Michael Smith Presents John Mackie's Error Theory

I am going to be going through Michael Smith's article, "Beyond the Error Theory," where Smith examined four responses to J.L. Mackie's famous argument against intrinsic prescriptivity. (Smith, Michael, "Beyond the Error Theory," in A World Without Values: Essays on John Mackie's Moral Error Theory, (Richard Joyce and Simon Kirchin, eds.), 2010.)

Before looking at the responses, Smith gives us his interpretation of Mackie's main argument - the one that the responses are responding to. It is fitting that we look at the argument before looking at the responses.

Mackie's argument has two parts. The first part is a claim that intrinsic prescriptivity is built into the very meaning of moral terms such that all moral claims are understood to be claims about intrinsic prescriptivity. The second part is that intrinsic prescriptivity does not exist. This leads to the conclusion that all moral claims are false.

Of this argument, the part that I want to explain a bit more clearly is the claim of what this "intrinsic prescriptivity" is. What is it that Mackie is saying does not exist?

Mackie presents a number of different ways of understanding his claim about intrinsic prescriptivity. Smith simplifies this by looking at one in particular - that of the early 20th century moral philosopher Henry Sidgwick.

In Sidgwick's terms, Mackie's claim that our concept of moral value is the concept of an objectively prescriptive feature of things amounts to the claim that to conceive of (say) happiness as a moral value is to conceive of happiness itself as having the feature of being an end that is absolutely prescribed by reason.

From my own perspective, Mackie is correct. There is no such thing. There is no rationality of ends, there is only a rationality of means. We can only speak about a rationality of ends in the sense that an end is also, at the same time, a means to the fulfillment or thwarting of other ends, allowing us to rationally determine the means-potential of an end. (I am not so sure about the claim that intrinsic prescriptivity is built into the meaning of moral terms - nor of the consequence that all moral claims are false. However, on this part, I hold that there is no such thing as intrinsic prescriptivity.) See my previous post: 

I draw a close distinction between ends and desires. Specifically, I take the view that desires are propositional attitudes - a desire can be expressed in the form of "desire that P" where P is a proposition. A desire that P motivates the agent to realize P - to create a state of affairs where P is made or kept true. The "end" is that state of affairs where P is made or kept true.

So, a person can have a desire to be happy. A person with that desire has a reason to pursue happiness as an end. However, happiness in this sense is an end that is prescribed by the desire to be happy. It is not an end prescribed by reason.

Michael Smith put it in the following way:

Ends are the sorts of things that each of us has, insofaras we aim at, or desire, different things.

Two facts that are relevant here are:

(1) A state of affairs in which P is true is an end in virtue of the agent having the desire. If the agent's desires change, then the agent's ends change. This way of looking at ends does not have room for any end being recommended by reason alone.

(2) Nearly anything can be an end. In this article, Smith is looking at happiness being an end. Yes, a person can have a "desire that P" where "P" = "I am happy", in which case the agent has an end of being happy. But the agent can also have a desire that his children are healthy, that he is eating a steak dinner prepared in his favorite way, or that he is not in front of a large crowd giving a speech. With these desires, the agent acquires other ends, and happiness becomes one end among many.

However, the claim that happiness is intrinsically prescriptive says that the value of happiness does not require a desire that "I am happy", that it is an end "absolutely prescribed by reason" independent of the agent's desires.

But while it is . . . true that some of us have happiness as an end, as some of us do desire happiness, the claim that happiness has a feature of being an end absolutely prescribed by reason, "the same for all minds," would have to be made true by some further fact about happiness, a fact beyond this purely descriptive psychological fact.

Here, we get to Mackie's claim that these prescriptive properties that reason tells us about that an object of evaluation may have independent of desire is such a queer thing that we have reason to doubt its existence. And even if they did exist, we would have no way to know about them.

Admittedly, there really is not much to this argument. I see it more as a challenge to those who wish to claim that intrinsic prescriptivity is real to tell the rest of us what, exactly, they are talking about. However, this challenge is backed up by the assertion, "And as you try to tell us what this intrinsic prescriptivity is, you will come to discover that it is such a queer thing - such a bizarre thing - that there is no sensible way in which such a thing can be a part of the real world."

Thus, again, all moral claims are false.

I want to add again, because it is glossed over so often, that this is a problem for objectivity in the form of intrinsic prescriptivity. It is not a problem for objectivity in the form of objectively true statements about the relationships between objects of evaluation and desires. If the argument stands, it tells us that we have to throw out the former type of objectivity, not the latter.

Well, Mackie tells us that intrinsic prescriptivity does not exist,.that all moral claims are claims about intrinsic prescriptivity, and that all such claims are false. Smith is going to examine four responses to Mackie. We will take a look with him in future posts.