# Followup on the Moral Argument for Theism

A commenter named Nikki argued against my post Fundamental Reality XII: The Good, and the Not.

Nikki writes:

I don't think the post's argument works - I'd argue that non-theistic morality can be objective and well-grounded, or at least be no worse off in those regards than theistic morality is.

So the first part of this post that really jumped out at me is the claim that if morality is objective, it must be like a mind. Frankly, to me this seems not only false, but a category error. Morality is things like systems, principles, rules, etc. - I'm not sure what the exact best word choice is. The point, though, is it is a thing that minds use, but not in and of itself a mind. You describe morality as approving or disapproving certain things, but this seems to be conflating things like "this abstract system contains claims that X is good/bad," which could validly be said about morality, and "this abstract system itself consciously judges that X is good/bad," which could not. It is us who use morality to consciously make those judgements.

As an analogy, personality traits are part of minds, but not minds themselves - to speak of them, by themselves, being conscious, thinking, willing, etc. would be a fundamental mistake. (Though Inside Out was a pretty fun movie). I'll admit though, I don't actually think that's the best analogy. I'd argue the set of laws of logic or mathematics are an even better example of something that is a feature of minds - but is not, and could not possibly be, a mind in itself. However, you've said in an above comment that logic is also a description of God's character.

(Perhaps a bit of a sidetrack here, but I don't think this could be true either. I believe that you've stated elsewhere that while you believe God is metaphysically necessary, he is not logically necessary - but of course, it is logically necessary that the laws of logic or mathematics are true. I don't think the dependence you're arguing for could work, even if God exists in some sense. That said, as one might guess, I don't think God is metaphysically necessary in the first place.

In fact, I have doubts that there is even a "metaphysical necessity" distinct from logical necessity at all. I find Chalmers' arguments in his paper "Does Conceivability Entail Possibility?" fairly convincing in this regard. I do think there are some weak points, but it seems to me that at least it shows that even if there is a metaphysical modality separate from logical modality, we don't currently have a good reason to believe in it. I know there are several relevant arguments on this blog, but well, I can't discuss every single reason for and against the existence of God in this post, so here I'm trying to stick with things related to the original topic/what's been mentioned in previous comments on it. I might debate the other arguments later.

As a note, Chalmers' arguments there are important for the case he makes that consciousness is not physical, because they counter the reply of some materialists that consciousness is metaphysically the same as a physical property, even if it cannot logically be derived from other physical facts. Others have argued that this causes problems for theists who both defend the metaphysical necessity of God and the non-physicality of consciousness. I suppose this may not apply to you because you've said you can't rule out that consciousness is physical in some sense, as in what Chalmers calls "Type-B Materialism," but I did think it was interesting).

Alright, back to the main topic. Does an objective morality depend on God? The whole field of moral philosophy is certainly not something I can fully describe in one post, but I'll start with something interesting you said in your own previous post in this series:
"Even people who say there's no such thing as ethical truth suddenly sound quite different when somebody treats them unfairly."

I suspect that in that statement is at least a hint at what the basis for a nontheistic objective morality might be like. If there is an objective morality, I think it has something to do with the symmetry between you and others - if you don't treat others well, what's to prevent them from doing things to you that you don't want? Even if evil may sometimes have short-term rewards, people committing acts like theft or murder or terrorism ultimately make things worse for everyone, including themselves. Note that these statements do not depend on God to make them true. And I think several strands of thought, such the Golden Rule, Kantian morality, Rawls' veil of ignorance, and even some game-theoretic analyses, among others, all point towards something like this in a sense.

Now, this may not be very compelling - I'm being vague and have not spelled out a fully detailed nontheistic system. Furthermore, many of the systems I've cited actually contradict each other. Nevertheless, I think that there are important shared elements that don't depend on a belief in God to be convincing (well, Kant's morality was theistic and the Golden Rule is a part of many religions, but I don't think everything along the lines that I've mentioned is). So it seems that the claim that no secular account of morality can possibly succeed isn't very certain. I'll note that you linked in your previous article to the SEP's article on Moral Naturalism, but merely said those systems were "problematic" without really discussing the individual ideas presented there, although there are many important nonreligious thinkers whose ideas on morality are much more detailed than mine. (I won't complain about that too much though - after all, I'm not discussing every form of theistic morality in this post myself).

Some more notes: 1. Speaking of moral naturalism, even on atheism, that isn't the only option available for an objective morality. While I agree naturalism and atheism are often found together in practice, it is still possible for an atheist to be a non-naturalist, including about morality. So even if morality cannot be justified on naturalism, you would have to show that God specifically is the only one who can ground morality, not some other non-natural element.

2. Above, Scott Church argues that on naturalism, the universe does not care about us and we are fundamentally unimportant, so it cannot ground objective morality. But the universe itself does not have to care about us/be a moral agent for morality to be objective! I'd argue that if morality, say, applies to all rational beings, it is objective, and the universe not obeying it does not matter because the universe is not a rational agent. The laws of rationality themselves are a good analogy for this - the universe, itself, does not reason, and it requires minds to use reason, yet the standards of rationality are fully objective (and not derivable from physical equations, by the way). And even on theism, it is agreed that some things, like inanimate objects, are not and cannot be moral, yet again, that does not prevent morality from being objective. Related, while pure pleasure-maximization/pain-minimization has several well-known problems, so I doubt that's the full objective morality, I do think there are non-arbitrary reasons why those are at least important. They are necessarily important to us by their very nature - no one can truly be indifferent to them even if they claim to be. And even if the universe does not care about them, I take the anti-nihilistic view that it is precisely the fact we care that matters - it's not as if the universe has any rule against that!

3. I've seen this part stated before in some other comments on the blog, but I think it's important enough that I'll state it again (especially since unless I'm missing it, I don't think I've seen a response). Escaping the Euthyphro dilemma by saying that God is identical to goodness can only work if we have good reasons to believe that the two could possibly be identical. I don't think we have those (unlike for the triangle case, in which we do have reasons to believe that "having three sides" and "having three angles" are the same, even though those are logically necessary), but we do, in fact, have reasons to believe the opposite. As I wrote at the beginning of my post, if God is to be viewed as even like a mind, he cannot possibly be identical to morality even if he is an (ultimately) moral agent. For instance, one of the important reasons to consider God like a mind is that he is supposed to be able to take actions, but morality cannot, by itself, take actions. (Also, I'll admit I don't know whether your analysis of Plato is accurate, but even if it is, it's generally fine to take inspiration from an argument and adapt it to your own views. After all, in the original article, you said you used "Hume's Is-Ought dictum in a manner which he would have thoroughly disapproved of!")

As a final statement, I don't think theism is actually better at convincing people of being moral than secularism. There's some evidence that nonreligious people are even more moral than very religious people, but interpretations are controversial and I'm focusing more on purely philosophical points here. (I do suspect nonreligious people being more moral than the religious, if true, would be a particularly big problem for theism and theistic morality. I think the evidence at least shows that the nonreligious are generally not less moral than the religious, but you've agreed in another article that for some senses of "good," religion is not strictly necessary for it, so that may not be a big problem for you). But anyway, you've agreed that not all rational people might be convinced by theistic arguments, and it's been pointed out above that you can always ask questions like "Why should you follow God's commands?" so that seems to be an issue. Of course, you might very well always be able to ask similar questions about any nontheistic system, and rational people might not find it convincing. But my point was that secular morality is at least equal to theistic morality in this regard, and while this is a bit speculative, perhaps some of the reasons above might make the former even more convincing than the latter.

My reply got pretty long, so I'm turning it into a blog post.

Dear Nikki,
Welcome to my blog, and thanks for your interesting comment. However, I am not sure that your arguments are actually directed against the specific argument I am making. Here are some replies (not in the order of your points):

I. Objective Morality is a Premise in the Moral Argument

You make a good case defending this proposition: It is possible for a non-theist to rationally come to believe in the existence of an objective ethical system, without thereby coming to believe in God. However, I also believe that this is the case!

In fact, if this were not true, there would be little rhetorical point in presenting a Moral Argument for God's existence.  In order for an argument for God's existence to be capable of being convincing, there have to be some people out there who agree with the premises of the argument, but have not yet realized that the conclusion follows (or at least, is made more probable) by the premises.  I obviously do not deny the existence of non-theistic moral realists, because they are the target audience for my post!  (That is why I presented an argument for ethical realism in part XI before describing how  I think Theism grounds ethics in part XII.)

Now obviously, if the a nontheistic argument for objective ethics happened to take the form of an entirely satisfactory reduction of concepts like ethical obligation into naturalistically acceptable terms—e.g. in terms of physical facts of the sort that even Sean Carroll would accept—then the Moral Arguments for Theism would fail, since there would be no additional work for God to do in terms of grounding ethics.  (There might still be a need to ground the laws of physics in some way, but no additional and separate need to ground ethical truths.)  But of course, if you could show that this were true, you would have just solved a very famous and important problem in philosophy!  So I sort of doubt you really think that we can know this to be the case.  And if we cannot know it to be the case, then there is room for discussing non-naturalistic groundings of ethics, in a probabilistic argument for Theism.

You sketch some ways in which you think an non-theistic grounding for objective ethics might work (which fall into the rough family category of what I called "Kantian approaches'' to ethics in part X).  As I explicitly stated in that post, Kantianism is not as friendly to the Moral Argument, as Platonism or Aristotelianism is; although I don't think it is utterly hopeless on that front.  (Kant himself made a sort of pragmatic argument for Theism from Morality, but he didn't agree with metaphysical arguments of the sort I'm discussing.)   The only conclusion I explicitly drew from Kantianism was:

If Ethics can be deduced rationally as in the Kantian system, then one can at least deduce that if the Universe originates from something like a mind, that mind should also be able to appreciate ethical truths.

So the point you are making was to some extent already acknowledged in this series.  (Of course, on classical forms of Theism, where God is something like the ultimate Reason or Logos behind the Universe, this would still end up identifying God with moral goodness in some deep sense; but such classical views are necessarily bordering on Platonism anyways...)

B. Moral Naturalism and Non-Naturalism

By the way, I revisited the SEP article, and found to my dismay that it had been edited in a way that removed (without refutation) some of the critiques of Moral Naturalist positions. Here is the original version of the article.  If you look, for example, at the original article's section 4.3, you can see what appears to me to be a pretty desperate attempt by Jackson to make naturalistic ethics work, together with (what appears to me to be) a pretty strong refutation in terms of the permutation problem.  But the main point is not the refutation of that particular idea, but that I don't see any way forward mentioned in the article which doesn't seem to have serious problems.

You write:

Speaking of moral naturalism, even on atheism, that isn't the only option available for an objective morality. While I agree naturalism and atheism are often found together in practice, it is still possible for an atheist to be a non-naturalist, including about morality.

Yes, obviously.  Such views exist (which is why I mentioned them in part X of this series). In fact, individuals with such views (e.g. Moral Platonists) are closer to being the target audience of this post, then perhaps you are.

So even if morality cannot be justified on naturalism, you would have to show that God specifically is the only one who can ground morality, not some other non-natural element.

No, because as I tried to make it clear at the beginning of this series that I wasn't trying to present a deductive, logically watertight argument for Theism.  As I said in Part I:

Even if there are no strictly deductive arguments (from indisputable premises), there are still going to be plausibility arguments pointing in various directions.  It's irrational to put too much faith in plausibility arguments, but it's also irrational to be completely insensible to them.

So the mere existence of logically possible positions, besides the one I argue for, doesn't bother me.  The question is which positions are most credible.

On the plausibility front, it seems to me that once you start modifying your metaphysics in order to accommodate objective ethics, it would be irrational not to take that into account when assessing the probability of other metaphysical hypotheses.  Ethical Monotheism is, among other things, the belief that a fundamentally good being exists.  The plausibility of this statement depends in part on what we think moral goodness is.  For example, on the view that:

1. "Morality is a emergent and subjective set of feelings found in some of the higher apes, conducive to their evolutionary survival, but having no basis in any metaphysical reality"

then the idea that there exists a fundamentally good being outside the physical universe—which did not evolve—is totally absurd.  On the other hand, if:

2. "moral facts are necessary truths, which tell us something substantive about the structure of non-physical realities",

then the idea of a fundamentally good being is, though not logically compulsory, at the very least far more plausible than on viewpoint (1) than (2).  Do you agree with that?  If so, then you are necessarily agreeing with me that the Moral Argument for Theism has significant probabilistic force.

[Notes: I am not saying these are the only possible views.  Also, hypothesis (2) does not necessarily deny biological evolution, as it is possible for evolved systems to recognize necessary truths such as mathematical theorems.]

C. The Role of Analogies

Let me remind you a bit of the context of my argument in the Fundamental Reality series.  In parts II-VI, I argued that it is plausible that there exists some fundamental reality which explains everything else, I discussed some properties this entity should have, and after reviewing various candidates I suggested that (based on the mathematical character of the laws of physics) the two most plausible metaphors for understanding this fundamental reality are:

* something like an equation
* something like a mathematician

Now it is important to remember that both of these ideas involve metaphors!  Obviously, if a Naturalist says that some equation provides the deepest truth about the Universe, that doesn't mean this assertion is being made about a set of chalk lines on a blackboard.

Similarly, if a Theist says that God is like a mind, that doesn't mean that this Mind is like our mind in every respect.  In particular, Classical Theism proposes a mind for whom there is no distinction between its subjective beliefs and objective reality, and also no distinction between its subjective preferences and objective morality.  This is obviously very different from evolved primate minds like our own!

You wrote:

So the first part of this post that really jumped out at me is the claim that if morality is objective, it must be like a mind. Frankly, to me this seems not only false, but a category error. Morality is things like systems, principles, rules, etc. - I'm not sure what the exact best word choice is. The point, though, is it is a thing that minds use, but not in and of itself a mind. You describe morality as approving or disapproving certain things, but this seems to be conflating things like "this abstract system contains claims that X is good/bad," which could validly be said about morality, and "this abstract system itself consciously judges that X is good/bad," which could not. It is us who use morality to consciously make those judgements.

and

As I wrote at the beginning of my post, if God is to be viewed as even like a mind, he cannot possibly be identical to morality even if he is an (ultimately) moral agent. For instance, one of the important reasons to consider God like a mind is that he is supposed to be able to take actions, but morality cannot, by itself, take actions.

I think perhaps you missed the amount of qualifying words I put into my reasoning.  What I wrote was (emphasis added):

But now observe that morality is at least a little bit like a mind, insofar as it approves or favors certain things, and disapproves or disfavors other things. So a fundamental morality would have something analogous to will or desire, and in that respect it would be more like a mind than like an equation, as in Theism.

The point here is not that an objective morality is exactly like a mind, but that it in certain respects more similar to a mind than (say) the equations of the Standard Model are, namely that the Standard Model does not encode any judgements that certain states of affairs are desirable or undesirable (as opposed to probable vs. improbable).

Now, obviously, when we say that God is personal, and can do things like forgive or create, we are adding more to our concept of God then is implied by the mere abstract notion of a metaphysical objective morality.  In my understanding of God, we are adding more to our idea of divinity than the idea of a Platonic form of the Good, but we are not necessarily taking anything away.

In other words, in my conception of God, God is such that he is good, not in an accidental (happenstance) way, but in an essential way, because all goodness in the universe in some sense participates in his goodness, just as all existence participates in his existence.  (The latter claim, of course, obtains for any fundamental reality which is taken to explain all other things.)

D. God Transcends the Abstract/Concrete Divide

Another commenter, St. David Madison, replied to your comment by saying (in part):

"You draw an analogy between morality and personality traits and then point out that personality traits are not conscious and do not themselves think. However, personality traits cannot exist without a personality that possesses those traits."

This is certainly a reasonable distinction to draw in general; and we could indeed escape from the supposed category error by simply replacing the words "objective ethics" with "that which grounds objective ethics, whatever it is."  But I think I am instead going to double down on this idea, and say that this supposed category distinction between abstractions and concrete objects breaks down when one is speaking about divinity, just as the distinction between particles and waves breaks down at the subatomic scale.  If God is the source of all else that exists, he must unify within himself the perfections of both abstractions (necessary, eternal, unchanging) and concrete realities (which are causally active, definite, individual etc).

This is indeed, already implied by certain sorts of religious language, in which God is portrayed not as some good or beautiful thing, but as the Supreme Goodness or Truth or Beauty or Life etc.  For example, in the Gospel of John, Jesus asserts his divinity by saying that he is the Way, the Truth, and the Life, which is not the sort of thing that a Positivist philosopher would consider a well-formed statement (a person cannot be an abstract quality).  But I am not convinced we can restrict our language in the way the Positivists wanted to do (I don't think Positivism even satisfies its own criteria of meaningfulness).  What this religious language points to, is an insight into the nature of divinity as a necessary being, in which all other realities are grounded.  A proposition about a created being can be true, but only the ultimate reality can be the Truth.  In other words, denying the applicability of the concrete/abstract distinction is not something I am doing merely to avoid a logical puzzle, but is already implied by standard religious language about God.

This sort of language about God makes Classical Theism radically different from traditional forms of polytheism, in which the gods are simply regarded as more powerful individuals than us, who still can be born/killed, have conflicts with each other, make mistakes etc.  Yeah, obviously the preferences of finite beings like ourselves can't possibly ground objective ethics, which was the whole reason why Plato went in a platonic direction instead.

Furthermore, I don't think we can avoid postulating this sort of concrete/abstract unification, simply by rejecting Classical Theism, as Naturalism seems to me to imply exactly the same thing.  For example, if the fundamental reality is something like a mathematical equation, then we are asserting that it is both an abstract piece of mathematics—which can in principle be understood by humans—AND ALSO the governing principle controlling the universe.  In other words, when a Naturalist does physics, they are still are postulating that the fundamental reality is a λογος, i.e. a rational principle.

Of course, I'm not saying that the equations we write on the blackboard, or in our minds, are strictly identical to the actual laws of physics, which obviously exist whether or not we ever discover them.  But if we asked, "what are the fundamental laws of physics like" we can't point to anything other than to our abstract human formulation of the equations, and then lamely add "except that it also exists as an actual concrete reality, in a way which transcends our human abstractions".

In the same way, objective morality exists even apart from human processes to reason about what is or is not moral—So I'm not saying, that this latter, social process of reasoning is equal to God.  Rather it is goodness as it actually exists (which our human reasoning is a mere approximation of) that is rooted in God's nature, as the ultimate Goodness that other things participate in.

E. Implications for Euthyphro

Escaping the Euthyphro dilemma by saying that God is identical to goodness can only work if we have good reasons to believe that the two could possibly be identical.

This is a strange way to discuss this subject, given that the (modern) Euthyphro dilemma is typically phrased, not in the form of a deductive argument, but in the form of a challenge to Theists to explain their beliefs more clearly.  It's phrased in the form: "Do you believe A, or B?" (both of which have unpalatable consequences).  But if A and B are not, in fact, exhaustive possibilities, because some other option C is conceivable—and if in fact C was the belief of most ethical monotheists historically, as well as myself—then merely pointing this out is sufficient to defuse the dilemma.

That being said, there is a good reason to think that, if God exists at all, he can ground morality.  Recall that God is, by definition, the explanation for all entities other than himself.  (That's the whole point of Mono-theism, to have only one ultimate entity.)  So if God exists at all, he either grounds or creates all other realities.  Now if there is objective ethics, then ethics counts as one of these realities.  Since it doesn't make sense to create ethics (since at least some ethical principles are non-arbitrary, necessary truths) then he must ground it.  (The same argument would hold for logic or mathematics.)

Now, to be clear, this is an argument that God grounds ethics.  It is not an argument which explains how God grounds ethics.  To understand how God grounds ethics we would have to first have direct perception of the divine essence, which we don't possess.  Instead, we only know the things which proceed from the divine essence, and we have to learn about what God is like, as best we can, from that.

If you like, you can take "a concrete reality which grounds ethics" as a defining property of God, and then ask questions like i) what other properties would such a being need to have, and ii) is there good reason to believe that such a being exists?

If you will allow me to make a more meta-level argument.  It seems to me that giving the Euthyphro dilemma as an objection to Classical Theism is historically obtuse.  It's like proposing the Equivalence Principle as an objection to General Relativity, when the Equivalence Principle was in fact the motivating thought experiment that led to GR in the first place.  In the same way, the question of what the gods (or really God) has to be like in order to justify treating piety as a virtue, was the underlying question motivating the Euthyphro dilemma.  But somehow atheists never say to themselves, "Geez, the fact that this famous philosophical argument was introduced in a Platonic dialogue, by a theist whose ideas laid the groundwork for the most mainstream philosophical formulation of Monotheism, maybe is a reason to think I've missed something and the argument isn't actually a knock-down in favor of Atheism."

(To be sure, arguments aren't "owned" by philosophers and there is no reason in principle why an argument by a philosopher P can't sometimes be turned against P's own worldview.  So sure, maybe there is some very subtle reason why GR is still inconsistent with the best formulation of the Equivalence Principle.  But if somebody sends me and email about why they think GR is inconsistent with the EP, and it shows no awareness of why some people have historically thought that GR satisfies the EP, then it's unlikely that their "gotcha" question about how the EP refutes GR has much merit.  Ditto for Classical Theism and Euthyprho.)

F. Metaphysical vs Logical Necessity

Now to be fair, you did explain why you don't believe in scenario C.  In addition to your "category error" assertion, you add this:

In fact, I have doubts that there is even a "metaphysical necessity" distinct from logical necessity at all. I find Chalmers' arguments in his paper "Does Conceivability Entail Possibility?" fairly convincing in this regard.

So on your recommendation, I read through this Chalmers article and I found it pretty unconvincing.  Why should reality be fundamentally scrutable to us?  Or said another way, why can't there be propositions P which are necessary, but only a mind fundamentally more powerful than the human mind could see why they are necessary?  It seems hubristic to think that human reasoning has access to every possible necessary truth.

Ironically, the reason I don't believe in Chalmers' thesis here, is actually very similar to the reasons why I side with Chalmers over Dennett when it comes to Consciousness.  While Dennett makes an interesting philosophical case for the reducibility of conscious experience to neurological facts, ultimately I concluded that Dennettism can only work if Dennettism is true by logical necessity.  In other words, that once you've specified all the physical facts then Dennett's views on consciousness follow automatically.  And it seems to me that this is simply not the case.

Similarly, Chalmers' idea that if we specify all the physical nonmodal facts, then a single set of views about modal necessity must logically follow (to idealized human reasoners) seems plainly false to me.

(Assuming it even makes sense to distinguish between "modal" and "nonmodal" facts in this way.  This is an important distinction between analytic philosophy and traditional medieval philosophy.  Analytic philosophy sees modality as primarily a feature of certain propositions, and only secondarily as a property of things.  While Aristotelian/scholastic philosophy sees modality as primarily as a property of things, while only secondarily as an attribute of propositions.  A scholastic might argue that the analytic habit of immediately jump to always reasoning about maximal "possible worlds" obscures the role that modal concepts play in causal reasoning, which involves specific concrete entities.)

Anyway, since you hold to something like Chalmers' view, here's a dilemma for you: Is the proposition expressing this view itself a logically necessary truth?

(P) There are no metaphysically necessary truths, other than logically necessary truths.

If you say that P is logically necessary, then there must be a proof that it is true which follows deductively from the definitions of the words.  What is that proof?  As far as I can tell, none exists.  Certainly Chalmers doesn't give a logically conclusive proof in that article, he just gives some reasons why he considers belief in P to be plausible, which is not the same thing.

On the other hand, if is not logically necessary, then either it is contingent (which is inconsistent with the usual S5 rules for modal logic) or else it is an example of a metaphysically necessary (but not logically necessary) truth, in which case it refutes itself.

One could make a similar, superficially less "meta" argument for the same conclusion by considering the proposition:

(N) A necessary being exists.

A standard analytic argument from S5 modal logic implies that either: i) N is necessarily true, or ii) N is necessarily false.  So which of these is logically necessary?  I say neither, but if you disagree then what do you think the proof of N or its negation would look like?

G. Can God be the grounds of Logic?

I believe that you've stated elsewhere that while you believe God is metaphysically necessary, he is not logically necessary - but of course, it is logically necessary that the laws of logic or mathematics are true. I don't think the dependence you're arguing for could work, even if God exists in some sense.

This is a little compact, but I'm guessing your argument is something like the following:

1. A contingent truth cannot ground a necessary truth.*
2. God's existence is logically contingent.
3. But logic itself is logically necessary,
4. Therefore, God cannot ground logic.

[*I suppose there is some sense in which, if a Cat walks onto a Mat, this arguably grounds the necessary proposition: "Either the Cat is on the Mat or the Cat is Not on the Mat" by virtue of being a truthmaker for one of its disjunctives.  But I won't pursue this possible counterexample further, since I don't think it is relevant to the sense in which God grounds logic.]

But this argument is fallacious, because when I say that God grounds logic, I am making a metaphysical statement rather than a logical one.  From the perspective of metaphysics, both logic and God are (in my view) metaphysically necessary, and it is not at all impossible for a necessary statement to ground another necessary statement.  In other words, we have to distinguish between:

1a: A logically contingent truth cannot logically ground a logically necessary truth.

which is true, and:

1b: A logically contingent truth cannot metaphysically ground a logically necessary truth.

which does not in any way follow from 1a, and I would say it is false.

H. What Metaphysical Necessity Means

Actually, there is a better way to put this which makes the concept of "metaphysical necessity" somewhat less mysterious.  The right way to talk about this is to make Aristotle's distinction between that which is necessary to us (axioms of human thought) and that which is necessary in itself (propositions which could not have been otherwise).

When we say that a proposition is metaphysically necessary, we merely mean it falls into the latter category.  The adjective is misleading since, unlike the cases of "logical necessity" or "nomic necessity" (which mean necessary given certain specific principles), the phrase "metaphysically necessary" simply means whatever is necessary simpliciter, i.e. that which (without adding any qualifications) could not have been otherwise (whether or not the reason for its necessity is known to human beings.)

On the other hand, logical necessity is an example of what is necessary to human beings, i.e. an axiom of human reasoning, or a particular technique L used to prove the impossibility of certain propositions.

So, the proposition P from earlier boils down to:

(Equivalent to P): If a proposition cannot be proven to be impossible by technique L, then it really is possible.

while I see no reason to believe that technique L is sufficient to uncover all possible cases of necessity.  Especially since technique L does not even seem to be powerful enough to refute the statement that no concrete entity whatsoever exists.

This relates of course to cosmological considerations as well.  As is well-known, if P is true, then the basic principles of existence are just contingent "brute facts" which means they are not true for any reason at all.  So there is an obvious reason to postulate a necessary concrete entity, which is that it serves as a starting point to explain why anything else exists at all.

This reason to want a necessary being, does not seem to depend on us being able to know why the being is necessary.  This is the Thomistic viewpoint on the Cosmological Argument, and it seems to me to be the only possible middle ground between Anslemian positions (there is a valid Ontological Argument for a necessary being from pure logic) and explanatory nihilism (there is no good reason why the universe exists, it just does).

(Now you could just double down and say, I have no idea what you mean by the phrase: could not be otherwise'', please explain it to me; and then refuse to accept any answer I give other than one which reduces it to logical implication.  But the same technique could be done to motivate skepticism towards practically any other concept, including the other concepts in this discussion like "mind" or "good" or "abstract" or "grounds".   (It is not even clear that logical necessity can be fully explained without an infinite regress, as  St. Lewis Carroll pointed out in his Achilles and the Tortoise dialogue.)  I don't claim to have a definition of metaphysical necessity that would satisfy Socrates, but if we make that the standard, there aren't going to be very many philosophical terms left!)

I. An Irrelevant Topic

As a final statement, I don't think theism is actually better at convincing people of being moral than secularism.

This is just so totally irrelevant to the metaphysical questions behind the Moral Argument for Theism, that perhaps I should simply refuse to respond to this entirely.  It's really just a complete change of topic.

God could be the metaphysical grounds for morality, even if every single human being on Earth were an atheist, or even if every single theist were morally worse than every single atheist.  These motivational questions really have nothing whatsoever to do with the question about what metaphysical theses are made more plausible, if we subscribe to moral realism.  I wrote my blog post Is it Possible to be Good without God? precisely because I was annoyed by how regularly people seem to conflate these totally unrelated questions.

(I'm not saying that the degree of goodness of religious people can't potentially be used as an evidential argument for or against the existence of God.  What I am saying is that it is a mistake to allow such sociological questions to contaminate our interpretation of the thesis that God grounds ethics.)

That being said, I''ll take the bait and say I do think there is some pretty serious question begging required for a non-circular argument that atheism is fully compatible with moral behavior.  For one thing, if a being such as is described by Classical Theism in fact exists (a perfectly wise and holy and good being, who created us and is the source of all our goodness), then we have the moral obligation to worship and obey that being, and to reflect God's holiness through a life of prayer and repentance, dedicating our earthly activities to the glory of God.  It is difficult to see how an atheist can satisfy that obligation, because for the atheist these activities are just distractions from a different, more secular understanding of what the good life consists of.

(To be sure, if the atheist has some intellectually honest reasons why they think God does not exist, then this may well be a mitigating circumstance that reduces—or even eliminates entirely—their culpability for this omission.  But if we are discussing the question of which beliefs make it easier to be moral, then usually mitigating circumstances are considered mitigating precisely because they make it harder to be moral.  Furthermore, a lack of culpability does not remove all of the causal consequences of trying to place our ultimate happiness in things other than God—what Christians call idolatry.)

I do suspect nonreligious people being more moral than the religious, if true, would be a particularly big problem for theism and theistic morality.

From the standpoint of Christian doctrine, it is not actually clear why this should be.  Merely having knowledge of God's existence does not necessarily translate into obedience, and in some cases knowledge can make people morally worse since they ought to behave better but don't.  As Jesus' brother St. James said:

You believe that there is one God. You do well. Even the demons believe—and tremble!  (James 2:19)

and as Jesus himself said:

“Not everyone who says to me, ‘Lord, Lord!’ will enter the kingdom of heaven, but only the one who does the will of my Father in heaven.  On that day many will say to me, ‘Lord, Lord, didn’t we prophesy in Your name, drive out demons in Your name, and do many miracles in your name?’  Then I will announce to them, ‘I never knew you! Depart from me, you lawbreakers!’ ”  (Matthew 7:21-23)

The Pharisees were among the most "religious" people in Jesus' day, and many of their leaders handed Jesus over to Pilate to be crucified.  See also St. Paul's observations of religious people in Romans 2.

According to Christianity, what people need to be transformed morally, is not so much knowledge as grace.  Knowledge is good if it helps us acknowledge our need for grace, but not so much if it makes us look down on other people.

I think the evidence at least shows that the nonreligious are generally not less moral than the religious...

I'm not sure what evidence you are referring to here, or how you could actually know this to be the case.  If your claim is just that religious people can be morally weak and inadequate, well I already knew that from my own life, without looking at anybody else's.

If it refers to survey data, you have the problem that what many polls of religious affiliation captures a lot of individuals who only identify as religious in a nominal sense.  Polling nominally religious people, and asking about their rates of divorce, adultery etc. is sort of like asking whether watching the Olympics on TV makes people more physically fit!  It's the wrong question to study.

If you are referring to personal experience, I can only say that while I know good and bad seeming people (emphasis on "seeming", it's not my place to judge them) who are both religious and non-religious, the most loving and self-sacrificial people I know seem to be religious.  And religion also often plays a significant role when very bad seeming people repent and turn their lives around.  Furthermore I have very often heard people refer explicitly to God when they explain why they did something morally difficult, while I cannot ever recall in my personal experience ever hearing somebody say that they did something morally difficult because atheism is true.  (I mean, I could imagine such a motivation: e.g. God isn't going to save this person, so I have to.  But I don't think I've ever heard anyone explicitly say this "in the wild" so to speak.)

By comparison, studying secular ethics seems to itself have little observable consequences in terms of making people better.  This could be taken as a critique of secular ethics, but it might be better taken as a critique specifically of what modern analytic philosophers mean by ethics as a discipline (as opposed to ancient philosophies, which were typically viewed as a way of life that had to be put into practice, in order to be understood).  I mean, why should studying little numbered arguments about whether ethics is objective, or arguing about what to do in some controversial edge case involving trolleys, actually help one to build habits of life that make one treat your fellow human beings better, and a community which helps support you in doing so?  Religion is one of the few ways of getting such support in the modern era.  (There are some others, but they are getting sparser in an increasingly disconnected age.)  While this isn't necessarily an argument for God's existence, it does make your thesis that serious religious practice is totally orthogonal to ethical accomplishment seem pretty implausible.

I called this an "irrelevant topic" because it isn't terribly relevant to the validity of the Moral Argument.  But of course, from the perspective of what ultimately matters, it is this section that is most important, and the rest which are of lesser relevance.  If Christianity is true, then what will matter the most in the end is not whether you are persuaded by this or that specific argument for Theism, but more whether your heart is open or closed to God at a deeper level than that.  Jesus has promised that those who truly seek God will find him.

If you take it as a goal to be as moral of a person as you can possibly be, then that is at least a start along that road—even if the final destination is going to be, in some ways, quite different than what you expected when starting out on that journey.  But somewhere along the way comes the recognition that you can't actually be good, and need help to do better, and that is where concepts like grace and salvation start to make more sense...

Blessings,
Aron