About Me

My photo
Australian philosopher, literary critic, legal scholar, and professional writer. Based in Newcastle, NSW. My latest books are THE TYRANNY OF OPINION: CONFORMITY AND THE FUTURE OF LIBERALISM (2019); AT THE DAWN OF A GREAT TRANSITION: THE QUESTION OF RADICAL ENHANCEMENT (2021); and HOW WE BECAME POST-LIBERAL: THE RISE AND FALL OF TOLERATION (2024).

Saturday, April 09, 2011

William Lane Craig vs Sam Harris

As some of y'all know, William Lane Craig debated Sam Harris in the last day or two on a topic relating to the source of morality - the precise proposition that was to be debated is still not clear to me even after listening to it, though it was doubtless stated at some point in the longish introduction. It was something to the effect of, "Are the foundations of morality best explained as natural or supernatural?" though it seemed, in practice, to be more like, "Given that morality is objective, can it come only from God?"

Damion has a review of it here ... and there is another review here (this one by John Loftus). I can't judge who would be seen as having won the debate from the point of view of a reasonable person actually in the audience, as you lose a great deal from having only an audio recording. From my viewpoint, merely listening to the audio, both speakers were fluent, sounded confident, and probably put the strongest arguments they reasonably could have in the timeframe available.

All in all, I'd score it as a draw. No knock-out blows were delivered. Neither speaker struggled at any point. Either speaker could have sounded plausible to someone predisposed to be sympathetic.

Sam Harris conspicuously failed to engage with one of Craig's points - that he (Harris) rejects both libertarian and compatibilist free will, so one might ask whether this must force him into something like an error theorist position. After all, if you reject even compatibilist free will, is there any sense in which any of us can ever act otherwise than as we do? And if not, doesn't this remove the presupposition that we are responsible for our actions And is that presupposition, Craig can ask, not inherent in ordinary moral discourse?

It's a bit surprising that Harris didn't address this, but perhaps he thought it best to avoid opening up a can of worms relating to free will and determinism ... and he was arguably within his rights to ignore it because atheism does not directly entail that there is no free will. At worst, Harris might be forced to adopt at least some deflationary concept of free will and/or moral responsibility, and he actually does this to a point in The Moral Landscape - he gives an account of how we can be held morally responsible for our actions in many or most circumstances. Perhaps it was better to do what Harris actually did, and let Craig waste time on this issue while concentrating on making a positive case for a naturalised objective morality.

Harris failed to attack Craig strongly where Craig was weakest - in showing just how God delivers an objective morality. Craig kept saying saying that he was not, for example, making a moral semantic claim that "morally good" means "in conformity with God's commands" ... but this left it rather confusing just what he does think it means. He kept insisting that he was making an ontological claim about the source of morality, not a moral semantic claim, but the actual content of the ontological claim ended up being very vague. To someone who was not already sympathetic, it might have sounded like no more than rhetorical gobbledegook.

Which leaves a question as to whether Harris was tactically correct to let it go save for the briefest mention. Enough was said to at least raise the point, while leaving the detailed criticism unargued - anyone who was sympathetic to Craig on this would, perhaps, have been beyond the reach of any more elaborate argument that Harris could have put. Conversely, it's doubtful that Craig's woolly explanation convinced anyone who was initially sceptical. That left Craig's main argument rather inconclusive, to say the least.

Still, someone scoring the debate formally might have deducted points from Harris for failing to engage more over this.

Let's take this slightly further than Harris did. It seems clear enough that commands from a powerful being such as God could have something like the status of weak categorical imperatives, like the norms of law, etiquette, professional practice, and positive morality. The point is: all of these can be ignored by someone who can get away with ignoring them and who doesn't care about their authority. "What is etiquette to me?"

The commands of a powerful and omniscient being such as God could not be ignored in practice - if we ignored them God could punish us. But that gives us only a subjective reason to obey, based on fear. If someone says, "What is that to me?" there will be an answer, but it will appeal to the person's desire not to be harmed. In that sense, it's still subjective.

Craig needs a reason to obey that is not based on such things as desires or fears. He might, of course, have it if the commands (or their content) were good in an objective sense even prior to being issued by God. But that's mysterious, and besides it's a line that Craig can't take. If he admits that that scenario is even possible, he is back to saying that God's issuing of commands is not the ultimate source of objective morality after all.

Perhaps he would have said more about this conundrum if Harris had pressed him harder on the issue, but Harris never did. Again, leaving aside someone formally keeping score, that may have been a smart move on Harris's part, leaving Craig stuck with a claim that sounded weak and unconvincing to anyone who was even mildly sceptical, while allowing Harris time to develop his positive account. Perhaps how we score it depends on who Harris was really trying to persuade to his viewpoint.

Though he seemed rather weak on showing just how God's commands are objectively binding in any interesting and plausible sense, Craig was effective in attacking other possible sources of objectively binding moral authority. Conversely, I thought that Harris was very strong when it came to developing his positive case for objective morality based on purely naturalistic considerations. In the end, I was not convinced - surprise! surprise! - and Craig made the telling point that the argument seemed to rely on something like faith: Harris presupposes the controversial claim that we ought to do whatever maximises well-being. Craig has a point when he says that this is not the sort of thing that can simply be stipulated as an axiom.

Still, it's a point that many people (including some of my regular readers) would grant Harris ... and once you let him that far he is very cogent and eloquent in developing the implications.

In all, this was an impressive performance by both speakers with some tactical (I assume) decisions by Harris that may have cost him points in some people's eyes ... but may not have been bad choices, all things considered. I'd be interested to listen to, or preferably watch, another debate between these two.

Edit: Thanks to the commenter who pointed to a YouTube video.It looks like there are actually nine of them if you want to follow the whole debate. To be honest, I'm not likely to do this soon, having spent a couple of hours yesterday listening to the audio version ... though I'll try to get to it at some stage in the future.


Anonymous said...

A slip?

You wrote:

Sam Harris conspicuously failed to engage with one of Craig's points - that he (Harris) rejects both libertarian and compatibilist free will, so one might ask whether this must force him into something like an error theorist position.

After all, if you reject even compatibilist free will, is there any sense in which any of us can ever act otherwise than as we do?

And if not, doesn't this remove the presupposition that we are responsible for our actions And is that presupposition, Craig can ask, not inherent in ordinary moral discourse?

Looks like the only error you have in view is, given the truth of Harris’ hard determinism, the error that humans are any more morally responsible than goldfish or toasters.

But error theories are meta-ethical views about the semantics of moral claims.

Incompatibilism is not and never has been one of those.

setAD7 said...

Here is the video.

mathewvaridel87 said...

I think the proposition was, "is good from god?" If that's true, I really think Harris should argue the following:

1) Euthyphro dilemma/Craig's semantic word games.
2) The unnecessary suffering and scripture arguments that he gave.
3) Give an alternative for where morality comes from:
a) Evolution - highlight precursors in other species.
b) Ability to empathise and big brains allow us to reason our way through morality.

I'd just note that if the proposition was what I think it was, I think Harris won. Craig didn't make a convincing enough argument as to why god should be the guide for morality. Harris also showed why god's actions are not what we want to base our actions on through unnecessary suffering and scripture. Thus, Harris wins as the proposition leaves the burden of proof on Craig.

I might be a little bias though :p

Russell Blackford said...

Gaius, there could definitely be a moral error theory that says:

P1. The practice of making moral judgments assumes the truth of some form of free will.
P2. But no form of free will exists.
C. The practice of making moral judgments is pervaded by error.

Richard Joyce discusses various such possibilities in his books and articles on error theory.

It's true, though, that this is not among the most common kinds of moral error theory, which are more along the lines of:

P1. The practice of making moral judgments presupposes that there are desire-transcendent reasons for action.
P2. But there are no desire-transcendent reasons for action.
C. The practice of making moral judgments is pervaded by error.

Or, a closely related variant that you might prefer:

P. All moral judgments assert the existence of desire-transcendent reasons for action.
P2. There are no desire-transcendent reasons for action.
C. All moral judgments are false.

Obviously there are variants.

Joyce briefly discusses a variety of possible moral error theories, though the one usually discussed on this blog (and overwhelmingly in the literature) is the last one described above, and variants of it. (Mackie, for example, may have believed a weaker conclusion ... perhaps more along the lines of: "All moral judgments include cognitive content that is false." Mackie seems to have thought that moral judgments contain a mix of cognitive and non-cognitive content, and that some of the cognitive content might even be true. But he certainly thought that the practice of making moral judgments was pervaded by an error along the lines of a false belief in desire-transcendent reasons for action.)

It's still true that Craig was accusing Harris of holding to a proposition (there is no free will of any kind) that is counter to a presupposition that we need (according to Craig) to make genuine moral judgments (i.e. some kind of free will exists). Thus, Craig is saying that Harris is committed, without knowing it, to the belief that all our moral judgments presuppose something that is false.

Does that help explain what I was getting at?

Marshall said...

Harris certainly put up a better front than Krauss, who didn't appear to have done any preparation at all, not even buttoning up his shirt. One point about Craig is that he actually varied his presentation according to the respective topic. I also notice that he gives nods to evolution here and there. His main strength seems to me to be that he is able to make arguments in terms that are sensible to the "informed layman". Whatever his actual understanding of metaethics and mathematics, he is able to relate his points about them to the common understanding, which Krauss and Kazez conspicuously are not even interested in doing.

How would God "deliver an objective morality"? Craig asked Harris how secularism can deliver such, using Mackie's Argument from Queerness. Working off my reading backlog, I just noticed this:
"In recent years, China's own activists have identified freedom, democracy, human rights and human dignity as "universal values": this is one of the core ideas in Charter 08, the reform document the government has tried so hard to suppress. China's rulers have countered by claiming that 'so-called' universal values are merely "tactics peddled by the West." *
The point is that the concept of "human flourishing" has nothing to say about this practical secular problem of trading off collective security vs. individual freedom, and it would be a queer sort of physical fact that did.

Whereas God as conceived by Christians would deliver morality as part of the process of forming humans: "I will give them an undivided heart and put a new spirit in them; I will remove from them their heart of stone and give them a heart of flesh." -Ezekiel 11:19; see also eg Matthew 13:15 for Jesus and 2 Corinthians 3:3 for Paul. Sam asked, what if God said it's OK to mutilate small girls? Since our hearts are formed after God's heart, we can be confident that the abhorrence we feel is a reflection of his desires. Could God have created our hearts otherwise? He could have, but he didn't.

"Craig needs a reason to obey that is not based on such things a desires or fears." Craig and Harris have the same reason, the natural desires of the human heart. Craig has an understandable reason why the "desires of your heart" are what they are, I don't see that Harris does.

*New York Review of Books, March 24, 2011, p.21, "How China Fears the Middle East Revolutions" ... not online, apparently

Anonymous said...

RB, as you seem perhaps to have guessed, I was thinking of Mackie when I added my above comment.

Personally, I think that if he is right and the error is that moral claims attribute to actions, action-kinds, etc. special properties that do not actually exist then he should have insisted that moral terms don’t denote such properties and neither do they denote anything else.

The result is that, contrary to beliefs that are essential to morality, moral claims don’t express anything that could be true or false, don’t express reasons for or against action, and don’t express anything that could be even believed much less known.

For example, it is part of the moral delusion, so understood, to believe that “murder is wrong” is true, can be believed or known to be true, and affords a reason not to murder.

Also, Mackie seems to have held the non-existence of moral properties to be merely contingent, making their alleged non-existence no different from their non-exemplification.

That seems to have left open the questions why they are not exemplified as well as whether they might have been in the past or might be in the future.

And if he can argue against such properties by appeal to their queerness surely we can argue against this view that their non-existence is contingent by noting that it is very queer, indeed.

I think he would have done better to argue that, given the work they are supposed to do, it is a necessary truth no such properties exist and, so to speak, that that that is why no such properties are exemplified.

On the other hand, if the properties in question did exist then a crucial component of a version of moral realism fitting the common moral consciousness would also be true.

Add an innate human capacity to recognize such properties in actions (or maybe action kinds), fill in a bunch of further details, and you have a moral realist story that no more necessarily depends on God than anything Ewing or Ross could have thought of, though it would fit admirably well, overall, the traditional view of morals common to Western man right back through the Christian and into the pagan era.

And all that without succumbing for a moment to naturalism in meta-ethics.

By the way, just for the sake of full disclosure, I am not myself a moral realist on any understanding of that term with which I am familiar.

I think Mackie was fundamentally right about morality, though mistaken about crucial details; and that we do have values some of which pertain to actions, characters, and the rest of the subject matter of morality – and, for that matter, politics.

I think a true account of non-moral value claims understands them to be subjective and relative and a revisionist moral skeptic could certainly make use of moral language in a similar way to express his true values, though this would unavoidably involve a certain measure of perhaps justifiable, or at least excusable, dishonesty.

After all, words mean what they mean.

They don’t mean what we want them to mean just because we want them to.

Anonymous said...

By the way.

There is no doubt that Christianity has entrenched the notion of agent causation and the idea that libertarian free-will is necessary to moral responsibility, though both ideas are older than Christianity.

But not, I think, to the point of making such notions integral to the very meaning of moral language.

So I think there is no dishonesty in compatibilism.

And I also think compatibilism is in a better position to accept such ideas as that of diminished responsibility or irresistible compulsion than libertarianism is.

Everyone is what he is and not another thing.

Why then should we wish to be deceived?

Russell Blackford said...

And in fact, Harris isn't really far from being a compatibilist. He thinks that what most people mean by "free will" is libertarian free will. But he doesn't really deny that what compatibilists are describing exists. And what compatibilists are describing seems to me to be enough for moral responsibility (and you may be right that it does a better job of accounting for gradations of responsibility - that's an interesting point).

Sean (quantheory) said...

I would have appreciated it if Sam had pushed a little harder towards Euthyphro, although I think he expressed that he was actually glad that it hadn't come up, which isn't surprising given that a similar argument constantly gets leveraged against his own moral ideas. In any case, he would have had a decent case to make that wellbeing is fairly close to what many people think of as morality, whereas an entity that "deserves to be worshiped" (or is both a conscious entity and a sort of platonic ideal of the Good) doesn't really make much sense.

Thalamus said...

With regards to free will, why is everyone ignoring the GIGANTIC elephant in the room here?

The reconciliation of God's omniscience with the concept of Free Will. If God knows everything including the future, then we're not really free to choose what we do. Does God know what I'm having for dinner tomorrow? A theist would have to say of course "Yes he does, because God knows it all. God knows you will eat Y for dinner". But what if tomorrow night I 'choose' to eat X? Then God will have been wrong and therefore God didn't really know my future.

Theists cannot believe in free will either!

Russell Blackford said...

I don't know what WLC thinks, but some theist philosophers actually do claim that God doesn't know the detail of the future. They might think God is powerful to intervene successfully whenever he wants to, but they think there's a whole lot of stuff he doesn't know precisely because he doesn't know what our free choices will be.

But that seems like a watered down concept of omniscience to me. It's a long way from the view of God as outside of time and seeing it all like seeing a panorama spread out beneath you from the top of a mountain - which was more how Aquinas thought about it.

Anonymous said...


What I had in mind is that compatibilism often accompanies the idea that if not the only then the chief purposes of punishment are (a) to discourage repeat offenses by the person punished and (b) to discourage like offenses by others.

The ideas of diminished capacity and compulsion enter into cases where the criminal’s ability to refrain from the criminal activity in order to avoid future punishment is inadequate.

In such cases punishment can be moderated or even waived in favor of some other alternative, depending on what the criminal behavior involved actually is and what society seems justified in doing to protect itself.

Tobacco addicts and kleptomaniacs who steal cheap items may be at one extreme while pederasts and serial killers are at the other.

That diminished capacity and irresistible impulse are even possible raises no special theoretical difficulties for compatibilists, committed as they already are to the idea that human behavior is subject to causes the individual ultimately does not control.

He can maintain that sometimes these causes operate through rational consideration and choice but sometimes they can subvert or even wholly overcome that normal process.

These options are not available to the libertarian.

Thalamus and RB,

RB has pointed out that some people deal with the apparent clash between divine foreknowledge and human freedom by denying that future contingents have truth values and accepting God does not actually know the future.

Others insist even the possibilities of guessing, surmising, or inferring the future and not just knowing it suppose there is a determinate future to be guessed, surmised, or inferred and hence that future contingents do have truth values.

And most insist that divine foreknowledge is, for theologians, an inescapable Biblical commitment.

Positions of theologians who agree that future contingents have truth values – and so there is a future to be known by God and guessed at by ourselves – vary considerably.

Many say that it is God who determines the future while human choices are nevertheless not subject to causation by events or other factors other than the agent, himself.

They see the determinate future as specifying not what you must do but what you will do; they rightly insist there is no logical inconsistency between the determinacy of what you will do and the claim that you could do otherwise.

Thus they insist the doctrine of a determinate future does not, in fact, conflict with the ideas of agent causation and libertarian freedom as developed in opposition to the idea that human choices are subject to causation that escapes our control and instead controls us.

But this view still leaves it true that God, limited only by logical possibility, writes the whole script of the world and that we write nothing so that our crimes and sins are more his fault than ours.

To circumvent this objection Alvin Plantinga famously – and I think incoherently and unsuccessfully – insisted that while God writes the script his writing is limited not only by logical possibility but by the “trans-world depravity” of mankind.

Our freedom escapes God in that while he can foresee, so to speak, in any given possible circumstance that we will freely do A though we could as well do B, he cannot choose what we will do.

So God can foreknow how things will go in the world he chose to create, but he cannot fully determine how they will go and, in particular, he can foresee but not prevent the wicked choices we freely insist on making.

Todd said...

I was surprised when Craig used Anselm’s ontological proof of God’s existence to substantiate that God is the basis for goodness in the moral sense. Wouldn’t our understanding of the concept of “greatest good” presuppose the existence of God if (as Craig argues) God is the basis for good? It seems that using Anselm to argue morality in this way is circular. Or maybe I misunderstand.

Roger said...

"Given that morality is objective, can it come only from God?"
But is morality objective, and if morality is objective, why should it come only from god? If morality only comes from god, then surely it is god's subjective view of morality. If it has another source, then morality does not come only from god. If morality comes only from god because god is moral, then there must be an external validation of god's morality.