On Desire 2018. Part 41: Susceptibility to Evidence

Concerning Alex Gregory's defense of his thesis that "To desire to Φ is to believe that you have normative reason to Φ," Gregory's second argument in defense of this thesis is as follows:

Second, DAB explains why desires are sensitive to evidence about what we have reason to do (cf. Fernández 2007; Byrne 2011; Moran 2001: 119). If you want to vote Conservative, I might get you to rationally abandon this desire by presenting you with evidence that there are no good reasons to vote Conservative. Or, for another example, if I ask you whether you want my spare plane ticket to China, you will respond by considering the reasons for and against taking this choice: the sights, the food, the weather, etc. DAB explains why desires are sensitive to evidence about reasons: because they are beliefs about reasons.

(Gregory, Alex, (2017), “Might Desires Be Beliefs about Normative Reasons for Action?” In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press.)

I discussed his first defense in the previous post. There, he wrote that normative beliefs have the power to motivate us to act, which I countered by suggesting that normative beliefs are beliefs about relationships between states of affairs and desires where desires provide the motivation to act.

The problem in this second defense is that, as a matter of fact, the desires-as-ends (that which we desire for their own sake) are immune to evidence. More complex forms of desire are subject to evidence because they are mixtures of beliefs and desires, and beliefs are sensitive to evidence.

As evidence that basic desires are immune to evidence, try reasoning a person out of extreme pain, out of their love for their child, out of an addiction, or out of their sexual orientation. What syllogism can you provide a person who prefers the taste of butterscotch topping to the taste of chocolate that will alter his preference so that he comes to refer the taste of chocolate? How do you persuade a child that she likes broccoli when they do not like broccoli?

Indeed, if basic desires are susceptible to reason, I would like to see the structure of the argument that has a change of taste or preference as its conclusion. Generally, if a person claims to have a belief that P, we can at least make some sense of what it takes to prove that "P" is true. So, if P is, "I have a normative reason to Φ," please provide an example of how such an argument can be constructed without making a viciously circular reference back to the agent's own desires.

This is not to say that desires are not subject to change. Of course desires can change. However, they do not change as a consequence of evidence. The most common way for a desire to change is by experience. If a particular action produces a reward - whether in the form of pleasure, or a drug-induced high, or praise from others; or if it produces pain, the agent at first may value (or disvalue) the action as a way of obtaining pleasure or avoiding pain. However, it comes to be valued for its own sake - one of the rules of behavior that the agent adopts. But the relationship between experience and desire is nothing like that of evidence and belief. There is no "valid inference" between experience and desire as there is between evidence and belief. There is merely cause and effect.

When Gregory claims to see desires susceptible to evidence, what he actually sees are cases of complex mixtures of beliefs and desires being subject to evidence. However, this is because beliefs are a part of the complex structure.

For example, we should consider the distinction between means and ends. A person "wants" a half-inch wrench to tighten a bolt, then discovers that it is a 9/16 inch bolt. He no longer "wants" the wrench. It seems that his desire is “sensitive to evidence.” However, this is only because we are using a shorthand way of speaking. The agent never did want a half-inch wrench. Instead, he wanted to tighten the bolt and he believed that, with a half-inch wrench, he could tighten the bolt. His belief is sensitive to evidence. But his desire to tighten the bolt is not. The belief that the bolt was a half-inch bolt provided none of the motivating force. That came from the desire to tighten the bolt. The desire is not susceptible to reason (unless, of course, tightening the bolt is a means to some further end, and there is evidence that tightening the bolt will not, in fact, serve that end).

For another example, consider the person who wants to own a Picasso painting. He believes that the painting up for auction is a Picasso, so he "wants" that painting. Then he is given reason to believe that the painting is a reproduction and not an original Picasso. This, too, looks like evidence that his desire is susceptible to evidence. Yet, evidence did not alter his desire to own a Picasso painting. Rather, the evidence altered his belief that the painting up for action is a Picasso. There is no mystery behind the fact that beliefs are susceptible to evidence. But the belief that the the painting was a Picasso did not provide any motivating force. That required the desire to own the Picasso.

The examples that Gregory gives us of desires being sensitive to evidence - the case of "wanting to voting Conservative" or "wanting a ticket to China," these "wantings" are packed with beliefs sensitive to evidence. They are like "wanting" the half-inch wrench, or the painting that is up for auction. There is no mystery behind the fact that evidence regarding the relationships between voting Conservative or travelling to China and the agent's desires might alter the agent's opinions about those relationships.

Indeed, rather than sensitivity to evidence providing a reason to look at "desire as belief" theories, the invulnerability of ends to reason suggests an argument against "desire as belief" theories.