A lot of philosophers, apparently including Hume of all people, have embraced the doctrine that we always choose under the guise of a good. There are two ways of taking this doctrine:
- If x opts for A, this is because A is in fact a good.
- If x opts for A, this is because A is taken to be a good.
- It is unclear how something's actually being a good can motivate—surely even if something is a good, it only motivates through the intermediary of the agent's taking it to be a good, and
- there seem to be too many counterexamples to (1)—people can opt for such non-goods as racial purity, enlightenment of astral selves, revenge, etc.
First, suppose Patricia believes that drinking a potion will create an astral lung in her. This by itself is not enough for her to act on according to the guise of the good theory. On version (2), what we need to add is that the astral lung is believed by her to be good. But I think this by itself is not enough. We have all kinds of beliefs, some of which are buried below much mental detritus, and which it would take a fair amount of careful work to recall. It's not enough to just have the belief buried somewhere in order to be able to act on it. The belief must be involved. Involved where? Well, there is a distinction between the will and the intellect. It is not even enough that one be occurrently thinking the belief while acting. (This is an important point, but the best argument I have right now is to invoke split brain patients—if I occurrently think p in respect of my left hemisphere while I am a split brain patient, that does nothing to help me act on account of p in respect of my right hemisphere.) The belief must enter into the will. But of course the will doesn't believe—it pursues. So the belief must enter into a description of what is pursued. In other words, it is not just that Patricia happens to believe that it is good to have an astral lung and simultaneously pursues getting one, but that Patricia pursues a valuable possession of an astral lung qua valuable. That it is valuable is a part of the description. (Most likely, it is not just "valuable" that is part of the description, but something thicker, such as "spiritually uplifting".) But then Patricia not only pursues an apparent good, but she pursues a genuine good, something which it would be actually good to achieve. For the valuable possession of an astral lung would be, tautologically, valuable to have!
I am inclined think that this pattern is always found when an agent seems to be motivated by the mere appearance of the good. The will wills things under the guise of the good, and hence the good—either thinly as "the good" or thickly as "the courageous thing" or "the generous thing" or whatever—enters into the description under which one wills. Those who pursue racial purity do not merely falsely believe racial purity to be good. They seek, instead, a society that is well-ordered in being racially purity. Their goal is impossible of achievement, because (i) racial purity cocneptually breaks down, and (ii) even if racial purity did not conceptually break down, it would not make a society well-ordered to make it racially pure. But their goal is one that is such that were it per impossibile achieved, it would be a good thing at least in some respect—it would be good for society to be well-ordered. Similarly, the person seeking revenge probably seeks an appropriate revenge, a well-deserved revenge, or at least revenge worth having. Well, all of these are goods: were they achieved, they would be good to have, at least in some respect. However, they may, in fact, be unachievable, for it may be that revenge is never deserved, worth having or appropriate. But a good is no less good for being impossible (having perfect virtue would be good even if it were impossible).
Thus, objection (4) can be handled by the defender of (1): we always pursue goods, but some of these are impossible goods. Objection (3) is also handled. For when we pursue goods, that means that we will things under descriptions that involve a good (thick or thin). Note: It may follow from this that we believe the descriptions to be satisfiable.