Monday, March 30, 2009

Defining Reasons for Action

A member of the studio audience forwarded a question to me about desire utilitarianism.

What does "there are reasons for action that exist such that..." actually mean?

1. There are reasons why people do act such that... 2. There are reasons why people should act such that... 3. Something else. Please explain.

1 is just describing people's mental states, which makes this subjectivism. 2 makes the definition circular. Is there a 3?

Let’s look at the first option presented to us. We need to clarify (1) because there are two main types of reasons. There are what we might call blind reasons – reasons without intention. And there are goal-directed reasons or reasons that form intentions.

A blind cause is something that just happens without an objective. If a mountainside should give way, rock and dirt into a valley that blocks a river and creates a lake, the goal of gravity was not to create a lake. Gravity did not say, "I would like to have a lake here, I know, I'll bring the mountain side down and then I will have my lake." There was no intention – no purpose – behind the avalanche.

Goal directed action, on the other hand, is something where an end is identified, then the agent begins looking for means to achieve that end. The end, perhaps, is to control flooding, or to provide water for irrigation, or to generate electrical power. For each of these a means is selected of dropping a bunch of material into a river and forming a lake behind it.

"Desires are the only reasons for action that exist," points to the fact that desires are the only entities that direct action by identifying an end and searching for means to that end. These goal-directed reasons are used as the explanation for intentional action.

We use beliefs and desires all the time to explain why the people around us did what they did, and to predict what they will do in the future.

The objection is that if I select option 1 that this "makes this subjectivism". I have always stated that there is one sense of the word "subjectivism" that applies to desire utilitarianism. Desire utilitarianism holds that value depends on desire – that there is no value independent of desire. If all of the desires were to disappear, then all value will disappear, since all value exists in the form of relationships between states of affairs and desires.

However, it is still the case that desires are real, and we cannot understand (objective) reality except as a place that contains desires and relationships between desires and states of affairs. Value depends on mental states, but mental states are real. And so are values.

The type of "subjectivism" that I complain about are those that reduce moral terms to relationships between objects of evaluation and the desires of the agent as if these were the only relationships that exist. This form of subjectivism needs to be rejected because, what also exists, are reasons for action for promoting or inhibiting malleable desires, and relationships between objects of evaluation and desires that people generally have reason to promote or to inhibit.

And all of these relationships exist. We can make objectively true and false claims about all of them. Call it "subjectivism" if you will. It is "subjectivism" with right and wrong answers, a great many of which are independent of the attitudes of the speaker.


Luke said...

Yes, but then you are only describing the reasons for which people actually do things. This is a descriptive theory of intentional action. How do you get from descriptive to prescriptive?

In your book it seems you argue something like this:

1. Desires are the only reasons for action that exist - they are the only reasons that people act one way or another.

2. Value exists as a relationship between desires and states of affairs.

3. "Good" = "such as to fulfill the desires in question."

Here, we have made a sensible step into situational normativity, aka goal normativity, aka hypothetical normativity. If you want to get from L.A. to San Diego, you should travel South.

That is a sensible kind of (limited) prescriptive statement. But moral realism, it would seem, requires universally binding prescriptivity. So, to talk about moral goodness, we note that:

4. "Morally good" = "such as to fulfill the desires in question, and the desires in question are all desires"

Is that roughly correct, or have I misunderstood you?

Emu Sam said...

I suspect Alonzo will add a qualifier to your definition, such that

"morally good" = "more and stronger of all desires"

I am not aware that there is currently a weighting system for desires, but I think we agree that some desires (such as a desire to not die) are stronger than others (such as a desire to lay in a hammock for an hour reading the newspaper).

It may be necessary when answering your question to point out that the object of moral evaluation - the thing that is morally good or morally not good - will always be a desire. Acts are judged based on the desires we can reasonably believe led to them, and based on whether a person with good desires would have done it.

When judging whether a desire is morally good, we would then attempt to weigh all the desires it tends to affect, and the degree to which it affects them. We would look at a world in which no one had that desire and try to figure out if that is a better or a worse world.

Alonzo Fyfe said...


An act is right (in the moral sense) if it is the act that a person with good desires (insofar as desires are maleable) would perform.

A good desire is a desire that people generally have reason to promote - which means that it must be maleable (people have the ability to promote or inhibit it) and it tends to fulfill other desires (thus giving them reason to promote it) and not thwart other desires (which would give people reason to inhibit it).

The more and stronger of the desires a desire fulfills, the better it is. Goodness comes in degrees.

Martin Freedman said...


I prefer now the notion of good desire as that it is labelled good in virtue of most people having many and strong reason to promote this desire.

That is it does not have this label because of such reasons, such that because it is good that it should be promoted but rather the label is a referentially transparent restatement of this fact, or using "good" desire or "many and strong reasons to promote this" desire means the same in the sense of having the same truth value under a truth conditional model of meaning.

This is the link or the transformation over the descriptive and prescriptive dualism you are concerned about. This partly addresses point 3/4 in your list.

When we get to universally binding normativity it is still a hypothetical imperative, still I would restate 4 in terms that Alonzo has used recently but not IIRC in his book:

4: "morally good" = "such that most people have many and strong reasons to promote"

You don't promote what is morally good, what most people would promote if looking at all desires without bias or exceptions just is morally good.

(Of course one has to qualify this noting to exclude demographic biases of cultural or indeed whole current mix of humanity etc. otherwise this can look like moral relativism which it is not)

I am sure Alonzo can make my point more clearly.