MandM header image 2

John W. Loftus on The Christian Illusion of Moral Superiority Part II

May 8th, 2009 by Matt

In my previous post, John W. Loftus on The Christian Illusion of Moral Superiority Part I, I argued that Loftus’ position was based on a confusion between ontological and epistemological foundations. I will now address his arguments against divine command theory.

Loftus’ Arguments Against a Divine Command Theory
After misconstruing the divine command theory as an epistemological theory, Loftus offers a series of fairly standard criticisms of divine command theories. Here I will focus on the two most important,

1. The contention that acceptance of a divine command theory makes it impossible to coherently attribute goodness to God; and,
2. The claim that a divine command theory makes God’s commands and hence, moral obligations, arbitrary.

The emptiness of God is Good
To support the first contention Loftus offers two reasons which I term the Empty Tautology Objection and the Arbitrariness Objection. I will address each in turn.

Empty Tautology Objection
The first is that on a divine command theory claiming God is good becomes an empty tautology as seen in this statement, “If we think that the commands of God are good merely because he commands them, then his commands are….well….just his commands.”

Here we see Loftus mistakenly assuming that a divine command theory is an account of “goodness.” As I argued above it is not, it is an account of deontological properties such as right and wrong. Suppose, however, we recast the objection in terms of deontological properties, the objection still fails. Loftus contends that if something is right because God commands it then God’s commands become right by definition; and hence, the claim ‘God commands what is right’ becomes a tautology, it is just the assertion that “his commands are….well….just his commands.”

This argument is mistaken. It rests on the assumption that if something (in this case an action) has a property (in this case the property of being right) because it has another property Q (in this case the property of being commanded by God). Then the claim that Q is P is a tautology with no substantive content. However, this assumption is mistaken.

Consider the claim that water is H20. This is a statement that one property is constituted by and hence cannot exist independently of, another property. Yet one who makes the claim ‘water is H20’ is uttering a substantive, non-tautological statement. Claiming ontological dependence between two properties does not entail that a statement attributing such a relationship is a tautology.

Independent Standard
The second and perhaps major argument Loftus offers is that we cannot call God’s commands good unless we appeal to a standard independent of him. He expresses different versions of this objection in several places in his argument. He writes,

We cannot call them good, for to call them good we’d have to have a standard above them to proclaim that they are indeed good commands. But on this theory they are just God’s commands. God doesn’t command us to do good things, he just commands us to do things. “All that could mean is that God wants us to do what he wants us to do.

Responding to Moreland he makes the same argument,

But think of what Moreland is saying here. He’s saying that morality is grounded in God’s nature, not in his commands. But this is a difference that makes no difference. It does no good to step back behind the commands of God to God’s purported nature at all. For we’d still want to know whether or not God’s nature is good. God cannot be known to be good here either, without a standard of goodness that shows he is good. For unless there is standard that shows God is good beyond the mere fact that God declares that his nature is good, we still don’t know whether God is good.

Elsewhere he makes essentially the same point,

[T]he modified divine command theory becomes equivalent to the autonomy thesis: the Good (or right) is not good (or right) simply because God commands it. Furthermore, if this is correct, then we can discover our ethical duties through reason, independent of God’s command. For what is good for his creatures is so objectively. We do not need God to tell us that it is bad to cause unnecessary suffering or that it is good to ameliorate suffering; reason can do that. It begins to look like the true version of ethics is what we called ‘secular ethics.’

Loftus goes on to summarise his point as follows,

The bottom line here is that if there is no moral standard for us to appeal to when we’re assessing the claim that God is good, and all we have to go on is the fact that God said he was good, then we cannot asses whether or not God is good. We still haven’t been given as answer to what he means by the word good.

Much can be said about these arguments. First, Loftus again construes divine command theories as offering an account of axiological properties such as goodness. I have already noted this is mistaken.

Second, Loftus again confuses ontology with epistemology. He affirms that “God cannot be known to be good … without a standard of goodness that shows he is good.” if there is no moral standard for us to appeal to when we’re assessing the claim that God is good, … we cannot assess whether or not God is good.” Also, “we can discover our ethical duties through reason, independent of God’s command.” These are all statements about epistemological independence of our beliefs about right and wrong. They show that one can know right and wrong independently and prior to any beliefs one has about what God commands. The problem is that divine command theorists do not deny this. They do not claim that one needs to believe in the existence of divine commands to be able to tell what is right and wrong. They claim that the property of being wrong is ontologically dependent on the property of being contrary to God’s commands.

Turning again to the analogy regarding H20 and water, most people throughout history have known what water is without knowing water is H20. In fact, one cannot determine whether H20 really is water unless we have an independent standard of some sort in our minds as to what water is. It does not follow from this, however, that water is not H20 and given that something cannot exist apart from itself, it does not follow that water and H20 are ontologically independent. The same is true of divine commands; people can know what is right and wrong without believing in God. Further, to know whether God’s commands are right, one will have to have some standard of what things are right and wrong. It does not follow from any of this that right and wrong can exist independently of divine commands.

Loftus’ arguments for the claim that one cannot coherently attribute goodness to God then fail. These arguments rely on the conflation of ontological and epistemological dependence I mentioned above.

Arbitrariness Objection
The second argument Loftus appeals to is the claim that a divine command theory makes God’s commands arbitrary. Again, he offers two versions of this.

No Reasons for Commanding
Loftus claims that a divine command theory means that God cannot have reasons for his commands and hence, they are arbitrary,

If we say, on the one hand, that something is right because God commands it, then the only reason why we should do something is that God commands it. It makes God’s commands arbitrary, because there is no reason why God commanded something other than the fact that he did.

This argument is again based on a failure to distinguish epistemic and ontological questions. Loftus argues,

[1] According to a divine command theory the only reason an action is wrong is because God commands it.

He infers from this that,

[2] Therefore there can be no reason why God issued the commands he did; hence,
[3] God’s commands are arbitrary.

This argument equivocates on the word “reason”. William Wrainwright notes that the word reason can be used in two different senses; the “constitutive” sense, one affirms that the reason water has certain phenomenological properties is because it is H20. This use of the word “reason” denotes a special kind of ontological relationship. The second sense is a “motivational” reason, as in when I state that the reason I feed my daughter is because I love her. This sense is more psychological or epistemological.[19]

It is important to note that these two senses are not the same. Suppose my son pours water into a glass. If we ask what the constitutive reason is for him to pour water into the glass, the answer is  that he poured water into the glass because he poured H20 into the glass. This was not the motivational reason for him pouring water into the glass; the  motivational reason will be that he wanted a drink. His wanting a drink does not constitute water and water being H20 is not the motivational reason he wants the drink.

Turning to Loftus’ argument, when Loftus states in [2] that God has no reason for commanding as he does, he clearly means that he has no motivating reason; this is why he can draw the conclusion that God’s commands are arbitrary. If God has motivational reasons, such as concern for the welfare of others for issuing the commands he does then God’s commands are not arbitrary.

The problem is that [2] follows from [1] only if one interprets [1] as affirming that a divine command theory contends that the only motivating reason an action is wrong is because God commands it. But this is false. As noted, when divine command theorists claim that the only reason an action is wrong is because God commands it, they have in mind constitutive reasons. The only constitutive reason an action is wrong is that God prohibits it; it does not claim anything about motivating reasons. Hence, Loftus’ argument commits the equivocation fallacy.

Abhorrent Commands Objection
The second version of the arbitrariness objection Loftus cites is the claim that a divine command theory entails that abhorrently evil acts could be morally required.

God could’ve commanded something else, or even something contrary, or something horribly evil and simply declared it good. If God is the creator of morality like he’s purportedly the creator of the universe, then he could have simply declared any act good, and there would be no moral reason to distinguish such a God from the Devil. This presents us with the “seemingly absurd position that even the greatest atrocities might be not only acceptable but morally required if God were to command them.

Turning to the second version, Loftus proposes two premises;

[1]If a divine command theory is true then it follows that if God commanded something abhorrently evil it would be required.
[2] It is possible for God to command something that is abhorrently evil.

From this it follows that it is possible for abhorrently evil actions to be required.

I agree with [1]; it is an implication of divine command theory that if God commands an abhorrently evil action then the action is required. This is not as significant as it seems for two reasons. First, any meta-ethical theory, whether secular or religious, that contends that some property P is equivalent to moral obligation will entail an analogous conditional; it will be the case that if an abhorrently evil action has P, the abhorrent action will be required. Hence, [1] by itself does not render a divine command theory implausible unless any and every account of obligation is implausible.

Second, according to the standard account of counterfactual conditionals, a conditional, such as “if God commanded something abhorrently evil it would be required,” is true if the antecedent is impossible. If the antecedent is impossible, the conditional is on par with claims such as “if there were a round square, its area would equal the square of one of its sides”[20] or “if there were no prime numbers, all numbers would have been Prime.”[21] The real issue then is [2]; is it possible for God to command abhorrently evil actions?

The problem is that [2] is false. Divine command theorists contend that an action is right, if and only if, God commands it. God here is used to refer to a personal being that is omnipotent, omniscient and perfectly good. Hence, for [2] to be possible, there needs to be a possible world in which a perfectly good being would command an action which is gratuitously abhorrently evil. This is an incoherent notion.

I think Loftus does have a response to this; it can be seen in his reply to essentially the same line of argument proposed by Robert Adams,

If Adams wants to claim that it is goodness plus God’s command that determines what is right,” Pojman rightly asked, “what does God add to rightness that is not there simply with goodness…If love or goodness prescribes act A, what does A gain by being commanded by God? Materially, nothing at all.” It is at this point where both a modified divine command ethic and a secular ethic share the exact same grounding.

Loftus here asks, if the commands of a good and loving God constitute right and wrong, why can’t a secular ethicist simply appeal to goodness and love to ground ethics and bypass an appeal to divine commands? What do divine commands add to love and goodness that is not “simply there with goodness?” Loftus seems to think Adams is refuted by the mere raising of this question.

This is rather puzzling because in an article Loftus cites, Adams himself raises this question and answers it. Adams notes,

It may be objected that the advantages of the divine command theory can be obtained without an entailment of God’s existence. For the rightness of an action might be said to consist in the fact that the action would agree with the commands of a loving God if one existed, or does so agree if a loving God exists. This modification transforms the divine command theory into a nonnaturalistic form of the ideal observer theory of the nature of right and wrong. It has the advantage of identifying rightness and wrongness with properties that actions could have even if God does not exist. And of course it takes away the basis of my metaethical argument for theism.

The flaw in this theory is that it is difficult to see what is supposed to be the force of the counterfactual conditional that is centrally involved in it. If there is no loving God, what makes it the case if there were one, he would command this rather than that? Without an answer to this question, the crucial counterfactual lacks a clear sense I can see only two possible answers: either that what any possible loving God would command is logically determined by the concept of a loving God, or that it is determined by a causal law. Neither answer seems likely to work without depriving the theory of some part of the advantages of divine command metaethics.

No doubt some conclusions about what he would not command follow logically or analytically from the concept of a loving God. He would not command us to practice cruelty for its own sake, for example. But in some cases, at least, in which we believe the act is wrong, it seems only contingent that a loving God does or would frown on increasing the happiness of other people by the painless and undetected killing of a person who wants to live but will almost certainly not live happily.6 Very diverse preferences about what things are to be treated as personal rights seem compatible with love and certainly with deity.[22]

Adams’ point is that “the essential features of a loving and omniscient creator would seem to under-determine the creator’s preference.”[23] God is not required, by the facts of the world or by his own nature, to issue the exact commands he does. Of course there are some things a loving good God will not command, such as cruelty for it own sake but once these options are eliminated, there is still no one set of commands that a loving being in a world like ours must command. There are possible worlds like ours where a loving being would issue one set of commands and there are other possible worlds like ours in all other respects in which a loving being would issues other commands. Hence, if we ask what a perfectly good and loving being would endorse, no correct answer can be ascertained. As such, one cannot ground moral obligations in the notion of lovingness or goodness alone. One needs to ground these things in the commands of an actually existing, loving and omniscient being.

Conclusion
My conclusion, then, is that Loftus’ critique fails. Loftus mistakenly construes defenders of divine command theories as claiming an epistemological dependence of goodness upon belief in God. It, in fact, claims that right and wrong ontologically depends upon God’s commands. Loftus’ major objections fail as a result. He fails to show that the idea of a good God is incompatible with a divine command theory and he fails to show that God’s commands are arbitrary. Finally his critique of Adams consists merely of a rhetorical question and ignores the responses actually made to this question by Thomas Carson and Robert Adams himself.

[19] William Wrainwright Religion and Morality.
[20] Craig “This Most Gruesome of Guests” 172.
[21] Plantinga “Naturalism, Theism, Obligation and Supervenience” 32.
[22] Robert Adams “Moral Arguments for Theistic Belief” 3 http://philosophy.ucsd.edu/faculty/rarneson/Courses/ADAMS1phil1reading.pdf.
[23] Thomas Carson Value and The Good Life (Indiana: Notre Dame University Press, 2000) 243. Carson defends the idea of under determination in more detail in his The Status of Morality (Dordrect: Reidel, 1984).

RELATED POSTS:
John W. Loftus on The Christian Illusion of Moral Superiority Part I

Tags:   · · · · · · 5 Comments

Leave a Comment


5 responses so far ↓

  • Hey Matt. You said that,

    “if we ask what a perfectly good and loving being _would_ endorse, no correct answer can be ascertained. As such, one cannot ground moral obligations in the notion of lovingness or goodness alone. One needs to ground these things in the commands of an actually existing, loving and omniscient being.”

    Ignoring other problems with naturalistic moral theories for now, could an atheist not take the route of arguing from what a perfectly good and loving being _could_ do? Ie, rather than attempt to appeal to (allegedly) counterfactual actions or decrees, rather appeal to counterfactual possible actions or decrees? It seems to me like this is a high price to pay, since it leaves the atheist open to admitting that a lot of immoral behavior must be permissible under his theory—but then atheists are often willing to make concessions like that anyway.

    Regards,
    Bnonn

    Recent blog post: On the atonement, part 6: universal atonement fails to actually accomplish redemption for anyone

  • Hi Bnonn,
    I think its not enough for the naturalist to contend that its possible for moral properties to exist if naturalism is true. The real questions faced are two fold (a) whether on naturalism the claim that such properties exist is more likely or plausible than the claim that they do not. And (b) whether a given naturalistic account is explains our moral experience as well as or better than a theistic account does.

    Now re (b) as you note that a person who took the line you suggest leaves the atheist open to admitting that a lot of immoral behaviour must be permissible under his theory a divine command theory can explain the nature and existence of moral properties without “admitting a lot of immoral behaviour must be permissible” this would mean that the divine command theory provides a better explanation of our pre-theoretical understandings of morality than this particular naturalistic theory does.

    This exasperated by the fact that the existence of God also provides an explanation for various other phenomena such as the origin of the universe, its continued existence, apparent fine tuning etc. So theistic accounts seem to have more explain the phenomena better than the naturalistic one you suggest does, theist accounts also bring in an increase in explanatory power that the purported naturalist one does not.

    Re (a) I am inclined to think the position you mention faces some problems here. Take (i) the hypothesis that moral properties do not exist, but because belief in that certain practises are wrong increases the chances of adaptive behaviour humans have evolved with certain false moral beliefs. And compare this with (ii) the hypothesis that moral properties do exist (they facts about counterfactuals of the sort you mention), because belief in that certain practises are wrong increases the chances of adaptive behaviour humans have evolved with certain moral beliefs, and these beliefs happen to coincide with those principles a perfectly informed loving being could endorse under counter factual situations.

    Its not at all clear to me that (ii) is more probable on the conjunction of Naturalism and evolutionary theory than (i) is. (i) is clearly more economical and it seems quite probable given naturalism and evolutionary theory.Natural selection selects for adaptive behavour not true belief. (ii) on the other hand is less economical, moreover it seems less probable that of the many basic evaluative attitudes we could have evolved that would be conducive to survival we happened to evolve those which coincided with counter factual about what a perfectly loving being could endorse if it existed.

    Recent blog post: Christian Blog Ranking Report for April 09 – HalfDone

  • Well I learned something new.

    If I could ask a question though

    “Divine command theorists contend that an action is right, if and only if, God commands it. God here is used to refer to a personal being that is omnipotent, omniscient and perfectly good.”

    How is ‘perfectly good’ defined ?

    Also this seems to refer to a Generic version of a God rather than simply the specific Christian God or am I missing something ?

  • Paul, I wouldn’t offer a definition of what the word “good means” but in the context of a DCT when people say God is perfectly good, they mean God has certain character traits, such as being loving, being just, hating certain types of actions which are paradigmatic examples of wrong actions, and so on. These are descriptive characteristics.

    As to a “generic God” vs the Christian God, I am not sure these are necessarily a different God. But in the case of a DCT what is needed is a person ( capable of issuing commands) who is omniscient and is good in the above sense. Arguments from the moral order for a DCT, also suggest this being must providentially govern the world so that virtue and happiness are ultimately united.

  • Thanks Matt, I’m still trying to get my head around DCT without having to read 10 tomes on the subject.

    Any chance you could post the Powerpoint from your talk on the subject (the viewing angle of the video is not good) ?