Here is a simple and straight forward question that I got from Luke Muerhlhauser's studio audience.
Is Desirism a descriptive or a prescriptive theory?
The rule that links the two is that the only legitimate prescription is an accurate description of a relationship between some potential state of affairs and reasons for action that exist. Absent the details, it says that all legitimate prescriptions are descriptions.
Now, by a prescription I am referring to any should or ought statement. "Jim should go see a doctor." or "We should recognize and respect a separation between church and state," or "You ought to pay Alice back the fifty dollars you borrowed last week."
Any should or ought statement invites the question, "Why should I?" or "Why is it the case that I ought to?"
From this question, the only legitimate answer is a reason for action. In fact, the only legitimate answer is a real reason for action.
If you give an answer that is not a reason for action, then do not be surprised if the person you are talking to simply stares back at you in confusion.
"You should comb your hair before it dries."
"Because water has a freezing temperature of 0 degrees Celsius."
You need a reason for action to answer a "Why?" question. Not only that, you need a reason for action that bears the right kind of relationship to the state of affairs you are evaluating.
"You should comb your hair before it dries."
"Because I know you hate to eat fish."
"What? If I don't comb my hair before it dries then I will have to eat fish?"
And you need a reason for action that is real.
A person might give his answer by referring to God's will, categorical imperatives, a reason for action that is an inherent property of some states of affairs themselves, some hypothetical social contract, the wishes of an impartial observer, choices made by substantially ignorant people behind a veil of ignorance, and the like. However, these reasons for action do not exist. Therefore, none of them provide a legitimate answer to the "Why should I?" question.
Desires are the only reasons for action that exist.
Regular readers might be tired of hearing this, but a great deal is built on this.
Beliefs and desires are propositional attitudes - they are attitudes that an agent takes towards a proposition. A"belief that P" is the attitude that P is true - that P is an accurate description of the world. A "desire that P" is a motivational attitude that pushes the agent to realize states of affairs in which P is true. A parent's desires "that my child be healthy" is a motivational attitude that pushes the parent to act so as to realize states of affairs in which the proposition "my child is healthy" is true.
A "desire that P" is fulfilled in any state of affairs in which P is true.
So, now, a legitimate prescription is an accurate description of a relationship between a state of affairs and a set of desires such that it relates the propositions that are true in a given sate of affairs to the propositions that are the objects of real desires.
"You should comb your hair before it dries that way otherwise when you go out in public you will be subjecting yourself to ridicule and you hate ridicule. Even if you do not hate ridicule, people tend to treat those who look ridiculous with less kindness and respect, and you may find you have reasons to want their kindness and respect."
Now, we have a description of how a state of affairs relates to a set of reasons for action that exist - specifically, the desires of the agent we are talking to.
"You should comb your hair before it dries that way" is both a prescription and, at the same time, a description of how a particular state of affairs relates to a set of reasons for action that exist.
A moral statement, like any other prescription, has to follow this same formula. Any moral statement that is a legitimate prescription must also be an accurate description of a relationship between some object of evaluation and real reasons for action. Any theory of morality that deviates from this formula has wandered into a world of make-believe. Their moral prescriptions are as fictitious as the reasons for action they invent.
I have proposed that the states of affairs that moral 'oughts' are primarily used to evaluate have to do with the strength and prevalence of malleable desires. Malleable desires are those that can be molded using social tools such as praise and condemnation. A moral 'ought' prescribes malleable desires to the community at large.
It is important to note that they are not prescriptions for the given individual one is talking to. They are prescriptions for the community as a whole. As such, they consider all of the reasons for action that are real. A moral claim is not really a claim about what "you ought to do", but a claim about "what any member of the community in your situation should want to do," where praise and condemnation are used as tools to help bring it about that it is something that people generally do want to do.
At this point, the task is to come up with the relationship between an object of evaluation and a set of desires that best accounts for the way a term is used in public discussion.
Applying this methodology to the term 'health', we note that this term is used to evaluate mental and physical functioning. An accurate prescription in the health department is an accurate description of a relationship between some state of physical or mental functioning and the desires of the agent. A form of functioning that tends to thwart the desires of those agents who have them is an 'illness' if its cause is minute and impossible to recognize by simple means. If its cause is easy to recognize - like getting trampled by a horse or falling off a ladder - we use the term 'injury'.
If we study the term 'useful', we apply it to any number of things. The desires or reasons for action that are relevant to the usefulness of something are determined by the context of the sentence in which the term is used. However, we only use the term to refer to indirect relationships between states of affairs and desires. That is to say, something is useful only insofar as it has the ability to bring about something else - and that something else has true propositions that are the objects of those desires being referred to in context.
Looking at the way people use the term 'beautiful', for some reason this is only used to evaluate things heard or seen - not things that are smelled or touched or tasted. Furthermore, it is applied to states that fulfill the desires directly. Looking at them, or hearing them, appeals to us not because of what we can use them for, but in themselves.
The relationship that makes the best sense of how moral terms are used - without wandering off into the realm of fiction and fantasy - is the relationship between malleable desires (desires that can be molded using forces such as praise and condemnation) and all other desires. A moral 'ought' is not a prescription made to an individual. It is a prescription made to the community. It is not a statement about "what you ought to do", though it is often expressed that way. It is a statement about "What people in the community generally have reason to cause any person in your situation to want to do."
It is often brought up that we have the power to take the term 'moral' and apply it to something other than this. However, we also have the power to take the terms 'health', 'injury', 'illness', 'useful', and 'beauty' and apply them to other things as well. We also have the capacity to take the terms 'atom', 'malaria', 'planet', and every other term in our language and apply them to something else.
It is true that we can take the term 'moral' and apply it to something else. It is also not important. The person who thinks that this is important - that this reveals some significant secret about morality - does not understand language. If this were some great discovery proving the subjectivity of morality, then it also proves the subjectivity of chemistry, medicine, astronomy, and every other field of study.
So, is desirism a descriptive theory or a prescriptive theory? Well, any theory that separates the two - that claims that the two are mutually exclusive - can be thrown out. Those theories are dealing in the realm of make-believe. Legitimate prescriptions are accurate descriptions of relationships between states of affairs and reasons for action. If your moral prescriptions are not, at the same time, accurate descriptions, then they are fictions that have no relevance in the real world.