A recent debate gave me the opportunity to write a post I’ve been wanting to write for a long time.
Sean McDowell expounds the William Lane Craig version of the moral argument, which goes like this:
- If God does not exist then objective moral values do not exist.
- Objective moral values do exist.
- Therefore, God exists.
Like Craig, McDowell offers no arguments in favor of premise (2), but instead pumps the emotions of the audience with tales of Nazis and rape. However, because atheists almost always miss the point of the moral argument, McDowell begins with three clarifications:
- Atheists can be moral.
- Atheists can know morality.
- Believers can do bad things, and do them in the name of God.
McDowell’s central claim, then, is this:
If God does not exist, we do not have a solid foundation for [objective] moral values. But if God does exist, we do have a solid foundation for [objective] moral values.
His opponent, James Corbett, responds with many doubts about McDowell’s certainty in God’s commands – do we pretend that Sean McDowell knows God’s will better than the celebrated theologians of ages past who tortured unbelievers or held slaves? This is a good point, but it is irrelevant to McDowell”s argument.
Corbett also cites Gregory Paul’s 2005 paper showing a correlation between secularism and societal health. He then says that “If the Christian God is necessary for morality, it must follow that belief should correlate with positive social behavior.” But I see no reason to accept this.
Point after point, Corbett fails to engage McDowell’s argument. Instead, he rants against the straw men that McDowell clearly labeled as straw men in his opening statement. His performance is summed up in his own words quite nicely:
Is there another explanation for morality? I’m sure there are many. And the truth is, I’m not an expert in religion or morality.
Well, then. As long as you, a non-expert, are “sure”… I guess that settles it!
When will atheists stop embarrassing themselves in debate? This shows the problem with atheists believing they are, by default, more rational than believers. Atheists don’t think they need to study the relevant subjects, or pay attention to the logic of the Christian’s position. Instead, they just wander in and spout some irrelevant points about the Crusades and religious disagreement. Meanwhile, the Christian can put forth whatever argument he wants – whether it’s a good argument or not – because the Christian will clearly explain why the atheist’s arguments fail, but the atheist will not clearly explain why the Christian position fails. Thus the audience leaves believing the Christian has won. And basically, he has.
McDowell responds to Corbett’s irrelevance kindly but firmly, pointing out quite clearly why each of Corbett’s points were red herrings and did not engage the stated topic of the debate.
Corbett once again accuses McDowell of logical fallacies, even though McDowell has not committed any. He then repeats his red herrings from his opening speech. Hoo-boy.
McDowell responds with humor, and points out that he has nothing to respond to, because Corbett hasn’t addressed the topic of the debate. Corbett completely misunderstands that the debate concerns the ontology of moral values – what philosophers call the “sources of normativity.” Instead Corbett thinks the debate is about why we do things that we believe to be moral, even though McDowell clearly explained the difference in his opening speech, and throughout his rebuttals. Argh!
Then there’s some cross-examination and time for questions, but suffice it to say that Corbett keeps digging his own grave, and McDowell merely hands him the shovel.1
Because Corbett never addressed McDowell’s argument or provided a foundation for moral values without God, I will do so here and now.
Let us remind ourselves of McDowell’s central claim:
If God does not exist, we do not have a solid foundation for moral values. But if God does exist, we do have a solid foundation for moral values… God is the best explanation for objective moral values.
Now what does he mean by “objective”? McDowell says, for example, that
[Rape] is wrong independently of whether anybody believes it to be [wrong] or not.
William Lane Craig offers this illustration of McDowell’s definition:
To say, for example, that [the Holocaust] was wrong [is to say that] it would have been wrong even if the Nazis had won World War II and succeeded in exterminating or brain-washing everyone who disagreed with them.2
McDowell also proposes “three criteria that any adequate moral system must be able to account for”:
- Any adequate moral system must have a transcendent standard beyond human nature.
- Any adequate moral system must account for free will.
- Any adequate moral system must account for what makes humans special.
It would be rather startling to all the philosophers who have done ethics without God for centuries if McDowell walked into a room and announced that without God there could be no objective morality. They might look at each other and exclaim, “What did he just say? Did he just say the universe can’t have moral value without a cosmic overlord lurking in the background? Is he serious?”
Of course, there are many theories of objective moral value, and none of them invoke a deity. (Theistic morality is a form of ethical subjectivism; I’ll get to that in a moment.)
So before I respond to McDowell’s argument, let me sketch just one plausible theory of naturalistic, objective moral realism. I call it “desirism.”
I cannot sketch desirism in one blog post. Nor could I sketch it in 30 pages. Desirism interacts with thousands of pages of contemporary meta-ethics, and cannot be boiled down easily. One reason that divine command ethics is so attractive is that it’s easily understood – “God said it, and that settles it” – even if it’s implications are unclear (Slavery or abolition? Genocide or peace? Distribute resources to the poor like Jesus or build wealth through capitalism?)
Morality is about reasons for action. Reasons for action to feed the poor. Reasons for action to promote the truth. Reasons for action to not rape.
Desirism holds that, as it turns out, the only reasons for action that exist are desires. Other proposed reasons for action do not, in fact, exist. Categorical imperatives, hypothetical social contracts, divine commands, intrinsic value – these things do not exist. However, desires do exist, and they are reasons for action. There is only a reason for action to feed starving people because it would fulfill their desires for food. If people did not desire food but instead desired sunlight, there would instead be reasons for action to give them access to sunlight.
But individual desires are not moral reasons for action in isolation. My desire to successfully rob a bank is a subjective reason for action for me to bring a gun to the robbery, because it will help me fulfill my desire to rob a bank. (This is usually called ‘practical value’ or ‘prudential value’, not ‘moral value’.)
But morality is concerned with reasons for action as a whole. This would include the desires of the people in the bank to not get shot, the desires of those with a financial stake in the bank to not be robbed, and the desires of people in the community to live in a peaceful and safe society. When accounting for all reasons for action that exist – when accounting for objective moral value – it turns out that my robbing the bank would be objectively wrong.
But don’t confuse desirism with act utilitarianism or rule utilitarianism. I’m not saying that the right act is the one that will fulfill the most desires. Desirism does not claim that acts are the primary objects of moral evaluation. Rather, it says that the reasons for action themselves – the desires - are the primary objects of moral evaluation. So a good desire is one that tends to fulfill more and stronger desires than it thwarts, and a right act is one that a person with good desires would perform.
Now, reading such a brief summary, many objections will immediately come to your mind. This is because I do not have the space to properly spell out the theory. But for the sake of argument let us examine this question: If desirism is coherent, does it engage Sean McDowell’s argument?
First, note that desirism does offer an account of objective moral value as McDowell defines it. According to desirism, racist slaughter and oppression would still be wrong even if the Nazis had killed or brainwashed anybody who disagreed with them. Even if we all believed racist slaughter and oppression was morally good, they would still be morally wrong according to desirism, because they are actions that come from evil desires – desires that tend to thwart more and stronger desires than they fulfill.
Next, does desirism provide a transcendent standard of moral value?
What does McDowell mean by “transcendent”? If McDowell means just “supernatural”, as he seems to intend later in the debate, then he begs the question against the naturalist. But in intellectual charity I will assume he intends the meaning he gives at the outset of the debate: a moral standard is “transcendent” if it lies above and beyond human nature.
Desirism places the standard of morality in a universal consideration of reasons for action that exist, and this is a standard above and beyond human nature. Why? Because humans are not the only beings with reasons for action. Many other animals have desires, too, though perhaps to a lesser extent. Probably, they have strong desires according to the complexity and sophistication of their neurological structures. A chimpanzee has many desires, but a butterfly probably does not. So desirism’s standard of morality lies above and beyond human nature, and is therefore transcendent according to McDowell’s definition.
By the way, I agree with McDowell that evolution cannot provide a basis for objective moral value. First, because of reasons summed up by Sharon Street. Second, because evolutionary ethics faces its own Euthyphro dilemma. But desirism certainly does not say that whatever we evolved to do is therefore moral.
Next, must an adequate moral system account for contra-causal free will? That is, must an adequate moral system account for free choices that are not determined by the prior state of the universe?
McDowell cites, of all sources, Dictionary.com:
One of the definitions for ‘moral’ at Dictionary.com is: “capable of conforming to right rules of conduct.”
But this is fully consistent with determinism. Even if our choices are determined, we are still capable of conforming to right rules of conduct. This happens all the time. We make conscious choices to do things that are right or wrong. Whether the prior state of the universe or God’s foreknowledge or the will of a supernatural spirit had anything to do with these choices is irrelevant. Moreover, some have argued that contra-causal free will is incoherent – that is, logically impossible. And, scientific evidence is piling up to show that, whether or not God exists, free will does not exist.
More importantly, contra-causal free will is not needed for morality. You might ask: “Without free will, why should we praise someone for doing what is right, or condemn them for doing what is wrong?” The answer is: “Because that’s the whole point of morality. When you praise people for doing good, you affect people’s desires. You promote good desires – desires that tend to fulfill other desires (including your own). And when you condemn people for doing evil, you diminish evil desires – desires that tend to thwart other desires.”
This is quite obvious in the case of, for example, slavery. Slavery was banished from much of the world – after thousands of years as a standard practice – within a few generations! How? It wasn’t done just by enacting laws prohibiting slavery. It’s not like we all walk around today thinking “I wish I could have slaves, it’s too bad about those damned laws that prohibit it.” No, our desires have changed. They changed because people like Martin Luther King Jr. praised brave abolitionists and condemned racist bigots.
Finally, I would remind McDowell that vast populations of Christians and other theists have rejected the notion of free will and still affirmed humanity’s moral responsibility. Indeed, it’s hard to see in what sense we would have free will if God foreknows everything we will ever do.
So contra-causal free will is not required for objective morality.
Are humans special?
It’s unclear to me why an “adquate moral theory” must conclude that humans are special. And of course, humans are special in a number of ways. We have the biggest brains of any species. We dominate the world like no other. We can run longer distances than any other species.
More relevant to morality, we are the only species capable of consciously making moral decisions. So humans are special in that way, and desirism acknowledges this.
Desirism also acknowledges that human desires may have more moral significance than animal desires, because our desire factories (our brains) are more advanced than animal desire factories, and therefore seem to produce more and stronger desires.
Moreover, desirism seems to have a much better account for the moral significance of animals than Christians ethics does. According to Christian ethics, slaughtering dolphins or torturing livestock is not wrong because of the harm it does to dolphins and livestock, but perhaps because God commanded it or, as Aquinas argued, because it harms the humans involved by making them feel less averse to harming the animals that matter: humans. In contrast, desirism says that slaughtering dolphins and torturing livestock is wrong precisely because of the harm it does to dolphins and livestock.
Can God Ground Moral Values?
So I think I’ve offered a robust, plausible account of objective moral value without God. Now, I would like to argue that God cannot ground objective moral values.
Many religious people assume that God is a plausible foundation for moral values, and they do not understand why most moral philosophers have been doing ethics without God for over 300 years. (This includes famous theistic philosophers like Immanuel Kant and Henry Sidgwick.) Indeed, most philosophers living today are moral realists, and most are atheists. Let me explain some of the reasons why this is.
First, the Euthyphro dilemma. If things are good because God says they are good, then morality is completely arbitrary. Whatever God commanded would be morally right. So if he commanded genocide and mass rape – as he is reported to do in Number 31:7-18 – then genocide and mass rape would be morally right. But this evacuates the term ‘moral’ of any significance. ‘Moral’ just means ‘Whatever the boss likes.’ To call God morally good is just to say that he is God. It doesn’t say anything meaningful about his actions, because God would be ‘good’ no matter what he did.
On this point, Christian philosopher Gottfried Leibniz is most eloquent:
In saying… that things are not good according to any standard of goodness, but simply by the will of God, it seems to me that one destroys, without realizing it, all the love of God and all his glory; for why praise him for what he has done, if he would be equally praiseworthy in doing the contrary? Where will be his justice and his wisdom if he has only a certain despotic power, if arbitrary will takes the place of reasonableness, and if in accord with the definition of tyrants, justice consists in that which is pleasing to the most powerful?4
But if things are good for some other reason, then God is at most a messenger of moral duties, and at worst may be a selfish deceiver about genuine moral values. After all, he does supposedly command that we worship and obey him for all eternity – a highly suspicious ‘moral’ command. In any case, if things are good for a reason besides God’s will, then the ground of morality lies beyond God.
The usual response to this dilemma is to say that morality is grounded in God’s nature, which is merely expressed in his commands. But this hardly avoids the problem, for it would still be the case that whatever it was God’s nature to prefer would be right by definition. If it was in God’s nature to prefer genocide and mass rape, then it would be morally right to commit genocide and mass rape. And this would still diminish the significance of moral terms. To say call God ‘good’ would just mean that God accords to his own nature – which is hardly a praiseworthy accomplishment. God would still be good no matter what his nature was – even if his nature was to command genocide, eternally torture people who had never heard of him, and demand everlasting adherence to his arbitrary decrees.
Why else do most moral philosophers reject theism as a ground of objective moral facts? Though theistic ethics is ‘objective’ in the sense McDowell defines the term – moral truth transcends human moral beliefs – it is not objective in another common sense of the term. Theistic ethics falls under the usual heading of ethical subjectivism: the view that moral statements are made true or false by the attitudes of certain people – a particular person of culture (group of people). So unless McDowell wants to say that God is not a person, his theory of morality is really a theory of ethical subjectivism. In contrast, desirism defends a standard of moral value that exists independently of any person’s or people’s attitudes.
Can God Explain Morality?
Finally, let us remind ourselves of the topic of the debate: Is God the Best Explanation of Moral Values?
But in what sense is “God did it” the best explanation for moral values? Does McDowell propose a Bayesian defense of “God did it” as a good explanation for objective moral value? If so, I’d like to see it.
Or perhaps McDowell intends to offer God as an explanation in a more traditional way, by saying that the God hypothesis possesses certain explanatory virtues as a potential explanation for the existence of objective moral values. In the same way that Einstein’s theory of relativity possesses the explanatory virtues of testability, predictive novelty, explanatory scope, consistency with background knowledge, simplicity, and so on with regard to the diverse phenomena it is invoked to explain, perhaps McDowell thinks that the God hypothesis possesses some or all of these explanatory virtues when posited as an explanation for the existence of objective moral values.
But I do not see how this is the case.
First, is the God hypothesis testable? I’m not sure how. What specific, measurable predictions does the God hypothesis make that we could test? Theologians usually define God such that he is consistent with all data, rendering him untestable. If McDowell thinks the God hypothesis is testable, he must explain why he thinks the God hypothesis makes certain predictions about the world instead of others, and in fact these predictions should turn out to be true if they are to lend support to his explanatory hypothesis.
Second, does the God hypothesis offer any predictive novelty? Again, I’m not sure what. Theologians have used the God hypothesis to predict all sorts of things about the world which science eventually revealed to be false. I’ve never seen the God hypothesis render some precise and surprising new prediction which later turned out to be true – as was the case in many ways with Einstein’s theory of relativity.
Third, does the God hypothesis offer much explanatory scope? This would mean that the God hypothesis accounts not just for some of the data but for all of the data. To illustrate, this is often a problem with conspiracy theories. For example, conspiracy theories about contrails – the visible exhaust of compressed vapor left behind flying jets – claim that they represent a secret government plan to dump harmful (perhaps mind-controlling) chemicals on people. But this theory does not explain why contrails appear just as often over uninhabited sea and land as over inhabited sea and land, along with many other things. The correct explanation – which has to do with jet exhaust and condensation – does account for these facts.
So does the God hypothesis offer good explanatory scope? Again, because the God hypothesis is offered so vaguely that it could explain any state of affairs, it therefore explains nothing.
Does the God hypothesis exemplify consistency with background knowledge? It doesn’t seem so. The God hypothesis posits something timeless (!!??), spaceless (!!??), all-knowing (!!??), all-powerful (!!??), all-good (!!??) and so on. In short, God is by definition exactly contrary to everything we know and experience in all the usual ways.
Is the God hypothesis simple? Christian philosopher C. Stephen Layman identifies four facets of simplicity:5
- The number of things postulated. Leverrier posited one planet to explain the irregular orbit of Uranus: not two or three or four objects. (The posited planet was later discovered: Neptune.)
- The number of kinds of things postulated. A theory that posulates four kinds of subatomic particle to explain phenomena is simpler than a theory that has to postulate 50 different kinds of subatomic particle.
- The simplicity of the terms. A term is complex to the extent that it can be understood only by someone who understands other terms. The term ‘ball’ is defined by reference to very few other terms, whereas the term ‘quark’ is defined by reference to a great number of other terms.
- The number of statements that receive little or no support from other statements in the hypothesis. Basically, the more things you say, the more likely you are to say something false.
Is the God hypothesis simple in these ways – relative to naturalism? (Desirism is a naturalistic theory of ethics.)
No. The God hypothesis is more complex, on all four fronts. First, both naturalists and theists posit the existence of all the usual things. But theists must posit an additional thing – namely, God.
Second, theists must posit more kinds of things posulated. They must posit both natural and supernatural things. As already stated, they must posit a timeless (!!??), spaceless (!!??), all-knowing (!!??), all-powerful (!!??), all-good (!!??) being. Now that is quite extraordinary.
Third, naturalism is simpler with regard to the simplicity of terms. Naturalism explains things in terms of other things we already understand – other natural things for which we already have good evidence. In contrast, theism employs terms like ‘omnipotent’ and ‘perfectly good’ and ‘omniscient’ that are so difficult to make sense of that theists have written entire books trying to explain what these terms could even mean.
Fourth, naturalism is also simpler with regard to the number of statements. Naturalism basically just says that all entities are physical entities. Theism claims something like:
Theism: (1) There is exactly one entity that is (2) perfectly morally good and (3) almighty and that (4) exists of necessity.6
Thus, Layman concludes: “I agree that Naturalism… is simpler than Theism.”7
So if Layman wants to argue that God is the best explanation for the existence of objective moral values, he certainly has his work cut out for him.
Desirism and explanation
How does desirism fare with regard to these common explanatory virtues? Much better than theism, that’s for sure. First, desirism makes a large number of specific predictions about relations between desires and states of affairs, all of which are falsifiable. So far, they have not been falsified.
What about predictive novelty? It depends how you define predictive novelty, but for now I’m willing to concede that desirism may not offer any predictive novelty.
Does desirism have good explanatory scope? I think so. Desirism is grounded in the most well-attested theory of intentional action: the desire-belief model of intentional action, which has been successfully used to explain a wide range of human behavior.
Is desirism consistent with our background knowledge? Yes. It need not posit anything unfamiliar to us, but only beliefs and desires and certain common-sense relations between them.
Is desirism simple? Though I can’t explain the theory in just a few words, it is certainly simpler than theism by virtue of being a naturalistic theory of ethics. It doesn’t postulate a God in addition to the natural world – it only postulates the natural world. And it refers to commonly understood and accepted terms like beliefs and desires.
Once again, an atheist has embarrassed himself in an awful debate performance against a theist. Christian apologist Sean McDowell – perhaps the next William Lane Craig – presented a clear argument, and his atheist opponent did not attack the argument or present his own argument for a non-theistic explanation of the existence of objective moral values.
So I have provided an objective theory of morality that has greater explanatory virtue than McDowell’s theory of theistic ethics. I have also explained some serious problems with theistic ethics that I do not think McDowell will be able to overcome.
- I should note that during the cross-examination, Corbett did no better but McDowell did worse. He begins to expound the off-topic evolutionary argument against naturalism, for example. [↩]
- From Craig’s debate with Sinnott-Armstrong, page 17. [↩]
- This summary uses certain terms somewhat differently than I use them elsewhere, which will confuse people who are already familiar with desirism. I apologize. This is my attempt to summarize things in a way that may be more clear on a surface level. If we discuss desirism at a more detailed level, then we can be more precise with the meanings of our terms. [↩]
- Discourse on Metaphysics II, trans. by George R. Montgomery. [↩]
- Letters to Doubting Thomas, pages 23-24. [↩]
- Ibid, page 12. [↩]
- Ibid, page 29. Actually, Layman’s character Zach says this, but it’s clear that Zach is the character speaking from Layman’s position. [↩]
Previous post: Software for Philosophers