Why Are Desires So Important?

Eenauk has presented me with another question.

Why are desires so important? On one level they are important by definition: we do, after all desire them. But on another level, there are certainly other important things in our moral lives. Kant also came up with a "closed" ethical theory that was however based on "reason alone". Why are desires more important than reason? Why reduce homo moralis to desires alone?

Before I go too far, I need to offer an apology. No, it's not what you think. See, when I write, I sometimes hear somebody else's voice in my head, and I have an annoying habit to write in that voice. My wife and I spent the weekend catching up on last year's Stargate Atlantis series and, as a result, I have Rodney McKay's voice stuck in my head. Rodney is an obnoxious know-it-all. As much as I try to get rid of it, it is stuck there, like a song one hates but cannot quit thinking about. I want to apologize if Rodney's tone leaks into this post.

The Value of Desires

Anyway, I want to devote a posting to answering this question from Eenauk because the answer is central to the moral foundation of these posts.

That answer is:

I do not need to answer why desires are important because desires are not important.

The objects of desires are important to those who have those desires. However, the desires themselves have no value.

Except . . .

Well, like I said, the objects of desires are important to those who have the desires. If the object of a desire is itself a desire, then that desire is important to those who have the desire as an object.

In addition, things can have value because of their instrumental value. A trip to the value might not be desired, but it is a useful way of avoiding things (like pain) that one desires to avoid. The trip to the dentist, in this case, has instrumental value.

Desires themselves can have instrumental value. A desire, like a trip to the dentist, can be useful in fulfilling other desires. To the degree that a desire has such a quality, then the desire is useful, like a trip to the dentist is useful. An agent can have reason to promote such a desire, just as he can have a reason to go to the dentist.

These are the two ways in which a desire can be important. The desire itself might be the object of a desire, or the desire can be a useful tool for the fulfillment of other desires. These are the only two types of value a desire can have. These are the only two types of value that anything can have.

So, why do desires play such a central role in this theory. Why can't something else play the central role - either by itself or along side desires.

The answer . . .

Because desires are the only reasons for action that exist. Desires are the only entities we have so far discovered that motivate agents to make or keep true some set of propositions.

Belief-Desire-Intention

Of course, this claim depends on the assumption that the belief-desire-intention model of intentional action actually holds up. To the degree that BDI theory breaks down - to the degree that there are reasons for action other than desires - then, to that degree, morality can be based on things other than desires. However, so long as desires are the only reasons for action that exist, then any answer to the question, "What reasons for action are there for doing X?" has to either make reference to desires, or it is false.

This is consistent with one of the things that eenauk claims; "[T]here are certainly other important things in our moral lives."

There certainly are. Every separate desire identifies something of importance. That is what desires do . . . they identify that which has importance and they motivate the agent to bring about or preserve states of affairs that realize the objects of those desires.

Beliefs do not do this. Beliefs and desires are mirror images of each other. If an agent believes that P, and P is false, then the agent needs to modify his beliefs. Beliefs - or, at least, true beliefs - map to the outside world. On the other hand, if the agent desires that P, and P is false, then the agent has a reason to act so as to create a state of affairs that realizes P.

Reason does not do this. Reason maps means to ends, but has nothing to say about the choice of ends.

Well, almost nothing.

Evaluating Ends as Means

A desire that P identifies P as an end or goal for any person who has that desire. Insofar as we are talking about P as an end, reason has nothing to say on the issue. Desires are the only reasons for action that exist, and reason cannot provide us a reason to bring about or to avoid P other than those provided by desires.

However, every desire that P, at the same time that it identifies P as an end serves as a tool for (or in opposition to) the fulfillment of other desires. So, even though reason gives us nothing to grab on to when it comes to evaluating P as an end, reason can give us something to grab on to when evaluating the desire that P as a means.

So, we can do something like what Kant argues for. Kant says that the right act is the act that one can will to be a universal law. Desire utilitarianism says that the right act is the act that would be motivated by a desire that one can consistently will to be a universal desire. However, the process of evaluating a desire has nothing to do with a Kantian ‘categorical imperative’. In order to evaluate a desire we look at simple, traditional means-ends rationality or, in Kantian terms, ‘hypothetical’ imperatives. There is no such thing as an intrinsic value, and there is no such thing as a categorical imperative.

We have no way to assign importance to desires other than insofar as they are objects of (other) desires or useful (as a means – or a tool) for the fulfillment of other desires.

How To Refute This Theory

In order to refute this – in order to put something else into the position at the root of morality, they will need to come up with a theory of action. That theory of action needs to demonstrate that it can predict and explain intentional action better (more accurately, more simply) than the belief-desire theory that I have relied on in this blog. Once somebody has demonstrated that reasons for action other than desires exist, then one can begin to sensibly refer to those reasons for action to answer questions about what we have reasons for action for doing or for refraining from doing.

I am not even going to say that it can’t be done. In fact, I will assert the opposite. I will confidently predict that, someday in the future (if the human race lives that long) scientists will replace belief-desire theory with something better at explaining and predicting human actions. Science will make new discoveries. The science of the mind is in its infancy, with some large-scale changes in our understanding to be expected. So, I am simply not going to say that no better theory can be invented. I will only say that I am dealing with (I hope) the best we have available today.

When a new theory of the mind does come forward, those people will need to look at the reasons for action in that theory. When they ask the question, “What reasons for action exist for supporting or opposing some policy P?” the answer – at least among all rational people – will be to turn to the reasons for action that exist. Reasons for action that do not exist simply are not relevant.

Summary

So, nothing is important in the real world except insofar as it is the object of a desire, or it is useful for bringing about the realization of a state that is the object of a desire. Desires themselves are not important in the real world except insofar as they are the objects of desires or useful for bringing about the realization of other desires. The reason why everything gets evaluated in relation to desire is because desires are the only reasons for action that exist. Desires are the only real-world entities that identify possible ends and motivate the agent to realize those ends. In order to turn to something else in answering questions about how to act, we have to show that this ‘something else’ is a reason for action that exists in the real world.