example the question is should Agent C have the desire to fulfill two of Agents A's desires at the cost reduce agent B's desire fulfillment by on desire. The answer to the should - does C have a reason to act - depends on whether this is a good desire or not. Now since C's desire would thwart B desires it is a bad desire and not one to be commended let alone carried out by C. Therefore C should not do this.

Anonymous said...

martino

Doesn't C's failure to act thwart Agent A's desires more than acting thwarts Agent B's?

My background is in math and physics not philosophy, so this may be a naive question, but I'm concerned about the non-zero sum game. The example I've given is the two player version, but there are multi-player versions as well. The reductio ad absurdum that nags me is the state where a well meaning Agent C in the example above in a bid to maximize the sum total of all desires chooses to maximally thwart one person in favor of another.

Martin Freedman said...

@kmeson:Doesn't C's failure to act thwart Agent A's desires more than acting thwarts Agent B's?
OK, you are looking at acts of omission as well as acts of commission. If C has no desire and does not act, then the issue is to do with the desires of A and B. You asserted they were all equal which means they do not care which specific ones are realized but they both want more than less?

If there is any thwarting going on it is that the remaining unfulfilled desires of A thwarts's B's desires - that is those desires of A's are bad desires and C might or should instead condemn A for having such desires.

Martin Freedman said...

A question for Alonzo

Carrying on with kmeson's theme but simplifying it. Two agents A and B each have one desire. A is the states of affairs that P obtains, B is that state of affairs that Q obtains, where Q=~P.

According to DU these are both desire thwarting desires and therefore both bad and both to be condemned.

But what if the state of affairs P is that the car is sold to A, whereas with Q it is sold to B, there is only one specific car for sale and no other candidate purchasers?

Martin Freedman said...

I think I can answer my own question here since whilst I pointed out the difference between DU and DFAU I still got confused here! :-(

The question to ask is what is the right act that a person with good desires would perform here? Clearly my example does not apply to answering such a question. They could both be "good people" just bidding over car. One succeeds one does not and that is life. (I am excluding cheating and bribery etc. here to which DU would apply but that was not in my template).

On the other hand are there scenarios like my template here that do not involve something like trade?