The following is my most recent attempt to summarise my dissertation in a couple pages. I am only just approaching the half-way mark, so expect several revised versions of this summary in the near future.
My dissertation attempts to make sense of the idea that desires may be correct or incorrect by articulating and defending a version of the claim that desires aim at the good.
There is a widely held intuition that some desires are infelicitous, bad, or perverse. For example: there is something infelicitous about the desire to drink a can of oil in order to quench one’s thirst; there is something bad about the desire to take the life a known innocent; and there is something perverse about the desire to stick a sharpened pencil in one’s eye even though one believes no good could come from doing so and one recognizes that it would be extremely unpleasant. The first desire seems infelicitous on instrumental grounds; drinking oil is a poor way to quench one’s thirst. The second desire seems bad on ethical grounds; taking the life of a known innocent is morally wrong. The third desire seems perverse on hedonistic grounds; all things being equal, we would expect an agent to avoid unpleasant experiences. One way to capture the idea that a desire may be infelicitous, bad, or perverse—a theoretical proposal that is tied to a longstanding philosophical tradition—is to say that desires aim at the good, and that a desire is inappropriate, bad, or perverse just in case it fails to realise its aim. On this view, the good is the most abstract characterisation of aim of desire. Let us call this proposal the guise of the good theory of desires (henceforth, GG theory).
I wish to articulate and defend a plausible version of GG theory. There are three influential strategies for making sense of GG theory currently found in the literature: the Desire-as-Belief Thesis, the claim that the desire to φ is equivalent to the belief that φ is good; the Desire-plus-Belief Thesis; the claim that the desire to φ is always accompanied by the belief that φ is good; and the Desire-as-Perception Thesis, the claim that the desire to φ is equivalent to perceiving that φ is good. I argue that all three proposals are unacceptable as ways of making sense of GG theory. Instead, I argue that a desire plays the same role in our deliberation as being the recipient of a (self-issued) command, order or request. Let us refer to this proposal as the Desire-as-Imperative Thesis.
Significantly, the notion of an imperative implicated in the Desire-as-Imperative Thesis should not be confused with the Kantian notion of an imperative (i.e., a dictate of pure reason). Rather, the word ‘imperative’ is meant to pick out the category of non-assertoric speech-acts that is typically expressed using the imperative mood of English grammar; a category that includes orders, requests, commands, and entreaties. I hold that desires are like the class of speech-acts that are typically expressed by the imperative mood in at least three respects. First, like speech-acts in the imperative mood, desires are not truth-evaluable. For example, both the desire to close the front door and the request to close the front door is neither true nor false. Second, like speech-acts in the imperative mood, an agent assents to a desire, not by forming a belief, but by forming an intention. For example, one assents to the desire to close the front door or the request to close the front door by forming the intention to close the front door. Third, like speech-acts in the imperative mood, desires are governed by norms that determine if it would be correct or incorrect to assent to them. For example, both the desire to close the front door and request to close the front door are correct just in case it is good to close the front door. On this view, when we say desire aims at the good, we mean that the desire to φ is correct just in case it would be good to φ.