Desire Utilitarianism
I. Introduction.
I have always been an atheist. I have never had the thought that some immortal soul will carry my consciousness into the indefinite future, granting me immortality. So, if anything of me was to survive my death it would not be me, but it would be some influence that I have had on the world. I wanted to leave the world better than it would have otherwise been.
But, what is 'better'? I heard a lot of suggestions, but people kept fighting about this. Indeed, more conflict and misery has been generated in the name of the combatant's definition of 'better' than for any other reason I can imagine. So, I wanted to know what 'better' was.
This started a journey that lead through 12 years of college devoted to value theory. By the time it was over, and I had to go into the real world, I had formed an idea that I thought made a lot of sense. That theory is Desire Utilitarianism.
I would like to present the basic structure of that theory here.
II. What Is A Desire?
The best place to start is with the question, "What is a desire?"
A desire is a propositional attitude.
Fine. What is a 'propositional attitude'?
A 'propositional attitude' is an attitude towards a proposition.
That's not very helpful.
Let's try this.
A proposition is the meaning component of a sentence. "Snow is white", or "There are nine planets in the solar system" are examples of propositions. Propositions are capable of being true and false or, if you like fuzzy logic, sometimes somewhere in between. For the purpose of this exposition, I want to ignore all of this fuzzy stuff and assume that propositions are either true or false. Fuzzy logic does not change the basic theory. It just makes it a little... well... fuzzy.
This assumption, by the way, is the same type of assumption that physicists use when they talk about massless strings and frictionless surfaces. These other influences have an effect. However, given what we want to focus on, they add an unnecessary complexity.
Let's take a proposition P = "The sun will come up tomorrow."
Okay, now, there are two families of basic propositional attitudes: beliefs and desires.
A "belief that P" is a mental attitude that the proposition P is true. That is to say, a person who believes that the sun will come up tomorrow, has a mental attitude that the proposition "the sun will come up tomorrow" is true. You will see him acting as if the proposition is true.
A "desire that P" is a mental attitude that the proposition P is to be made or kept true. That is to say, a person who desires that the sun will come up tomorrow has a mental attitude that the proposition "the sun will come up tomorrow" is to be made or kept true. If faced with a choice between an option that will lead to a case where the sun will come up tomorrow, and one in which it will not, the person with the desire will choose the option that causes the sun will come up tomorrow. Or, at least, the option that he believes will lead to that result. Unless some other desires get in the way.
There are no more propositional attitudes at this level. All other mental terms can be reduced to statements about beliefs and desires. The mental terms "know", "understand", "think", "suspects", "doubts" all refer to states that fall within the general category of "belief". Whereas the mental terms "wish", "prefer", "want", "hope", "fear", "concern", and the like all fall within the general category of "desire".
These propositional attitudes are the causes of intentional human action.
(Beliefs + Desires) -> Intentions -> Intentional action.
So, you desire to go see a movie. You have a stronger desire for an action movie than for any other type. You believe that Indiana Jones IV is playing at the local theater. You also believe that the movie starts at 5:00pm. You desire to go with your significant other. You believe that this person has a cell phone, and you believe that the number is 555-555-5555. So, you call up your significant other, invite him or her to go to the movie with you, make plans to pick him or her up at 4:15, and drive to the theater.
All intentional action is the product of some set of beliefs and desires. Your beliefs are mental attitudes that certain propositions (e.g., "the movie starts at 5:00" and "Indiana Jones IV is playing at the local theater") are true (though some of them may be false -- maybe the movie starts at 4:30). Your desires are mental attitudes that certain propositions (such as "I am seeing an action movie" and "I am with my significant other") are to be made or kept true. The interaction of these brain states produces intentional action, and the next thing you know you're sitting in the theater watching an action movie with your significant other.
So, this points out a significant difference between beliefs and desires.
Beliefs can be correct or incorrect. If a person believes that P (e.g., "that the movie starts at 5:00"), but P is false (e.g., the movie starts at 4:30), then the belief that P is incorrect. A person acts as if all of their beliefs are correct. If a belief turns out to be incorrect, this could fail to fulfill his desires (arriving at the theater too late to see the movie).
Desires are not correct or incorrect. If a person desires that P (e.g., "that the movie starts at 5:00"), but P is false (e.g., the movie starts at 4:30), there is nothing 'incorrect' about the desire. The desire has been thwarted, but the desire is not 'wrong' in the sense that the belief is 'wrong'.
I am going to use a couple of common terms here in slightly uncommon ways.
I am going to say that if a person desires that P, and P is false, then the desire is thwarted. If a person desires that he is watching an action movie with a significant other, but the significant other is busy and cannot make the movie, that desire is thwarted.
I am also going to say that if a person desires that P, and P is true, then the desire is fulfilled. If a person desires that he is watching an action movie with a significant other, and he is in the theater watching the movie with his significant other, that desire is fulfilled.
A desire, recall, is a disposition to make or keep a proposition true. So, a desire that one is watching an action movie with one's significant other is a disposition to make it the case that the proposition, "I am seeing an action movie with my significant other" true.
III. Selfishness and the Fulfillment of Desire
With any type of utilitarianism, there is a question about what it is that we are going to maximize. Jeremy Bentham in the early 1800s argued that it was pleasure over pain. John Stuart Mill during the middle of that century argued for happiness over unhappiness. In the argument that follows, I will argue for desire fulfillment. In doing so, I am going to provide an argument against any notion that all humans are selfish because we seek only the fulfillment of our own desires.
Okay, it's true that we only act to fulfill our own desires. This follows from the fact that desires are brain states, and there is only one brain attached to my arms and legs in such a way that they can directly cause the movement of those arms and legs. Of course it is my desires that I am acting to fulfill. How can I possibly act to fulfill desires that are locked away as a part of somebody else's brain?
However, this does not make us selfish. Our desires can be (and many of them are) other-regarding desires. A person desires that their child be healthy and happy. Such a person is disposed to make it the case that the proposition "my child is healthy and happy" is true. If one calls "selfish" a person who only desires what makes others healthy and happy, and devotes their life to making the proposition "others are healthy and happy" true, then that individual is using a very bizarre definition of the word "selfish".
Here is a little thought experiment that should completely refute any residual notion that people are basically selfish, while it illustrates the case that desire fulfillment is what human action aims towards.
You, and somebody you care a great deal about (e.g., your child) have been captured by an evil extra-terrestrial mad scientist who is interested in conducting all sorts of experiments on humans. It offers you the following two options:
Option 1: "I will take this other person to another ship and perform all sorts of medical experiments on him. We have become well versed in the art of vivisection, I assure you, and the process will be painful and unending. However, you will be made to believe that your child has been set free and allowed to live a safe and happy life."
Option 2: "I will let your child free with enough gold to live a healthy and happy life. However, you will be made to believe that I have taken your child to another ship and that I am performing all sorts of medical experiments on him. You will be made to believe that we have become well versed in the art of vivisection, and that the process will be painful and unending."
Of course, I will also cause you to forget about this choice.
Which option do you choose?
The view that says that everyone is only after their own happiness would have to argue that everybody would select option 1. This is, after all, the option that provides the agent with the most happiness. However, this is contrary to fact. Most people go with option 2 -- they sacrifice their own happiness for the sake of the child.
We have no actual experiments to point to that involve locking people in a cell and asking them to make this choice. So, maybe, they would all choose Option 1. But, most people at least report that they would not, and no reason can be provided to doubt them. The selfishness theory needs to at least explain why so few people think (incorrectly) that they would choose Option 2.
The widespread choice of Option 2 is easily explained if we hold that desires are dispositions to make or keep the proposition that is the object of the desire true. The parent with a desire that their child is healthy and happy is disposed to make or keep the proposition "my child is healthy and happy" true. Option 2 is the option in this case that makes or keeps the proposition true. It is desire fulfillment that we are after in life, not pleasure, nor happiness.
IV. What Is Value?
Here, I am going to take what I said about desires above, and turn them into a theory of value. I will argue that value terms, and all true value claims, relate states of affairs to desires.
We have a universe in which the proposition, "Mike desires that the movie start at 5:00" is true. If this is the case, we also have a universe in which the proposition, "The movie starting at 5:00 is such as to fulfill Mike's desire" is true. These are, after all, two ways of saying the same thing -- in the same way that, "Mike is taller than Sally" says the same thing as, "Sally is shorter than Mike."
Here is my proposal:
"Good" = "Is such as to fulfill the desires in question."
"Bad" = "Is such as to thwart the desires in question."
This handles every instance of good and bad in the real world.
Now, for a word of warning. This is talking about generic 'good'. I am not talking about morality or any more specific type of good -- at least not yet. Certain specific value terms -- such as moral terms -- place limits on the objects of evaluation, or the relevant desires, or the types of relationships that exist. So, something can be good in the generic sense without being morally good. This 'good in the generic sense' is the sense of good in which it still makes sense for a psychopath to cut up an innocent victim and sigh, "Man, that was good!".
Or, "Man, that was such as to fulfill the desires in question, where the desires in question are my own, and only my own."
We still have to get from here to the more specific meanings of terms.
This is easy to do, because this account identifies four different criteria for any value term.
(1) What are the objects of evaluation that the term applies to?
(2) What are the desires that are relevant for that particular evaluation?
(3) Does the term state that the objects of evaluation fulfill or thwart those desires?
(4) Do the objects of evaluation fulfill or thwart those desires directly, indirectly, or both?
Every value-laden term, insofar as it can play a role in a true proposition, has these four elements. Here are some examples:
(A) Illness and Injury
Let us take, for example, 'illness' (and, by association, 'injury'). 'Illness' (and 'injury') are value-laden terms. That is to say, nothing can count as an illness (or an injury) unless it is something bad. This is no accidental relationship between being an illness (or an injury) and being bad. This is true in the same way that all bachelors are unmarried. It is part of the meaning of the word.
Now, to express 'illness' (and 'injury') according to the general criteria for value listed above:
(1) Illness and injury are both used to evaluate changes in mental or physical functioning (depending on whether we are talking about mental or physical illness/injury).
(2) The desires that are relevant in determining whether someone is sick (injured) are those of the agent. I could, perhaps, be the happiest person in the world that my brother has cancer. But my desires are not the desires that are used to determine if my brother's cancer is an illness. That depends on the relationship between the change in physical functioning and my brother's desires.
(3) Illnesses always thwart desires. An illness is a change in physical or mental functioning that is such as to thwart the desires of the person to whom an illness is being attributed.
(4) Illnesses can thwart desires directly (cause pain and discomfort), or indirectly (cause the loss of mobility or memory which, itself, thwarts desires that one can fulfill best by moving or remembering).
The difference between 'illness' and 'injury' can be handled by making a further refinement to the first criterion. An illness has a micro-cause (not easily seen by the naked eye), while an injury has a macro cause (something that we can see).
Cancer is a disease. A pre-technology culture can see the effects, but they have to guess at the cause. To the primitive eye, diseases (illness) just happen.
Injuries, on the other hand, have a macro cause that a pre-technology culture can easily see. A person who gets trampled by a horse, or who falls off of a building, or gets an arrow in his eye at the Battle of Hastings, has been injured.
(B) Useful
'Useful' is a value-laden term. No person can make a comment that something is useful without saying something about its value, in a sense. The criteria for something to be 'useful' are as follows:
(1) The object of evaluation when the term 'useful' is used could be just about anything. It could be a power tool, a person, a mountaintop, or any random thing picked up off the ground.
(2) The desires that are used in making this evaluation are determined in context. Whenever a person uses the word 'useful', you have to look at what is happening -- at other things he is saying or what he is doing or what you know he wants to do, in order to determine which desires are being referenced.
(3) Useful things are useful in the sense that they fulfill desires. A useful thing can also thwart desires (e.g., salt is useful in torturing somebody), but the 'desires in question' in this case are not those of the torture victim, but those of the person doing the torturing. These are desires that the salt fulfills.
(4) Useful things always fulfill desires indirectly. Useful things can be pleasing to the eye, but it is not in virtue of this that they are useful. They are useful if they can be used to bring about some change, where the change (or some effect of the change) fulfills particular desires. The Mechanic's wrench is not useful because it looks pretty. It is useful because it can be used to make modifications to the engine, to make the engine more powerful, which can then bring about a state of affairs that fulfills desires.
The key element here is that 'usefulness' never evaluates things according to their ability to fulfill desires directly, but only in virtue of its ability to bring about other things that are desired. They only fulfill desires indirectly. Useful things can also fulfill desires directly, but this is not relevant to its usefulness.
(C) Beauty and Ugliness
(1) The term 'beautiful' is applied to things that are seen or heard. They are not applied to things tasted, felt, or smelled. Why? I don't know. I am not the person who invented the English language. There exists, in the universe, relationships between things seen and heard and certain desires (to be discussed below), and we opted to give these relationships the name 'beauty'.
(2) The desires that the object of evaluation are being related to are those of the viewer or listener. "Beauty is in the eye of the beholder."
(3) Beautiful things fulfill desires. Ugly things thwart desires.
(4) Beautiful things fulfill desires directly (and ugly things thwart desires directly). They are immediately perceived as beautiful or as ugly. The ability something has to bring about some future state which fulfills desires makes something more useful, but does not make it more beautiful. Ugly things thwart desires directly.
V. Moral Concepts
Ultimately, what I am most interested in, are moral concepts. So, what do moral concepts evaluate?
(A) What Is The Object Of Evaluation?
The ultimate object of moral evaluations are desires themselves. Desires have the capacity to be 'such as to fulfill other desires' or 'such as to thwart other desires'. Because of this, we have the ability to evaluate desires as 'good' or 'bad'. Examples of good desires would be charity and honesty (a desire to prevent the thwarting of other people's desires, and an aversion to representing to others as 'true' propositions that one sincerely believes are 'false').
Of all of the desires that we may have, moral evaluations focus only on those that are malleable -- that can be changed by environmental influences. Morality is not concerned with desires that are hard wired. Desires that are subject to environmental influences are desires that can be changed, and it is only with respect to desires that can be changed that it makes sense to ask "What desires should we have?". So, for example, our aversion to pain sits as a background condition to moral considerations, like acceleration due to gravity and the chemical composition of tree bark.
Why focus only on desires that can be changed through environmental influence? This is because the question that lies at the root of all moral concepts is, "What types of desires should we have?" Morality is concerned with using environmental factors to control the desires that exist, promoting desire-fulfilling desires and inhibiting desire-thwarting desires.
It is also relevant that, by far, the most powerful tools to use in influencing the desires that exist are reward and punishment. Praise and reward an action, and the desires that motivate that action become stronger. Condemn and punish an action, and the desires that cause that action are repressed. One does not even need to condemn or punish an actual instance, but simply make it known that a particular action is 'of a type that is to be condemned', and an aversion to that act type is nourished. We learn an aversion to rape and to killing innocent people, most of us, without ever having to be actually punished for a rape or a murder.
We do, of course, morally evaluate actions. However, I will argue below that the only sensible way to evaluate an action is to derive its value from the value of desires.
By the way, the value of an action is NOT derived from the desires that cause the specific action. The value of an action is derived from whether or not a person with good desires would have performed that action. An action that a person with good desires would have performed would be an 'obligation' or 'duty'. An action that a person with good desires would not have performed would be 'prohibited'. And an action that a person with good desires may or may not perform is 'permissible' in the sense of 'neither obligatory nor prohibited'.
This, by the way, handles issues of negligence -- the issue that the late 19th Century philosopher James Martineau stumbled on and that his contemporary Henry Sidgwick criticized him for. Martineau argued that the value of an action depends on the value of the motives that caused the action. Sidgwick argued that this makes no sense in cases of negligence, where the motives for the action are common motives (a desire to move a load of lumber from one place to another), but a lack of caution causes dire consequences (the load is not secured well and people get hurt).
These cases can be handled by recognizing that the negligent person lacks a desire that he should have had -- that a good person would have had -- that would have caused him to fasten the load more securely.
A right act is an act that a person with good desires would have performed, and a wrong act is an act that a person with good desires would not have performed. A good desire is a desire that tends to fulfill other desires, and a bad desire is a desire that tends to thwart other desires. The evaluation of actions is derived from an evaluation of desires, and an evaluation of desires stands at the heart of morality.
(B) What Desires Are Relevant?
Moral terms evaluate desires relative to all other desires that exist.
This is where most versions of common subjectivism make their mistake. While it is true that all evaluations relate the object of evaluation to some set of desires, they do not all relate the object of evaluation to the desires of the speaker. 'Beauty' does this -- it relates the object of evaluation to the desires of the person seeing or listening. However, 'injury' relates the object of evaluation to the desires of the person injured, not the desires of the person viewing the injury.
I would consider a person beautiful if their appearance is such as to directly fulfill my desires regarding what I see. But I would consider a person 'injured', not by relating their change in physical functioning to my desires, but by relating that change in physical functioning to his own desires. It might even please me that a person is injured. I might have been the one to injure him. But the concept of 'injury' does not depend on what I like and do not like.
Neither does the concept of 'duty', nor does another person's 'rights' depend on what I like and do not like. If it did, then "I prefer chocolate ice-cream to vanilla" would be a moral statement, no different from "I prefer the execution of murderers to life in prison." If it did, then inferences of the form, "I want Jenny to have sex with me, therefore Jenny has a duty to have sex with me," would be valid. We would see no difference between these types of claims, if moral claims related objects of evaluation solely to the desires of the speaker.
(C) Do the Objects of Evaluation Fulfill or Thwart The Desires in Question?
Obviously, we have a number of moral terms -- some of them used when we make positive evaluations, and others used when we make negative evaluations. Those that we evaluate positively ('duty', 'obligation', 'right') are such as to fulfill the desires in question. Those that we evaluate negatively, ('wrong', 'prohibited', 'contemptible') are such as to thwart the desires in question.
(D) Do the Objects of Evaluation Fulfill/Thwart Desires Directly Or Indirectly?
Conceptually, I do not think it matters. By far the most common way that a desire can fulfill or thwart other desires is indirectly. The person contributes money to charity, which then fulfills the desires (or, at least, prevents the thwarting of desires) of the person to whom charity is given. Honesty spreads true propositions that, as I have demonstrated, are useful in fulfilling desires. An aversion to taking things that belongs to others simply prevents the thwarting of desires that would come from taking what belongs to others (and from living in a society where one has to live in fear that something may be taken).
VI. The Fact/Value or Is/Ought Distinction
Now, I have to deal with two avenues of attack on this line of reasoning. G.E. Moore, if he were around to read this, would charge me with committing a 'naturalistic fallacy'. While David Hume, writing in the 18th Century, would require that I explain how I can possibly derive an 'ought' from an 'is'.
(A) G.E. Moore's 'Naturalistic Fallacy
G.E. Moore argued that "good" is a basic unanalyzable property, and offered as proof of this his "open question argument".
According to G.E. Moore, any proposition of the form "Good is X" can be defeated because the question, "Object is X, but is it good?" remains an open question. If "good" really is "X", then the question, "Object is X, but is it good?" would be a nonsense question. Since I am offering a definition of 'good' that takes the form 'good is X' (specifically, 'Good' = 'is such as to fulfill the desires in question'), I need to get around Moore's objection.
The problem with Moore's Open Question argument is that it is an example of the masked man fallacy.
This masked man fallacy gets its name from its explanation. Imagine that you live in a community where the news recently has been filled with accounts of a masked man who robbing people on the highways outside of town at night. Somebody comes to you and says, "Your brother is the masked man."
You respond, "That's not possible. Last night I was wondering myself who the masked man was, but I was not wondering who my brother was. So, they cannot be the same person."
Obviously, your wonderings have nothing at all to do with whether the masked man in your brother. That you did not know that your brother was the masked man is a property of you, not a property of your brother or the masked man. As such, it does not provide proof that the two are not one and the same person.
Similarly, G.E. Moore's wonderings about "Object is X, but is it good?" has nothing to do with whether goodness is ultimately understandable in terms of X.
To answer whether good is X, we need to perform the same type of investigation that we would perform in determining if your brother is the masked man. We need to look for evidence that the same entity plays both rolls -- that of brother and masked man, or that of 'good' and 'is such as to fulfill the desires of question'.
I believe that the previous section at least provides a start in this direction. I have shown how the concepts of 'illness', 'injury', 'harm', 'hurt', 'beauty', 'ugliness', and moral concepts can all be replaced by different species of the genus 'is such as to fulfill (thwart) the desires in question'.
(B) David Hume's 'Is/Ought Distinction'
In an often-quoted paragraph, Hume argues that he can defeat all 'vulgar' systems of morality by noting that the person begins with all sorts of propositions that describe "is" relationships (e.g., God created man"), and then suddenly and inexplicably shifts to "ought" relationships (e.g., man ought to obey God), without explaining how the latter can be derived from the former. Some sort of explanation is needed, and until one is provided, we can dismiss these 'vulgar' attempts to derive 'ought' from 'is' as illegitimate. They are grounded on faulty logic.
Here is the explanation Hume has asked for:
'Ought' is simply one 'is' relationship among many, and the derivation of 'is' from 'is' is not at all problematic. 'Ought', as a species of the genus 'good', which, in turn, is 'such as to fulfill the desires in question'. This means that 'ought' statements are 'is' statements, and this means that they can be derived from other 'is' statements.
However, insofar as 'ought' (and 'good') relate states of affairs to desires, the derivation of 'ought' from 'is' must take the following form:
(1) 'Is' statements about some set of desires
(2) 'Is' statements about the object of choice
------------------------
(3) Therefore, 'ought' statement that describes the object of choice as the
object that fulfills the more and the stronger desires in question.
So, of Archie wants a flat-screen monitor at a reasonable price, and Dell offers a flat-screen monitor at the lowest price, then he should get the Dell monitor -- meaning only that the object of choice is better able to fulfill the desires in question.
Now, the main objection in deriving 'ought' from 'is' is that no set of facts can entail motivational force. I can tell you all sorts of facts about the house, none of those facts entail that you should buy it.
However, if I include facts about your desires, and can show you that your owning the house 'is such as to fulfill the desires in question', then I can conclude that you should buy the house.
That is, IF it is true that the house 'is such as to fulfill the desires in question', THEN it is true that you have a motivating reason to buy the house. In other words, you 'ought' to buy the house. And if there is any reason why you 'ought not' to buy the house, then the only reasons that make sense are those that take the form 'is such as to thwart the desires in question' for some set of desires. All of this fits quite comfortably in the universe of 'is' propositions.
This admittedly superficial description can be further argued by bringing up a significant problem with the traditional interpretation of Hume's famous paragraph.
With respect to the relationship between fact and value, between 'is' and 'ought', we are faced with three options.
(1) Value reductionism. Values exist somewhere in the world of fact. There are only two options, either 'is' or 'is not'. And values are either part of what 'is', or a part of what 'is not'. If we think we are going to have difficulty explaining and predicting events in the physical world without value terms, then the only option is that a value term refers to something that 'is'.
(2) Value eliminativism. Values are a part of what 'is not'. Values do not exist. There is no 'value' that is a part of any reasonable explanation for my writing this article. There are just a bunch of quarks bouncing off of each other -- some of them (for example) causing my fingers to bounce off of a keyboard. We can eliminate all talk of 'value' and delegate it to the realm of myth like our talk of Pegasus and Zeus. When we look inside ourselves to determine the value of things, we are looking at something that we make up -- something in the realm of "let's pretend".
(3) Fact/value (is/ought) dualism. Whatever values or 'oughts' are, they exist in a realm distinct from fact, and yet have an amazing ability to interact with it. Somehow, these 'values' (which are not facts) can alter the flow of material in the physical world, causing actions that would not have otherwise happened. Somehow, some non-fact entity called a 'value' is, at this very moment, causing me to write this article. And yet this entity exists nowhere in the world of 'fact'.
Of these options, (3) is the only option that is compatible with the traditional interpretation of a hard fact/value or is/ought distinction. Yet, (3) is much more problematic than (1). Particularly, if we have a (1) available, where we have an 'is' statement that seems to be able to play the same logical role as any 'ought' statement -- a fact proposition that seems indistinguishable from a value proposition capable of operating in the real world and influencing real-world events.
To accept Hume's 'is/ought' distinction as law, is to accept fact/value dualism as proved. That is a very difficult conclusion to accept.
(C) G.E. Moore: 'Is Desired' vs. "Ought To Be Desired"
G.E. Moore raised an objection against John Stuart Mill that is relevant here.
Mill argued that all value is based on desire. Moore raised the criticism that Mill's view fails to adequately distinguish between what is desired, and what ought to be desired. Certainly, there are a great many things that a person may desire, which are not to be counted among what ought to be desired, with sadistic desires perhaps topping the list.
The account that I am presenting here respects that distinction, because it recognizes that desires not only have the capacity to be fulfilled or thwarted, but that desires have the capacity to fulfill or thwart other desires. Therefore, it is sometimes reasonable to ask, "What desires is it good for us to have?" or "What should we like?" This is a question about whether a particular desire -- a desire that we have the capacity to choose (in the same sense that we choose where to live to and what career to go into) -- is a desire that we should choose.
The only way that an entity can have value in the real world is to be "such as to fulfill the desires in question", but there is nothing in this that conflicts with the possibility of a desire being "such as to fulfill (or thwart) other desires".
(D) Summary
So, now, we have gotten around the objections that spring from traditional interpretations of Moore and Hume. Though, actually, if you read Hume carefully, you may come to the conclusion that Hume would read this and say, "Yeah! That's what I was talking about! Finally, somebody understands what I meant, that value requires looking at the passions -- at desires -- and that whether or not something is such as to excite the passions (or to fulfill a desire, if you insist on using your 21st century dialect) is the only type of fact from which we can derive 'oughts'." It is not an unreasonable interpretation of Hume.
VII. Utilitarianism
I have talked a lot so far about desires, and their relationship to value. This is half of the issue of desire utilitarianism. It is time that I get to the 'utilitarianism' part of the equation.
I have already said one thing about this form of utilitarianism that is important to keep in mind. We are not seeking to maximize pleasure over pain, or happiness over unhappiness. We are seeking to maximize desire fulfillment over desire thwarting. The purpose is to make true the propositions that are the objects of desire, and to focus more on those propositions that are the objects of stronger desires.
To get a closer look at what else desire utilitarianism requires, I would like to look at its chief rival, 'act utilitarianism'. Specifically, I want to identify the chief problem with act utilitarianism, and illustrate how desire utilitarianism solves that problem.
(A) Act Utilitarianism and Desires
Act utilitarianism states that one should do that act which produces the highest utility (desire fulfillment).
Now, there is a principle of morality that says 'ought' implies 'can', and that 'cannot' implies 'it is not the case that one ought'.
So, for example, you are at the site of a fire. There is a child trapped inside. If you could teleport the child out of the building to some place safe, then you ought to do so. But no moral system can sensibly label you a sinner, a perpetrator of evil, because you do not do that which is not causally possible. So, among the things that you ought to do, teleporting the child out of the building is not one of them.
Now, above, I wrote the formula:
(Belief + Desire) -> Intention -> Intentional Action.
If you say that a person ought to do X, then this is only possible if the causes are in place for the agent to do X. In other words, "Ought to have done X" implies "Ought to have had those desires that would have caused one to do X."
Insisting that the action change, while the causes of the action remain constant, falls in the same category as insisting that a person teleport a child out of a burning building. It violates the principle of 'ought' implies 'can'.
However, if we change the desires, we also have to look at the consequences of those desires. Desires do not spring into and out of existence on a whim. They are persistent entities, and so the same desire that would motivate a particular action at one time would be there to influence other choices at other times -- and not necessarily for the better. So, if we are demanding that a particular desire be present at time T1, we had better make sure it is a desire that we hope not to be present at time T2, T3, and T4.
(B) Impossibility of Act Utilitarianism
There is only one type of person who can always choose to do that act which produces the most utility. That is a being with only one desire -- a desire to produce the most utility. If he has any other desire -- any desire at all for anything else -- then there will be instances where that desire will conflict with the desire to maximize utility, and will cause the agent to do something other than the act-utilitarian best act.
Let us look at an example, a person with a strong desire for alcoholic drink. Let's make this a particularly strong desire -- so much so that, if offered a drink for detonating a nuclear weapon in another city, he would detonate the bomb to collect his drink.
Let us now weaken the desire. Maybe he won't detonate the bomb, but he would kill somebody. A little weaker -- he would take money from his place of business. A little weaker -- he spends a substantial portion of his disposable income on liquor.
We can continue to weaken this desire, but there is only one point at which the desire for alcohol will not, under certain circumstances, cause the agent to choose a drink at the expense of maximizing happiness. That is at which he has no desire for alcohol.
Act utilitarianism not only requires no desire for alcohol, it requires no desire for anything other than to maximize utility. If the agent likes the taste of steak better than hamburger, then there will be an instance in which he will sacrifice maximum utility for a steak. If he has a strong preference, it will have the same effect as a strong preference for alcohol. If he has an aversion to pain, a desire for sex, a particular interest in the well being of his children, there are instances in which she will sacrifice her desire to maximize utility to obtain fulfillment of any of these other desires.
I hold that a moral commandment to act as an act-utilitarian is no different than a commandment to alter the gravitational constant to a number that maximizes utility, or a commandment to move the Earth to an orbit that would produce a more pleasing climate. If it cannot be done, there is no sense in saying that it ought to be done.
(C) The Utility of Desire
Now, I want to combine this with another consideration. Let us take a person, and say that he has two choices to make.
Choice 1 (C1) is between option 1 (O1) and option 2 (O2). Let us say that O1 will produce more utility than O2. This means that, according to act-utilitarian theory, the person should choose O1.
Choice 2 (C2) is between option 3 (O3) and option 4 (O4). Let us say that O3 will produce more utility than O4. This means that, according to act-utilitarian theory, the person should choose O3.
So, ultimately, the person should choose O1 and O3.
However, actions are caused by desires. Let us say that the following desires are possible. (Recall, I suggested earlier that morality is concerned with choosing desires, and this only makes sense in the context that desires can be modified by environmental factors, such as praise, blame, reward, and punishment.)
Desire D1. An agent who has desire D1 will be motivated to choose O1 in the first case, and O4 in the second case.
Desire D2. An agent who has desire D2 will be motivated to choose O2 in the first case and O3 in the second case.
Desire D3: An agent who has desire D3 will be motivated to choose O2 in the first case and O4 in the second case.
A desire that will motivate a person to choose O1 in the first case and O3 in the second case is not causally possible. In the previous section, I argued that there is no such thing as an act-utilitarian -- that only a person who desired nothing but to maximize utility will always choose the act-utilitarian best act. This says the same thing -- that it is not possible for a person to have the desires that will cause them to do O1 and O3.
Let us look at the three possible desires, and decide which desire produces the most utility.
We can already see, just by looking at what we know, that D3 is out of the running. However, the case so far leaves it open whether or not D1 is better than D2. For the purposes of this analysis, let us look at cases where D1 is better than D2.
What does this lead to?
Case 1: We see a person faced with the O1 vs O2 option choosing O1 rather than O2. This means he has desire D1. We praise him (and let everybody know that people who choose O1 will be praised). Through this, we strengthen desire D1.
Case 2: We see somebody else choosing O2. This person either has desire D2 or D3. We don't know which, but we know that both desires are desires that we want to inhibit. So, we condemn the person who chooses O2. In this case, and the case above, our praise or condemnation follow the act-utilitarian model, giving act-utilitarianism some measure of accuracy.
Case 3: We see somebody who must choose between O3 and O4, and he chooses O3. O3 is the option that will maximize utility in this instance. However, only desire D2 (which is incompatible with choice O1 in other instances) can bring about D2. So, this person shows evidence of a desire that we wish to inhibit. This person, who is performing the act-utilitarian best action in this case, is to be condemned. In condemning this action, we inhibit desire D2, and strengthen desire D1 -- the desire that leads to the action that would be praised.
Case 4: Another person, who must choose between O3 and O4, chooses O4. Maybe he has desire D1. Maybe he has desire D3 (the worst of the three options). We cannot know this. Though we hold that he ought to choose O4, we visit those who choose O4 with qualified praise. He did what he had to do.
Now, I want to focus on Case 3 for a moment.
Doesn't this describe every type of intuitive objection ever raised against act utilitarianism?
There is the case of the sheriff who must decide between protecting an innocent prisoner and precipitating a riot where dozens of people will almost certainly be killed or wounded, or the Los Angeles jurist who must decide whether to acquit a police officer accused in the beating of a motorist named Rodney King (a police officer who, in the sincere opinion of the jurist, has not been proved guilty), even though he knows it will touch off a riot where he expects 50 innocent people to be shot, stabbed, or dragged out of their cars and clubbed to death.
We can easily see this as a case of a sheriff or a jurist faced with an "O3 or O4" choice in the model described above. In that model, we can easily identify a long list of O1 vs. O2 choices.
Again, given that there is no such thing as an act utilitarian, sheriffs are going to have other desires, such as a desire to drink (some of them), a preference for steak over hamburger (or vice versa), a desire to be home with their loved ones, a desire for a new car, a desire for the approval of the townspeople.
The agent who will sacrifice the interests of the prisoner to fulfill a desire to maximize utility in this case is at risk of sacrificing the interests of the prisoner to fulfill other desires the agent might have -- the desire for a good stiff drink, for a steak dinner rather than a hamburger, to get home with their loved ones, for money, for the approval of the mob -- in other cases.
I am going to suggest that the desire that we want the sheriff to have in this case is a desire to execute the will of the Court. We will not want the sheriff to have a desire that will also inhibit him from turning the prisoner over to the proper authorities -- the representatives of the state prison system -- when they come for the prisoner. So, a desire to protect the prisoner would not suffice. But, a desire to execute the will of the court will protect the prisoner from the mob, and from the sheriff's own desire for steak or a new car, while motivating him to stand aside for the prison officials when they show up.
The desire to execute the will of the court is our D1. We promote such a desire (or aversion) through the practice of praise, blame, reward, and punishment, focusing on instances that show evidence of D1.
Our "O1 vs. O2" choices are very common. Desire D1, to execute the will of the court, if sufficiently strong, will cause the agent to choose executing the will of the court even when it means hamburger instead of steak, a night at work when necessary instead of a night at home with his loved ones (though, certainly, will allow him to spend time with his loved ones when the will of the court can be served without him being there), and to turn down bribes and other forms of corruption.
However, in doing this, we have to accept the implication that there will be times when the aversion causing him to avoid sacrificing the interests of prisoners will also have to be strong enough to override the desire to maximize utility -- instances where the agent will choose O4 over O3.
It is very much relevant to note that these "O1 vs. O2" choices are far more common than the "O3 vs. O4" choice that the critic of act utilitarian raises -- the choice between protecting the prisoner from the angry mob or preventing the riot. And, in fact, in all of these counter-examples, it is no strange coincidence that the "O3 vs. O4" option the critic raises are rare to the point of nonexistence -- be it a doctor who must decide killing a healthy patient to use his organs to save five others, or the visitor to a third world country who must kill one innocent villager to prevent the murder of 20 others.
Every one of the common objections to act utilitarianism can be put into this formula. They are all cases where the critic points out a case where an agent has a choice to make between some hypothetical Option 3 or an Option 4, where act utilitarianism demands Option 3, but which morality seems to demand an Option 4.
They are all cases where the critic correctly points out the fact that we cannot bring the agent to choose Option 3 without altering the agent's desires. And, of those desires are altered, we risk common, every-day choices between an Option 1 and an Option 2, where agents will then choose Option 2. All for the sake of making the 'right choice' in some hypothetical situation which will occur only rarely (if ever) in the real world.
VIII. Conclusion
So, here, we have a theory which is compatible with the most widely accepted theory of human action -- that they are caused by beliefs and desires, and humans act as if their beliefs are true, and as if the propositions that are the objects of their desires are to be made or kept true. It is a theory that respects both the fact that actions aim to fulfill the desires of the agent, but that the agent can desire the well-being of others.
It is a theory that can be applied to every value-laden concept, from illness and injury, to usefulness, to beauty and ugliness, to moral concepts such as 'duty' and 'right'.
It is a theory that explains the distinction between fact and value -- between 'is' and 'ought' -- without reducing to the absurdity of fact/value dualism, while respecting the fact that values are dependent on desire. Value claims are that subset of fact claims that relate objects of evaluation to desires, so there is no value without desire, but insofar as desires exist in the world of fact, then so do values.
It is a theory that takes all that has merit in act utilitarianism, but which avoids the classic problems. It handles the counter examples by recognizing that you can't change an action without changing the desires that caused it, and a change in desires may motivate worse choices elsewhere.
I think that this theory has a lot to recommend it, and I invite you to give it due consideration.
References:
Adams, R.M. 1976. "Motive Utilitarianism", Journal of Philosophy, 73: 467-81.
Hume, D. 1969. A Treatise of Human Nature. Edited by Ernest C. Mossner. London: Penguin Books.
Martineau, J. 1901. Types of Ethical Theory, Third Edition, Revised. Clarendon Press, Oxford.
Mill, J. S. 1998. Utilitarianism, edited with an introduction by Roger Crisp. New York: Oxford University Press. Originally published in 1861.
Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press.
Sidgwick, H. 1907. The Methods of Ethics, Seventh Edition. London: Macmillan. First Edition 1874.