N. M. L. Nathan's argument that IDP utilitarianism, if universally adopted, is inconsistent, does not succeed. The argument requires that if an IDP utilitarian has only self-regarding desires, then none of these desires can be informed. This rests on a partial misuse of the expression 'satisfaction of desire'. For an individual attempting to realize his self-regarding desires, the satisfaction of the 'satisfaction of a desire' is unmeaning. The naming of an object of the desire is an intrinsic part of the phrase 'satisfaction of desire'. Further, contrary to Nathan's claim, this suggestion does not trivialize IDP utilitarianism.
In 'Projectivist Utilitarianism', ~ N. M. L. Nathan offered an argument against 'IDP utilitarianism'. A person of this belief takes utility as the satisfaction of informed desires, and also accepts a projectivist doctrine of evaluation in general. As a utilitarian he thinks that only utility is valuable in itself, so as a projectivist he wants nothing to exist as an end but utility. And so, interpreting utility as he does, he wants nothing to exist as an end but the satisfaction of informed desires .... I argue that if everyone were an IDP-utilitarian, no IDP would know that any of his own desires were informed. (p. 207)
An informed desire is one not based on a false belief, characterized in accord with three principles Nathan sets forward at pages 207-208.
Nathan's proof of his contention distinguishes several different cases. One of these is that (supposing that everyone is an IDPutilitarian), "N desires that p as an end (NDEp)" (p. 208). About this case, Nathan states: "Since N is an IDP-utilitarian, the only kind of thing he desires as an end is the satisfaction of desires. So if N desires that p as an end, p is that some other desire, either one of his own or one of someone else's, is satisfied. Let us call the object of this other desire pl" (P. 208).
So far, Nathan is perfectly correct. But he seems to me to fall into fallacy at his next step:
We can then say that if N desires that p as an end (NDEp) then what he desires as an end is that Nt's desire that p~ is satisfied, (NDSN1Dp~) where N~ may or may not be identical to N. Now either N does actually desire that Pm or not. If not, then by (H3) N's desire that p is based on a false belief, which runs counter to our original hypothesis.