The following simple and valid argument came out of discussions with Mark Murphy (who has a forthcoming book that contains related arguments, though perhaps not this one).
According to the identity version of Divine Command Metaethics (IDCM), to be obligated to A is to be commanded to A by God (or to be willed to A by God or to be commanded to A by a loving God--details of this sort won't matter). But:
- If p explains x's being F, and to be F is the same as to be G, then p explains x's being G.
- My being commanded by God to follow Christ explains my being obligated to follow Christ.
- It is not the case that my being commanded by God to follow Christ explains my being commanded by God to follow Christ.
- Therefore, it is false that to be obligated to A is the same as to be commanded by God to A. (By 1-3)
The argument more generally shows that no normative-level answer to a "Why am I obligated to A?" question can provide a property identical with being obligated. Thus, sometimes at least the answer to "Why am I obligated to A?" is that Aing maximizes utility. Hence, by an exactly parallel argument, being obligated to A is not the same as having A as one's utility maximizing option.
The argument is compatible with constitution versions of DCM on which the property of being obligated to A is constituted by the property of being commanded to A. But such theorists then have the added complication of explaining what the constitution relation means here, over and beyond bidirectional entailment (after all, many non-divine-command theorists will agree that necessarily x is obligated to A iff God wills x to A).
Greetings,
I apologize if I've made a mistake here, as I am a layperson, but it seems to me like we are to infer that IDCT fails because the third premise is circular. Is that correct? If so, the argument is still sound, isn't it? After all-- if x, then x is necessarily valid in classical logic.
Is the circularity of 3 sufficient to make the argument false? Maybe it is, or maybe I missed the point. Either way, I appreciate your clarification.
Thanks!
The argument (1)-(4) is against IDCM. I am defending the premises, including (3), as true. The reason I think (3) is true is that although it logically follows from "God commanded me to follow Christ" that "God commanded me to follow Christ", the claim that God so commanded me is not self-explanatory. One way to see that it's not self-explanatory is that explanations can be given for it--one might give the reasons for which God commanded me this, or one might explain what it is to command.
Can't the defender of IDCM just deny (2)? We might think that the one explained the other, but that was before we realized they were identical.
Analogy: the ordinary chap says that his hunger for pies explains his excessive consumption of them. But the (a posteriori) behaviorist says not so: your excessive consumption just is your hunger.
Are you saying that (2) carries independent intuitive weight?
Matt:
The story about pies seems a good reason to deny behaviorism.
As for (2), I think there are three lines of thought that lead to it:
i. It is intuitively plausible, and were divine commands not explanatory of some obligations, divine command theory wouldn't be plausible even at the normative level.
ii. Some obligations are explained by things other than divine commands. Thus, my obligation to A might be explained by my having promised to A. Consider this little question and answer session:
Q. Why was Laban obligated to allow Jacob to marry Rachel?
A. Because he promised to do this if Jacob worked for him for seven years, and Jacob did work for him for seven years.
The divine command theorist should not disagree. Of course, she'll want to add that the moral force of promises comes from a divine command to keep promises, but the explanation given here is just fine.
Suppose that after asking the question about Laban, the questioner asks:
Q. And why were the Israelites obligated to refrain from eating pork?
A. Because God commanded them.
And so there are cases of actions that a person is obligated to perform because she has been commanded to. And for my argument I need only one case--(2) was just an example.
iii. The following argument appears sound:
2a. If a legitimate authority x issues to y a valid command to A, then y is obligated to A because x was commanded by y to A.
2b. God is a legitimate authority over me.
2c. God has validly commanded me to follow Christ.
2d. Therefore, I am obligated to follow Christ because I was commanded by God to follow Christ.
--------------------
There is also another version of my main argument.
5. One is never obligated by being obligated. (One may be obligated by having promise, or being commanded, but not by being obligated alone.)
6. I am obligated (say, to follow Christ) by being commanded by God.
7. Therefore, being obligated is not the same as being commanded by God.
Perhaps a defender of the IDCM theory might reply as follows: Your argument confuses the property of being X with the concept or meaning of being X. For example, the property of being composed of H2O is identical to the property of being water. But the concept of being composed of H2O is not identical to the concept of being water.
Why is the stuff in Lake Michigan a clear, odorless, and tasteless liquid that we can drink and that quenches thirst, etc? Because the stuff in Lake Michigan is composed of H2O (and is subject to certain physical laws). 'X falls under concept A' can explain why 'X falls under concept B', even if (perhaps unbeknownst to us) A and B pick out the same property.
Aaron:
Again, my argument doesn't use the concepts "concept" or "property". It doesn't even seem to contain any referring expression that refers to a concept or a property. It talks of "to be F", but I don't know that "to be F" is a referring expression (it's an infinitive of a predicate). (I also use expressions like "explains x's being F", which really should be "explains that x is F".)
I take it the suggestion is that I use "to be F" types of expressions equivocally. Thus, maybe, in (1) "to be F is the same as to be G" is a statement of concept identity while in (4) we have a denial of a claim of property identity (the other way around wouldn't be plausible), and concepts cut more finely than properties do.
I am not sure this move works, though. Suppose we read (1) as talking of property identity. That sure seems plausible. If P=Q, then to explain why x has P is to explain why x has Q (I know that Dan Johnson disagrees with me here--he did so in a comment on my blog's version of the argument--but I think the point holds for the sort of explanation I am talking about). Or at least this is plausible in the special case in which I need the principle, where "P" is an intrinsic name for a property ("obligation") and "Q" is a name based on the proposed account ("being commanded by God").
Let's take the water case.
Q. Why is there water on the floor of the garage?
A. Because there is H2O on the floor of the garage.
That's not a good explanation. How about:
Q. Why is there a clear, odorless, and tasteless liquid that we can drink and that quenches thirst, etc. on the floor of the garage?
A. Because there is (liquid) H2O on the floor of the garage.
(Nevermind the oddness of saying you can drink it. And if it's on the floor of the garage it's probably not clear. That's just detail.)
This is a good explanation. But while it is true that to be water is to be H2O (or liquid H2O?), it is false that the property of being a clear, odorless, and tasteless liquid that we can drink and that quenches thirst, etc. is the same as the property of being H2O. Here we are dealing with plainly different properties--if the former is a property at all (I like properties to be sparse). I don't even know if they have the same extension in our world, much less in all worlds, but even if we find a way of filling it out so we get the same extension in all worlds, the one is a conjunction of phenomenological properties while the other is not.
Of course, you could go with Lewis's unstructured proposition theory and identify together all necessarily extensionally equivalent properties. Then, in effect, IDCM would just be claiming that necessarily A is obligatory iff A is commanded by God.
"If p explains x's being F, and to be F is the same as to be G, then p explains x's being G."
Let p = Lois Lane dates superheroes (or, perhaps better, Lois Lane loves Superman).
p explains Lois Lane's being on a date with Superman.
To be on a date with Superman just is to be on a date with Clark Kent.
But p does *not* explain Lois Lane's being on a date with Clark Kent.
Soooo.... premise 1 is false. What do you think?
I think the claim that p does not explain LL's being on a date with CK is only plausible if to be CK is a different thing than to be Superman. If they are the same, then I think we should say that p does explain LL's being on a date with CK, but does not do so as perspicuously as it explains being on a date with Superman. But if they are different, then by the same token to be on a date with Superman and to be on a date with CK are different properties.
Well, that's certainly one way you could go. It doesn't look very plausible to me at the moment, but I'm open to changing my mind! :-)
So you think that p (Lois Lane loves Superman) explains why Lois Lane is on a date with Clark Kent. That strikes me as pretty implausible. I would have thought that p counts as an explanation of that fact only in conjunction with some other proposition q (namely, that Superman is Clark Kent). By itself, p doesn't do the trick.
After all, imagine someone who is in the dark about Superman's being Clark Kent, but who works with Clark at the Daily Planet and observes that Lois is on a date with Clark. This person wonders "Why is Lois on a date with Clark? That guy is such a nerd." Merely telling this person that Lois Lane loves Superman is not a sufficient explanation. It's not an explanation at all. This person could, in full rationality, know that LL loves SM and still wonder why LL is on a date with CK. I take it that a sufficient explanation would prevent someone from doing that in full rationality.
Of course, I'm just relying on my understanding of the ordinary English word "explanation." (Premise 3 in your original argument seems to rely on that sort of ordinary understanding.) But maybe you have some technical sense in mind. If so, what is it? If not, I can’t see how Lois Lane’s loving Superman all by itself counts as an explanation of Lois’ being on a date with Clark Kent. And so it still looks like we have here a counterexample to your premise 1.
You seem to think there are non-perspicuous explanations. Let's call them "obscure explanations." I guess each of these creatures of darkness manages to explain some proposition p while still allowing one who learns of the explanation to wonder “What’s the explanation of p?” in full rationality. That’s pretty weird. I don’t see the difference between an obscure explanation and no explanation at all.
But if there are such obscure explanations, I wonder why the identity-divine command theorist can’t say that your premise 3 is false since this is indeed an explanation, albeit an obscure one: that I am commanded by God to follow Christ explains my being commanded by God to follow Christ. Sure, you’re still left wondering why God commanded that, but this is the nature of obscure explanations!
Why couldn’t this count as an obscure explanation, on your view? But if it can, then premise 3 is false.
"I take it that a sufficient explanation would prevent someone from doing that in full rationality"
That "sufficient explanation" would then have to be a pretty beefy sort of sufficient explanation. Most of the things we ordinarily call explanations don't satisfy that.
Q: Why did the apple fall?
A: Because he pushed it off the table.
Of course, if you don't know about gravity, you will not get the explanation, no matter how rational you are.
Or suppose Pierre asks us: "Pourquoi Londres est belle?" And we tell him: "The buildings in London were designed by superb architects." In full rationality, he is still puzzled. He wouldn't be puzzled had we said the same proposition in French. But explanation is a relation between propositions and sentences. So it is possible to have an explanation and yet be in full rationality puzzled.
This suggests that what I said in my PSR book about explanation being a mystery-remover needs to be qualified. Explanation contains the crucial ingredients for mystery-removal, but one needs to "get it" in order for the mystery to be removed.
“That ‘sufficient explanation’ would then have to be a pretty beefy sort of sufficient explanation. Most of the things we ordinarily call explanations don't satisfy that.
Q: Why did the apple fall?
A: Because he pushed it off the table.
Of course, if you don't know about gravity, you will not get the explanation, no matter how rational you are.”
Sure, so it might be, in this case, that the sufficient explanation for a subject S is not just “he pushed it off the table,” but rather that proposition plus some stuff from S’s background knowledge. We don’t mention in our explanations propositions that we assume our interlocutor already knows; we trust that the information we convey in our explanations will ‘hook up’ with that stuff from the background knowledge and together form a sufficient explanation.
If S didn’t have any knowledge of gravity, I’d think it’s pretty clear that “he pushed it off the table” would not count as an explanation for S of why the apple fell. S is still baffled, after all. So I don’t think we have here a counterexample to my claim that a sufficient explanation of some event E for some subject S must prevent S from wondering “Why did E happen?” in full rationality.
So, still, I can’t see why you think “Lois Lane loves Superman” could count as an explanation of Lois’ being on a date with Clark Kent, for a subject who does not know that SM=CK. So, still, I think we have a counterexample to your original premise 1.
------------
“Or suppose Pierre asks us: "Pourquoi Londres est belle?" And we tell him: "The buildings in London were designed by superb architects." In full rationality, he is still puzzled. He wouldn't be puzzled had we said the same proposition in French. But explanation is a relation between propositions and sentences. So it is possible to have an explanation and yet be in full rationality puzzled.”
Well, I’m not so sure about this claim that explanation is just a relation between propositions and sentences. Suppose I code up some proposition in a special little idiolect. I let the sentence “BONK” express the proposition that it’s raining outside. Someone asks me why the streets are wet. “BONK!” I reply. On your view, I’ve explained to them why the streets are wet (it’s just like the Pierre case in all relevant respects). But surely I haven’t given them an explanation of why the streets are wet. So I’m not a huge fan of this view of explanation that you mention.
Also, I’m worried that your view of explanations makes them too easy to come by. Let the proper name “Bobby” refer to whatever proposition it is that explains the entire universe. Sadly, I don’t know how to spell out that proposition perspicuously, but by fixing the sense of “Bobby” in this way I can at least refer to that proposition and say things about it. So you ask me “Man, what’s up with the universe? Why is there something rather than nothing?”
“Easy,” I reply. “It’s because Bobby is true!”
As far as explanations go, surely this wouldn’t help you very much at all. But on your view, this comes out as a sufficient (if obscure) explanation. Right? This is just a case in which neither person “gets” the explanation. Or have I misunderstood your view? This looks relevantly like the Pierre case, in which you said there was a successful explanation. If I’ve understood your view correctly, I’d think this is a pretty bad result for the view.
------------
”This suggests that what I said in my PSR book about explanation being a mystery-remover needs to be qualified. Explanation contains the crucial ingredients for mystery-removal, but one needs to "get it" in order for the mystery to be removed.”
OK, but if there are such obscure explanations (successful explanations that one can fail to “get”), I wonder why the identity-divine command theorist can’t say that your premise 3 is false since this is indeed a successful explanation, albeit an obscure one: that I am commanded by God to follow Christ explains my being commanded by God to follow Christ.
Sure, you’re still left wondering why God commanded that, but this is the nature of obscure explanations! The source of your puzzlement, according to the identity-divine command theorist, is that you don’t “get” the explanation, since you don’t “get” the relevant identity between God’s commands and our obligations. It’s just like the Superman/Clark Kent case, in which you said there was a successful albeit obscure explanation.
Why couldn’t this count as an obscure explanation, on your view? But if it can, then your premise 3 is false.
I should have said that explanation is a relation between propositions, not sentences. Sorry.
Of course, to have something get explained to you, you have to understand the sentences to at least some degree. I don't know that I understand the sentence "BONK"--it's like a sentence in a foreign language. But what's interesting about the Pierre example is that we can rig the case so he satisfies the conditions for understanding "London" and "Londres", yet without his understanding that the two terms are synonymous.
"Sure, you’re still left wondering why God commanded that, but this is the nature of obscure explanations! The source of your puzzlement, according to the identity-divine command theorist, is that you don’t “get” the explanation, since you don’t “get” the relevant identity between God’s commands and our obligations."
But even if I get the identity between God's commands and our obligations, I can still want to know why God commanded me to obey Christ. And the explanation of why God commanded me to obey Christ isn't that God commanded me to obey Christ, but it has something to do with the kinds of reasons that God had for commanding this to me (perhaps that it would be good for me to obey Christ).