In his paper, “The Guise of the Good”, David Velleman argues that desires aim, not at the good, but at the attainable. Velleman begins by drawing two important distinctions. Firstly, he distinguishes between cognative attitudes (i.e., attitudes in which a proposition is grasped as patterned after the world), and conative attitudes (i.e, attitudes in which a proposition is grasped as a pattern for the world to follow). Examples of cognative attitudes include beliefs, assumptions, and imaginings. Examples of conative attitudes include desires, hopes, and wishes. Secondly, he distinguishes between the direction of fit of an attitude (i.e., that in virtue of which it is a cognative or conative attitude) and the constitutive aim of an attitude (i.e., that in virtue of which it is correct or incorrect).
With the preceding pair of distinctions in place, Velleman then advances the following three-stage argument: First, he argues that the constitutive aim of belief is what distinguishes it from all other states with a cognitive direction of fit; namely, the fact that beliefs are correct just in case they are true. Second, he argues that what distinguishes desire from all other states with a conative direction of fit is not the fact that it aims after the good, since this is something it shares with all other conative states. Third, he argues that what distinguishes desire from all other conative states is the fact that desire aims at the attainable. He therefore concludes that desires aim at the attainable.
Stage 1: Velleman on Constitutive Aims
Velleman takes as his starting point the stipulation that the constitutive aim of a psychological state-type is whatever sets it apart from all other psychological state-types; to wit, what makes a psychological state-type the state-type that it is. Hence, to say that belief aims at the true, according to Velleman, is to say that the truth-aim is what distinguishes belief from all other states with a cognitive direction of fit. Velleman motivates this claim by exploiting a comparison of believing that p, on the one hand, and fantasising that p and assuming that p, on the other. Velleman notes that all three psychological states have a cognitive direction of fit; all three, according to Velleman, involve a proposition being grasped as patterned after the world. However, while believing that p is deemed correct if and only if p is true, fantasising that p and assuming that p are not deemed correct if and only if p is true. Thus, by Velleman's lights, it is the fact that a psychological state-type has the correctness conditions that it does that makes it the psychological state-type that it is.
Stage 2: Velleman's Negative Thesis
Next, Velleman argues that the good cannot be the constitutive aim of desire since desires are not the only psychological states that aim at the good. Velleman takes as his point of departure the assumption, widely held by proponents of GG theory, that desires aim at the good in virtue of their direction fit. On this view, to say that desires aim at the good just is to say that desires have a conative direction of fit. However, Velleman points out that having a conative direction of fit is something that desires have in common with all other conative attitudes, including wishes and hopes. The upshot is that all conative attitudes may be said to aim at the good. Given that the constitutitive aim of desire is what makes it the psychological state-type that it is, it follows that the good cannot be the constitutive aim of desire. We may call this Velleman’s negative thesis.
Stage 3: Velleman's Positive Thesis
Finally, Velleman argues that the constitutive aim of desire is the attainable. By this, Velleman does not mean that one can desire that p only if p is attainable. Rather, he means that one can desire that p only if one believes that p is attainable. Thus, Velleman allows that I may desire something that is in fact unattainable, if I mistakenly believe it to be attainable. For example, I can desire that I arrive in the airport at 3pm, even though it is not physically possible for me to arrive in the airport at 3pm, so long as I believe (albeit mistakenly) that it is physically possible for me to do so. Velleman observes that one can wish for something that one believes to be unattainable. For example, I can wish I were never born even though I believe that undoing my own birth is not something that is attainable. However, according to Velleman, one cannot desire something that one believes to be unattainable. Hence, I could not desire that I were never born if I believed that undoing my own birth was not attainable. According to Velleman, this distinction between desires and wishes generalises to all other conative states; only desires aim at the attainable. Thus, Velleman concludes that desires stand to the attainable as belief stands to the true. We may call this Velleman's positive thesis.