This is the cardinal mistake of desirism - it classifies desires as abstract as "good" or "bad" according to arbitrary criteria, instead of addressing the real desires of real humans.
This is a common objection that I hear. It can be found in another recent comment from a member of the studio audience:
[Y]ou have said that when we consider moral questions, the desires in question become "all desires that exist." Why? . . . How do you get from practical-ought to the moral-ought, without already instilling your own values into the equation (the value of considering all desires that exist)?
The fact is that I can drop the terms 'good' and 'bad' - and moral terms like 'good' and 'evil' completely out of this theory and it will not change the theory whatsoever. The terms 'good' and 'bad' are merely code words.
When the first person pointed to a rock and said 'rock', he set up the question, "Okay, now, what else is going to count as a rock?" There is no law of nature that dictates what the word 'rock' means. People simply make these things up. They decide, "Heck, it sure is convenient, when w want to talk about these naturally occurring hard things that seem to be everywhere, to use the term 'rock', so that's what we are going to do."
Every word that exists is assigned its meaning by means of arbitrary criteria.
Do you think that there is some experiment that people used to determine that a 6-proton atom would be called 'carbon?' It is not as if they peered through a carbon atom through a microscope and read the name off of it. Chemists decided, quite arbitrarily, that they would assign the word 'carbon' to the 6-proton atom. They did not prove that 6-protons are properly called 'carbon' by means of any natural experiment. They simply made an arbitrary decision, and agreed among themselves to follow it.
To accuse me of doing the same thing that geologists and chemists do does not provide me with any reason to reject any theories that I may have about the nature of rocks or of atoms or of morality.
If this is to be a meaningful objection, then the person who raises it cannot be accusing me of arbitrarily assigning meanings to words. That is a necessary part of language - and anybody who participates in the institution of language necessarily participates in the same arbitrariness. If I am guilty of doing the same thing that geologists, chemists, and physicists must necessarily do, then I am just as guilty as the geologists, chemists, and physicists.
Instead, a meaningful objection must be one in which I am accused of arbitrarily assigning a property to an object or state of affairs - a property that it does not have as a matter of objective fact.
Yet, if this is the accusation being made, the accuser has the obligation of identifying the property that I am allegedly assigning arbitrarily to some object or state of affairs. They have to make an accusation that takes the form, "When you apply the term T to state of affairs S, you are claiming that S has property P. Yet, the assignment of property P to S is arbitrary and cannot be demonstrated as a matter of objective fact."
That would be a problem; it would be a particularly significant problem for me because I hold that arbitrarily assigned properties do not exist. The only realm where properties can be arbitrarily assigned or removed from objects or states of affairs is in the realm of fiction - not the realm of fact.
Note that I have written several times that I do not believe that there is a mutually exclusive is/ought distinction. I think that such a distinction is nonsense. It smacks of dualism - of the existence of an odd realm of existence distinct from yet capable of interacting with the realm of what is. Instead, I hold that there is and is/is not distinction. Either moral properties fit into the realm of what in, or we must assign them to the realm of what is not and move on.
In light of that assumption, the accusation that I am guilty of arbitrarily assigning some property to some state of affairs would be particularly damaging.
If the accusation that I am guilty of arbitrarily assigning a property to an object or state of affairs, then name that property.
If the accusation that I am guilty of expressing my theory by using language (the practice of arbitrarily assigning definitions to symbols), then, I am as guilty as any geologist, chemist, and logician.