What Should I Do?

Why should I do what you tell me I should do?

Give me a reason.

What is a "reason" anyway? And how can we tell the difference between a good reason and a bad reason?

These questions are going to take a few posts. However, I will start here with the question, "What is a 'reason'?"

The Stanford Encyclopedia of Philosophy (SEP) has an article on this subject: Reasons for Action: Internal vs. External. However, it contains a serious error - or at least a serious source of confusion that leads to a lot of error.

I know . . . for me to say that the SEP contains a serious error is a bold claim. However, I will back it up below and you can decide for yourself.

This entry in the SEP entry presents what it calls, “the revised Humean theory of reasons” (RHTR). The term “Humean” refers to the 18th century Scottish philosopher David Hume (1711-1776). If anybody wants to know which famous historic philosopher comes closest to the desirism, that would be David Hume.

The SEP presents the RHTR as follows.

Revised Human Theory of Reasons (RHTR): If there is a reason for someone to do something, then she must have some desire that would be served by her doing it, which is the source of her reason.

In other words, it says that if there is a reason for you to read this blog posting, then you must have a desire that would be served by reading this blog posting. I am going to reject this claim in what follows. But, to understand its initial plausibility, we can imagine that you want to know something about the nature of what it is to have a reason to do something. In that case, reading this posting will (hopefully) serve that desire, so you have a reason read this posting. What the SEP calls "serving" a desire I call "fulfilling" a desire. That is to say, if we express a desire as a "desire that P" and then fulfilling a desire is acting to make or keep "P" true.

It is obvious what this has to do with desirism. It reduces reasons to desires. This entry appears in the SEP because a lot of philosophers think that anybody who wants to hold a theory that reduces everything to desires has to believe something like this RHTR. Unfortunately, RHTR as stated is problematic. Consequently, many philosophers conclude that a desire-based theory of morality like the one you see here is problematic.

But . . . RHTR is . . . well . . . not the way I would put it.

The problem is that it relates "there is a reason" (seen at the start of the description) to "she must have some desire" found in the second half of this association. There is no justification for this. In fact, it is clearly false - and can be shown to be false with some very obvious counter-examples.

The whole problem rests on the distinction between "There is an X" and "Agent1 has an X". Clearly, it is possible for it to be the case that "There is an X" to be true and "Agent1 has an X" to be false. Think about a $5 bill. It is possible for it to be true that "There is a $5 bill on the table" to be true and "Agent1 has a $5 bill on the table" to be false. This happens whenever it is the case that the $5 bill belongs to somebody else, or belongs to nobody at all.

The same is true with reasons and desires. The first part of the statement says that "there is a reason". It is perhaps clearer to understand this as "there exists a reason" or "a reason exists." Let's grant that. Assume that Agent2 has a desire to live in an orbiting space station. This means that Agent2 has a reason to live in an orbiting space station. Because Agent2 has a reason to live in an orbiting space station, there exists a reason for Agent2 to live in an orbiting space station. Because there exists a reason for Agent2 to live in an orbiting space station, there exists a reason for Agent1 to help Agent2 live in an orbiting space station.

Whoa! you say. That last sentence doesn't follow form the earlier statements. For the moment, I am simply going to stipulate that it does. But it clearly seems to have a second meaning as well - a meaning that is a source of all of the confusion. I will get to that second meaning shortly. At this point, all of the statements in the previous paragraph, as I am using the terms, refer to the same desire - Agent2's desire to live in an orbiting habitat.

This is the statement that I want to disambiguate. I want to separate the sense that is consistent with the rest of the paragraph from the sense that is not and end the equivocation between these two senses that is causing so many problems. The sense that is consistent with the rest of the paragraph does not imply that Agent1 has a reason to help Agent2 live in an orbiting space station. In fact, in this example, Agent1 does not have a reason. "There is a reason" is true, but "Agent1 has a reason" is false. Thus, the RHTR is false in this sense. We have to reject it.

On the other meaning, RHTR is true, but it is trivially true. On this second meaning, "there is a reason for Agent1 to do something" is just another way of saying, "Agent1 has a reason to do something." If Agent1 has a reason to do something" is true in this sense, then "Agent1 has a desire that would be served by doing it" is also true. But this sense makes the false assumption that "there is a reason" implies "Agent1 has a reason" - and this is true only if all of the reasons that exist are those that Agent1 has.

Here, I would advise the critic to adopt a clearer, more precise language. When "there is a reason for Agent1 to do something" means "Agent1 has a reason", then say so, and quit using the phrase "there is a reason".

In my next post, I will be showing that what the SEP calls "the central problem" in moral philosophy rests on a failure to be clear on this matter, allowing people to equivocate between two different meanings of "there is a reason for Agent1 to do something" - the "there is a desire that would be served" meaning and "Agent1 has a desire that would be served" meaning.

And that is why contemporary moral philosophy is all messed up.

More specifically, I will be using the Revised Humean Theory of Reasons Further Revised:

The Revised Humean Theory of Reasons Further Revised (RHTRFR): If there is a reason for someone to do something, then there must be some desire that would be served by doing it, which is the source of that reason. And if Agent1 has a reason to do something, then Agent1 must have some desire that would be served by doing it, which is the source of that reason.

This is the theory that Desirism uses.

Note that this version relates "there is a reason" to "there is a desire that would be served" (without specifying whose reason/desire it is) and it relates "Agent1 has a reason" to "Agent1 has a desire that would be served."

This is going to clear up a lot of problems.