Desirism is built on a theory of value that holds that value exists in the real world as relationships between states of affairs and desires. Different types of value claims describe different types of relationships between different states of affairs and desires.
On this account, value claims are objectively true or false. The relationships exist as described or they do not. Furthermore, whether a value claim is true or false can be substantially independent of the beliefs or sentiments of the person making the claim.
At the same time, there are no objective values in the sense of values independent of desires.
Note: the fact that it is possible for value claims to be objectively true even though there are no objective values is just one example of the terminological quagmire that moral philosophy must navigate. Another example is that moral philosophy allows intrinsic value to depend essentially on extrinsic properties. We will get to this case further below.
Specifically, any true value claim contains four elements:
First, it makes reference to some object of evaluation. It is going to take some state of affairs (e.g., one in which some painting exists, one in which somebody experiencing a sunset, or one in which a prisoner is being tortured) and identify it as "good" or "bad". In all true value claims, something is being evaluated.
To call something "good" or "bad" in the generic sense implies that some reason exists for acting in a way that realizes or avoids realizing that state of affairs respectively. Good things are to be realized. Bad things are to be avoided.
Desirism holds that desires are the only reasons for action that exist. Other types of reasons for action that have been proposed - divine commands, intrinsic values, categorical imperatives - none of them are real. Therefore, any claim that they provide agents with reasons to act so as to realize or prevent the realization of some state of affairs is false.
Thus we get the second component of true value claims - they make reference to a set of desires and describe a relationship between the object of evaluation and that set of desires.
What is the nature of this relationship?
There are two types of relationships between states of affairs and desires - direct and indirect.
In both cases, we begin with the claim that desires are propositional attitudes. That is to say, they can be expressed as a statement that takes the form, "Agent desires that P" for some proposition P.
A desire that P is objectively satisfied by any state of affairs in which P is true.
So, when we evaluate a state of affairs S, relative to some desire that P, we are looking at whether P is true in S. When P is true in S, an agent with a desire that P has a reason to act so as to realize S.
Note that, among the set of possible desires, an agent may have a desire that not-P (or, in other words, an aversion to P). If an agent has a desire that not-P, and P is true in S, then, in this case, the agent has a reason to prevent the realization of S.
This leads to the third component in the concept of value: Does the object of evaluation objectively satisfy those desires?
For the agent with a desire that P, is it the case that P is true in S? If P is true in S, then S objectively satisfies the desire that P and the agent has a reason to act so as to realize S. If, on the other hand, P is not true in S, then the agent has no reason to bring about P. Finally, when the agent has a desire that not-P, and P is true in S, the agent has a reason to act so as to prevent the realization of S.
All of these are accounts of direct relationships between a desire and an object of evaluation.
Many authors refer to this as "intrinsic value". However, this easily leads to confusion. Another common definition of "intrinsic value" is "value independent of all extrinsic properties". Yet, we see that direct value in this sense very much depends on a set of extrinsic properties - it depends on desires.
One way to avoid this confusion is to call the relationship above "direct value", and to leave the term "intrinsic value" for "value independent of extrinsic properties". In this sense, intrinsic value does not exist. However, direct value exists.
Some people want to call value independent of extrinsic properties "objective value." J.L. Mackie had this definition in mind when he argued that there are no objective values. He goes to great pains to make it clear that he is talking about objective, intrinsic prescriptivity.
This gets confusing because we can certainly make objectively true or false claims about the relationships between things - that is to say, about extrinsic properties. "Objective values" in Mackie's sense, do not exist. However, objectively true value claims are still possible - as Mackie himself admits.
Direct value, as described here, is to be contrasted with indirect or instrumental value.
An instrument is a tool that can be used to bring about or preserve something else that has value. An agent has a desire that P. P is true in S. In order to realize S, the agent needs to use tool T. In this situation, T has instrumental or indirect value. The agent has a reason to act so as to acquire T.
Objects are not the only things that have instrumental value. Institutions, laws, relationships, and just about anything can be useful in bringing about a state S where P is true in S.
Desires themselves are useful in this sense - they have value as tools. There are some desires which, if common in others, make it easier for an agent to realize S. Insofar as an agent has a desire that P, and P is true in S, and desire D1 helps to bring about S, the agent has a reason to promote desire D1.
Similarly, there are some desires, if common in others, that make it more difficult to realize S. Insofar as an agent has a desire that P, and P is true in S, and desire D2 commonly gets in the way of bringing about S, the agent has a reason to inhibit desire D2.
This, then, is the fourth component of value. Does the object of evaluation objectively satisfy a desire directly (have direct or "intrinsic value")? Or is it something that can be used to realize a state of affairs that objectively satisfies a desire (has instrumental value)?
In summary, then, four questions get answered in any true value claim.
(1) What are the relevant objects of evaluation?
(2) What are the relevant desires?
(3) Do the objects of evaluation objectively satisfy or objectively dissatisfy those desires?
(4) Do they objectively satisfy or dissatisfy those desires directly or indirectly?
All value-laden terms can be evaluated according to this set of criteria. Moral terms (e.g., right, wrong, good, evil, obligatory, permissible, prohibited) are one set of value-laden terms among many. Moral terms have their own set of answers to these questions. Other value-laden terms (e.g., useful, beautiful, healthy, illness/injury, dangerous) have a different set of answers to these questions. We can distinguish different value-laden terms by identifying the differences in how they answer these four questions.