Having a Reason vs There Exists a Reason

I have often protested the fact that those who write on "reasons" fail to distinguish between "the reasons that a person has" and "reasons that exist."

Elsewhere, we make this distinction quite easily. We distinguish, for example, between the money that a person has versus the money that exists, and the job that a person has and the jobs that exist. However, in the realm of reasons, we see no such distinction.

Look at the way that the Stanford Encyclopedia of Philosophy article on Internal and External Reasons describes the Humean Theory of Reasons.

The Humean Theory of Reasons (HTR): If there is a reason for someone to do something, then she must have some desire that would be served by her doing it.

"If there is . . . " and "She must have . . . " - as if one can switch between one and the other without effort.

After all of that complaining, I would like to take a look at what reasons language would look like if it respected this distinction. We could, for example, divide the Humean Theory of Reasons into two parts; a theory of having a reason versus a theory of reasons that exist.

The Humean Theory of Reasons That an Agent Has (HTRTAH): If an agent has a reason to do something, then she must have some desire that would be served by her doing it.

The Humean Theory of Reasons That Exist (HTRTE): If there is a reason for someone to do something, then there must be some desire that would be served by her doing it.

Of course, the reasons that an agent has is a subset of reasons that exist, just as the money that an agent has is a subset of the money that exists.

This may look like it is begging the question with respect to one of the issues in the topic of reasons - the possibility of external reasons. If there are more reasons that exist then there are reasons that an agent has, it seems to follow that external reasons must exist.

However, that moves too quickly. Making this distinction would hardly count as proof that reasons internalizes such as Bernard Williams are wrong. Instead, Williams would be justified in waving his hands and saying, "Of course those types of external reasons exist, but they are not the type of external reasons I am objecting to."

Williams came up with the following theory of reasons (from Williams, B., 1979. “Internal and External Reasons,” reprinted in Moral Luck, Cambridge: Cambridge University Press, 1981, 101–13.):

(1) A has a reason to φ iff A has some desire the satisfaction of which will be served by his φ-ing.

This does not confuse having a reason and there exists a reason. Yet, Williams would have to agree that, if this is true, then the following is also true:

(2) B has a reason to φ iff B has some desire the satisfaction of which will be served by his φ-ing.

And, if B has a reason to φ, then it must be the case that there exists a reason for B to φ. And if there exists a reason for B to φ, then there must be a distinction between the reasons that exist and the reasons that A has.

It will also be the case, in some circumstances, that B might have a reason for A to φ – meaning, B has a reason to bring it about that A does φ. Here, following Williams’ original formula:

(3) B has a reason for A to φ iff B has some desire the satisfaction of which will be served by his bringing it about that A does φ.

All of these follow from Williams’ original thesis. When Williams says that there are no external reasons, we would have to consider him guilty of the greatest epistemic negligence if we were to interpret him as implying that (2) and (3) were not the case.

The following objection can still be raised: B’s reasons to φ may well be a subset of the reasons that exist. However, it is still the case that the “reasons that exist for B to φ” (if we take this whole phrase rather than its parts) is still limited to the reasons that B has to φ. Other reasons clearly exist – the reasons that other people have. However, none of those other reasons are “reasons for B to φ”.

However, I find the phrase, “reasons for B to φ” to actually be ambiguious on this matter. In one sense, it can be understood to refer only to the reasons that B has. However, in another sense, I would argue that it is sometimes used to mean “reasons that exist for B to φ”.

In fact, I would draw the distinction between the two senses of “reasons for B to φ” precisely on the line separating practical reasons from moral reasons.

If we are talking about practical “ought” – about what a person should do in the practical sense – then we are talking about the reasons that an agent has. In this context, the “reasons for B to φ” and the “reasons that B has to φ” are the same.

If, instead, we are talking about moral “ought” – about what a person should do in the moral sense – then we are talking about the reasons that exist. Harm to others is a morally relevant reason because harm to others refers to states that others have many and strong reasons to avoid. Consequently, in a moral discussion, talk of “reasons for B to φ” is talk of reasons that exist for B to φ – which includes more than simply the reasons that B has.

This is just another place where the ambiguity between the reasons that exist and the reasons that an agent has manifests itself – where people assume that words are being used consistently when, in fact, they are not.

Regardless of what we do with the language – regardless of how we decide to interpret the phrase, “reasons for B to φ”, the reasons that exist for B to φ, as distinct from the reasons that B has to φ, is a real thing. It is a part of the world. It is something for us to discuss. It is something that we have reason to talk about regardless of whether or not we currently have the terms to talk about them.

However, I do not think that native English speakers have left this whole realm of discussion – the discussion of reasons that exist – out of our vocabulary – leaving us with only those terms that refer to the reasons that an agent has. If we were to look more closely at the distinction between practical and moral ought, we would find in this a distinction between actions based on the reasons that an agent has, and actions based on reasons that exist.