Monday, September 12, 2011

A Response to Richard Carrier

(I had intended to post this as a comment in this thread, but it's either too long or uses characters picky-picky Blogger doesn't like. Please see the original post and comments for context.)

Dr. Carrier,

Thank you for your response!

It seems there are two basic issues under contention here:

1. What counts as morality.
2. The semantics of 'ought.'

We can set 'ought' aside for a moment. Even without ever mentioning that word, you could claim that morality is identified with — not merely associated with or marked by — what a person has most reason to do by virtue of her desires (or a certain class of her desires). I concede that once this identity is in place, your basic argument follows least on the individual level.

Why aren't metaethicists flocking to this identity? I suggest it's the same reason metaethicists aren't flocking to Sam Harris' externalist identity between morality and what promotes the flourishing of conscious creatures. Namely: many philosophers and non-philosophers think morality has a different identity, which only partially or contingently overlaps with your answer or Harris' answer. In the May 2011 Craig vs. Harris debate, Craig argued that Harris "isn't really talking about moral values at all. He's just talking about [the identity Harris favors]." Then, in the same debate, Craig implies that morality is identified with a humanity-transcending legal authority; if a proposed description of morality isn't about that, it isn't really morality!1

So whose identity is correct? Yours, Harris', Craig's, etc?

It begs the question to insist that other answers are incorrect because they don't line up with a person has most reason to do by virtue of her desires. Harris could say your answer is incorrect insofar as it doesn't line up with the well-being of conscious creatures. Craig could say your answer is incorrect insofar as it doesn't line up with God's transcendent laws.

Now we can get back to 'ought.' Your main answer to why your identity should be recognized over the alternatives appears to be that unqualified 'most ought' refers to your answer, and people are likely to admit that morality is identified with what we 'most ought' to do. This way, two paths are meant to lead back to your one answer:

Path A. Agreeing that morality = what a person has most reason to do.
Path B. Agreeing that the moral choice = what a person most ought (or actually ought) to do.

I'm trying to block Path B by denying that there is a single meaning to what a person most ought to do.
"Michael ought[some end] to contribute to UNICEF" can only be true if it converts to the conditional, "If Michael wants [some end], then P" where P is your proposition.
Except we use 'ought' in situations where it doesn't seem plausible we're talking about what the agent wants. Suppose Supervillain McGee empties Fort Knox, launders the gold bars, and successfully hides away in Argentina with his wealth. I expect a lot of people would still say, "Whoever stole those gold bars ought to bring them back" and mean it in a moral sense. But neither McGee's superficial nor his deep desires would be best satisfied by doing that, because he's so good at getting away with crime. Would these people take back their moral 'ought' claims if they knew about McGee's psychology and success? I don't think so!

Nor is it pointless to morally condemn agents who don't give a damn. Others who hear the condemnation might give a damn and conform — at least somewhat — to your negative attitude toward theft.
That's why there can never be a single true proposition of the form "Michael ought[1] to X" for any X, because we can invent countless "some ends" and thus countless "Michael oughts[1]" that are all (by your scheme) equally true, which is impossible (because they all contradict each other).
I agree that there can never be a single true proposition involving 'ought' without an end, just as there can never be a single true velocity without a reference frame. The utterance, "Michael ought to X" could be interpreted as an endless array of true and false propositions, depending on the end. In theory anyway. In practice, we tend to use a fairly small set of ends which can often be successfully inferred from the context of the utterance.
That's why your ought[1] cannot produce true propositions.
Just to be clear:

utterance: "Michael ought to go boating tomorrow."

propositional schematic: In order that [some end], it ought to be the case that Michael goes boating tomorrow.

a complete proposition: In order that Michael keeps his promise to his children, it ought to be the case that Michael goes boating tomorrow.

another complete proposition: In order that Michael appease Dagon, it ought to be the case that Michael goes boating tomorrow.

a third complete proposition: In order that Michael best fulfill his desires, it ought to be the case that Michael goes boating tomorrow.

Now, supposing Michael promised his children a day at the zoo, the first proposition may be false while the third is true. But since they don't logically conflict, this is ok. And of course I put the third one there to show that your supposed base case can easily be seen as a special case.
BTW, in regard to my use of the terms "objective" and "realism" (which are in accord with standard references in the field), see my discussion of the ontology of my moral theory on my blog: Moral Ontology.
Unfortunately, a number of quite distinct uses of those terms are consistent with professional usage. You (and I) could count as moral realists, but on the "low bar" side of the spectrum. Similar deal for 'objective.' These terms tend to get used as badges rather than useful classifications.


1 comment:

  1. It looks like you missed the layer underneath goals and desires that Carrier was talking about when he mentioned people being motivated or not.

    To help spell it out, I'll look at your last example with boating and Dagon and all of that. Here's the one proposition you are missing (and it could be reworded in terms of views that are pre-scientific, I suppose):

    In order that the laws of physics be obeyed, Michael's brain does what brains do.

    This will contradict things that can't physically motivate his brain to act in other ways. Thus the options that are not motivating are literally false, being physically impossible. I hope you can see what I mean.

    So, the next step is "In order that Michael's brain does what brains do, it ought to be the case that Michael______"

    It then simply becomes an emperical question about what brains do (Carrier and myself hypothesise it is about seeking the most satisfaction, so that all true "oughts" about our choices begine with "In order that this person best fulfill their desires") and that last blank gets filled in accordingly.

    Also, since brains are imperfect, we can only say so to a degree of probability. But Carrier is even talking about an idealized case by saying "if we are fully informed of the facts and are resoning correctly".