Two Types of Subjectivism

Tomorrow, I am going to attend a PhD dissertation defense by a University of Colorado graduate student, Jonathan Spelman.

When I learned that Mr. Spelman was delivering a presentation to the summer students at the university on "In Defense of Subjectivism about Moral Obligation" I wrote to him to ask if he had a paper on the topic I could review. He wrote back to say that it was his dissertation which he would be defending the following week. I got a copy of the dissertation - and read it.

My interpretation of Spelman's dissertation was that he gave a good defense of the claim that the facts relevant to a moral evaluation include facts about the beliefs of the agent.

Let's look at it this way:

A moral syllogism contains three parts.

(1) The prescriptive premise: A general principle governing what the ought to do.
(2) The descriptive premise: An account of the facts of the given situation
(3) The conclusion: A statement of what the agent ought to do in that situation.

An example of a moral syllogism - which comes from Spelman's paper - concerns the case of a doctor named Jill. Jill's patient Frank has a non-fatal skin condition. Jill knows that drug A will relieve the symptoms but not cure the disease. Of drugs B and C, Jill knows that one will cure the disease while the other will kill the patient. However, she does not know which is which. From this, we conclude that Jill ought to give Frank drug A. There is no sense in risking Frank's life to cure a disease where there is an effective treatment for its symptoms.

Spelman uses this as an argument for moral subjectivism over moral objectivism.

Let us assume that drug B will actually cure Frank's disease. However, Jill does not know this. There is a sense in which Jill should give Frank drug B in that this would produce the best outcome. However, given Jill's ignorance, this is not what she should do in the given situation. In the given situation - which includes the facts of Jill's ignorance - Jill should give Frank drug A as stated.

For Spelman, this supports the conclusion that beliefs are relevant to the question of what Jill should do. This is a form of subjectivism - a form of the view that something is right in virtue of the agent's beliefs, rather than in virtue of the facts of the matter.

Well, I write back with a comment.

One of those comments is that Jill's beliefs ARE facts of the matter. This is in keeping with my own position that "subjective" and "objective" are not mutually exclusive options. We have objective facts of the matter regarding mental states - such as the facts that describe what Jill believes regarding the effects of drugs A, B, and C.

I have an objection to Spelman's paper in that he does not seem to distinguish between two separate propositions.

Proposition 1: The truth of the conclusion of a moral syllogism depends on beliefs.

Proposition 2: The truth of the prescriptive premise of a moral syllogism depends on beliefs.

As I mentioned, Spelman produces several arguments that can be understood as showing that the truth of a moral conclusion depends on beliefs. In the example above, what Jill ought to do with respect to her patient depends crucially on her beliefs regarding the effects of drugs A, B, and C. Her ignorance over which of the drugs B or C will cure the patient and which will kill him are morally relevant. The conclusion that Jill should give the patient drug A depends on the fact of Jill's limited knowledge regarding the effects of drugs B and C.

However, one cannot infer from the fact that the truth of the conclusion depends on Jill's beliefs that the truth of the prescriptive premise depends on Jill's beliefs - specifically, in her belief in that premise. The prescriptive premise states that Jill ought to do that which - given her beliefs - would be best for her patient. This prescriptive premise is true independent of whether or not Jill accepts it. This identifies an actual and objective moral obligation.

Just to drag desirism into this conversation, this obligation exists in virtue of the fact that people generally have many and strong reasons to promote this sentiment using the social tools of praise and condemnation. But that does not impact the actual argument we are discussing here. There might be some other way to support the proposition that this objective moral obligation exists. What matters is that one cannot infer the proposition that the truth of the prescriptive premise depends on the agent's opinion from the fact that the truth of the moral conclusion depends on the agent's opinion. This is no more valid than inferring the fact that the truth of the descriptive premise depends on the agent's opinion from the fact that the truth of the conclusion depends on the agent's position.

The truth of the conclusion depends on the agent's opinion because the (objectively true) prescriptive premise says that it matters - and that provides the reason for including (objectively true) claims about the agent's beliefs in the descriptive premise.

This yields an objectively true conclusion that depends crucially on the agent's beliefs about the world (such as Jill's partial ignorance regarding the drugs B and C) in premise 2.

Well, I have exchanged some emails with Mr. Spelman - soon to be Dr. Spelman. He does seem to want to infer from the evidence he provides that the truth of the conclusion depends on the agent's beliefs that the truth of the prescriptive premise depends on beliefs. I think that this is a mistake. At the same time, I have to say that he has done a very good job proving that the truth of the conclusion depends on the agent's beliefs. Even if he draws an invalid implication from this fact, he has demonstrated what, within desirism, would be considered a very important fact.