WLC's Mistake in Formulating the Moral Argument

Actually, there is more than one mistake that WLC made in his formulation of the moral argument and there is more than one mistake that WLC made in his Q&A response, Formulating the Moral Argument.  I've addressed the moral argument a few times already; in The Modified William Lane Craig Moral Proof, The Logic of an Apologist, WLC3: When Morality Arguments are Bad, The Moral Argument in Graphical Form and WLC Being a Duffer.

In the first, I show how the same argument can be used to argue for the existence of a god via the existence of Kim Kardashian (if there is a better way to highlight how trivial an argument is, I'd like to know it).  I'll get to the second in a moment.  In the third, I summarise the first and second arguments and go into a little detail as to why I think WLC's formulation of the moral argument is questionable. In the fourth I demonstrated, in a diagram, the circularity of the moral argument.

The fifth is more about his Q&A response to a question about … well, it should be reasonably clear already -  it's the formulation of the moral argument, which I too questioned in The Logic of an Apologist.

Now, The Logic of an Apologist got some attention recently at Craig-Land, in a thread dedicated to lampooning my arguments, eventually focussing in on the moral argument and my assertion that WLC presents the moral argument oddly and, as I interpreted it in WLC3: When Morality Arguments are Bad, deceptively.

There's a bit of noise in the discussion, but through it is an interaction between me and the thread's originator, λ-calculus, that I want to use as my launch-pad to discussing WLC's errors.  Reasonably late in the thread (post #86) I challenged λ-calculus to answer the question I effectively posed in The Logic of an Apologist: why did WLC formulate the moral argument the way he did, if not to deceive?  I also reframed the question, to make it easier for him: Why did he word it in the form (~A->~B;B;∴A) rather than in the form (A->B;A;∴B)?

Someone else immediately chimed in and referred to WLC's Q&A response, Formulating the Moral Argument.  And this led me to take another look at the piece that was the inspiration for WLC Being a Duffer.

What a treasure trove!

My original comment on it was that WLC effectively torpedoes his own argument, but that is only one of many problems.  Here are the ones that I have identified:

Problem 1 - Equivocation on the term "objective"

This is a problem with the content of the argument.  By "objective" WLC really means "absolute", as opposed to "relative".  But he equivocates, perhaps unintentionally, when he tries to persuade atheists to accept that there is such a thing as "objective moral values and duties".  We can use metrics, generally agreed principles and cool reason to work out this action is better or worse than that action - that would be objective - rather than letting ourselves be swayed by personal opinion, hyperbole and how we felt on the day - which would be subjective.  When WLC gives examples of "objective values" saying this or that is "simply wrong", you can tell that he's really talking about absolute morality, rather than objective morality - or he wouldn't be raising emotive topics like rape or torturing children for fun.

Problem 2 - Equivocation on the term "conditional"

This is a problem with his explanation for formulating the argument the way he did.  He writes "Since, as our truth table reveals, the whole conditional statement comes out true if the consequent is true, then it doesn’t matter whether the antecedent clause is true or false."

Here he is mixing up what is known as a material conditional with what is known as a conditional sentence.  He's helped in this in that the terms both contain the word "conditional" and they can be symbolised the same way (A->B).  Wikipedia will explain the potential for confusion better than I:

The material conditional is used to form statements of the form p->q (termed a conditional statement) which is read as "if p then q" or "p only if q". It is conventionally compared to the English construction "If...then...". However, unlike the English construction, the material conditional statement p->q does not specify a causal relationship between p and q. It is merely to be understood to mean "if p is true, then q is also true" such that the statement p->q is false only when p is true and q is false.

These are quite different understandings of the construction "if … then …".

Problem 3 - Equivocation on the term "true"

When, in general parlance, we say that something is "true", we do not mean that it is "trivially true" or "vacuously true" - we mean that it is actually true such that its negation is not true.  WLC is equivocating by presenting as "true" something which is only vacuously true.

Problem 4 - Lack of specificity to morality (if Problem 2 is ignored)

This argument works (to the extent that it works) with anything, including the existence of Kim Kardashian.  It's not really a moral argument at all.

Problem 5 - Triviality due to negation (if Problems 1, 2 and 3 are ignored)

WLC himself points out that, from his point of view, his first premise is trivially true - because he believes that his god does exist.  But this isn't actually just from his (or any other theist's) point of view.  In terms of the argument, if it were to succeed, it makes its major premise trivial.

There may well be other problems, feel free to point them out.

The modified William Lane Craig moral proof

The text below is based on an original reply posted to an article on the Ethical Realism blog. JW Gray's article focused on two major problems with William Lane Craig's "proof" of god, which he (Gray) paraphrased as:

1. Either we must be reductionistic materialists or theists.
2. Reductionistic materialism can’t account for intrinsic values.
3. Theism can account for intrinsic values.
4. Intrinsic values exist.
5. Therefore, God exists.

The problems raised by Gray are:

1. Craig presents a false dilemma (the first premise), and
2. Craig presents a reductio ad absurdum argument (by seemingly forcing people who disagree with the fifth premise to feel obliged to abandon the fourth premise).

While it's true that there is a false dilemma, since there are other options beyond being a theist or a reductionist materialist, the claim of a reductio ad absurdum argument is based on a similarly false assumption - that everyone does or should agree with Craig's premise 4 (as paraphrased).  Gray agrees with premise 4 and argues that Craig's error lies in premise 1, where the false dilemma is presented.

I feel that there are indeed major issues with Craig's argument, but that the primary problem is that all four premises (as paraphrased) are unsupportable.  (Gray does actually make reference to this when he writes that the "(p)remises of an argument should be more plausible than the conclusion.")

Something that is of particular interest to me is that there’s a noticeable leap in Craig's argument from “theism” (being the belief in some sort of divine entity or entities) to an assumption of the existence of a specifically Christian God.  This is not just a figment of Gray's imagination ... the article by Craig that Gray is paraphrasing includes the following in his concluding remarks:

Theism is thus a morally advantageous belief, and this, in the absence of any theoretical argument establishing atheism to be the case, provides practical grounds to believe in God  ...

and

If ... we hold  ... that objective moral values and duties do exist, then we have good grounds for believing in the existence of God. In addition, we have powerful practical reasons for embracing theism in view of the morally bracing effects which belief in moral accountability produces.

Craig is cheating a little here.  He's credited as being a philosopher of theology, which allows him to use the word "believe" in two different senses simultaneously, or at least argue using one sense with one audience and another with a different argument.

Philosophical belief is "the attitude we have, roughly, whenever we take something to be the case or regard it as true".  Theological belief is something quite different, being linked to the idea of faith.

I can believe that a bus will arrive in the next five minutes, but I don't necessarily have faith associated with that belief.  Similarly, I might find that I believe something subconsciously, something that was taught to me as a child by someone I considered to be an authority at the time.  When quizzed on that belief, I would answer with what I thought to be true, for example if asked about the origin of the Angles (of Anglo-Saxon fame), I would have until quite recently identified Angles as a southern tribe of Celts.  That belief was wrong.  I accept it was wrong, and now can (hopefully correctly) identify Angles as being a Germanic tribe from a region just south of modern Denmark.

Theological belief is distinct from this mundane belief not only in so much as it is clung to with great tenacity, but also in that it is interwoven with the believer's world-view.  A theological type belief is more akin to a belief in gravity, although it's possibly more likely that an average physicist would be more willing to revise views on gravity than your average Christian theologian would be to revise an opinion on the existence of God.

Personally, I use "believe" in two senses, firstly to refer to things that I used to think were true, but now know to be incorrect.  My belief with regard to Angles is such an example.  Another example is that I used to believe that St Paul's Cathedral in London miraculously survived the Great Fire of London, as well as the London Blitz, probably because someone I respected at the time told me that convincingly.  My informer might have been confused because the London Blitz was also called "The Second Great Fire of London" - or it could be because miracle seekers would particularly like the idea that this beautiful piece of architecture survived miraculously not once, but twice.

The point is that it doesn't make sense to say that I used to "know" that the Angles were a tribe of Celts or that St Paul's Cathedral survived the (first) Great Fire of London, since this was a false belief.  Currently, I might say that I know something which I later find to be false.  The thing that I currently "know" will then turn into something which I "believed".


The other use of "believe" is to refer to things which I am not sure about.  I use the term this way sparingly, and try to either say "As far as I know, this is the case" or "I reckon this".  Informally though, "believe" works fine.

Which version of "belief" is Craig using in the quoted text?  Philosophical belief or theological belief?  Something previously thought to be true but since discovered to be false or something about which one is currently uncertain?  My suspicion that he is using the word so that "good grounds for believing in the existence of god" is interchangeable (in his mind) with "a good proof of the existence of god" is borne out by Craig's three part "proof" discussed shortly.

Craig's rush from "theism" into the existence of God makes his argument a fallacious one, in which he affirms the consequent.  He argues that because one has an explanation which (apparently) works, then that explanation is necessarily true.  Craig does this in a two step process, first by implicitly redefining “theism” to mean something along the lines of “acknowledging the factual existence of divine entities”, rather than “belief in divine entities”.  Then he argues that this single explanation for intrinsic values (the existence of a divine entity) is both true and comprehensive.

This is akin to saying:

1. If I have ripped a branch from a tree, then a branch will have been ripped from a tree.
2. A branch has been ripped from a tree.
3. Therefore, I have ripped a branch from a tree.

Actually, no, admittedly some of the smaller branches were ripped off by me, but the larger branch in my backyard at the moment was ripped off by a storm.  The logic works sometimes, but is not necessarily true and is certainly not comprehensive.

Craig's affirmation of the consequent also appears in the 3 part version of his “proof”, a version which more adequately reveals other issues with his arguments.  The tripartite "proof" is (in his own words):

1. If God does not exist, objective moral values and duties do not exist.
2. Objective moral values and duties do exist.
3. Therefore, God exists.

This version embodies the same problems as Gray's paraphrased "proof", in that the first two premises are far from controversial (premise 2 is Gray's paraphrased premise 4 and premise 1 is just a statement of opinion). 

What is more interesting is how Craig seems to want use this argument.  In my discussions with people interested in Ethics (the philosophical subject), it's quite obvious that many people seriously do believe that intrinsic values or objective moral values (and duties to a lesser extent) do exist.  It's not something which is limited to theists: Sam Harris, a quite vocal atheist, believes in the existence of objective morality and wrote a book defending the idea.  It seems to me that what Craig does is spike his argument with an apparently palatable, apparently obviously true second premise and spends his time arguing the first premise by redefinition and inversion.

Let's step through that.  We are presented with all three statements together.  We scan them and depending on our predispositions agree with or reject them.  Assume for the moment that we have already accepted the second premise.  Why have we already accepted the second premise so quickly, without even asking Craig to present us with evidence?

It's probably because anyone without an interest in philosophy who is not fundamentally broken (this excludes psychopaths, if you can find a psychopath who is committed to moral psychopathy, I'll update the blog accordingly) can think of something which they consider to be objectively wrong.  It could be murder, it could be rape, it could be abusing children, desecrating the dead or it could be something theological (such as failing to praise the correct deity).

Even some people with an interest in philosophy may agree that there are objective moral values, but these people will have a particular understanding of what an objective moral value is.  An objective moral value is a moral value which is independent of context, an objectively moral wrong is always wrong irrespective of context. It might be better to illustrate the point using an extreme example.

Is it wrong to kill a single innocent if by doing so you could save a thousand innocents?  If you say "yes", then you are a moral objectivist.  If you say "it depends on the larger context" or "I need more data to make a decision", you are probably a moral relativist (or possibly a moral universalist).  If you say "no", you are likely to be a moral consequentialist.

There is, however a subtlety that isn't amenable to the rough and tumble of a debate format.  I can agree that it is always wrong to act in order to kill an innocent, even though I might save a thousand innocents by doing so, and still act to kill the innocent, because it is also always wrong to let a thousand innocents die through inaction.  Since I am going to be wrong either way, I might as well be wrong with 999 more people surviving the experience (this is a form of meta-ethical pragmatism).  The point being that I can agree with the objective morality of never killing an innocent and, simultaneously, with never letting a thousand innocents die through my inaction - thus arriving at the moral consequentialist position.

On the other hand, I might be bound by the law.  The law takes a rather peculiar moral perspective on acts of commission and acts of omission.  Unless it's specifically my responsibility to save as many lives as I possibly can, I am not permitted to sacrifice one to save many.  If I am the Captain of a ship, I might have to close down an engine compartment and flood it with oxygen displacing gas, killing the lone inhabitant in order to ensure the safety of a much larger number of crew and passengers.  However, as the Captain of a ship, I am not permitted to push an obese member of the public from a footbridge in order to prevent a runaway train from plowing into an equally large number of innocent railway staff and passengers.  In one situation, I'd kill the one to save the thousand, in the other, I'd watch the disaster unfold with horror, along with my calorie-challenged friend.  In other words, although I still may agree with the objective morality of never killing an innocent nor letting a thousand innocents die, I could arrive at the moral relativist (or universalist) position.

So, in short, we can all (if we don't think about it too deeply) agree with Craig's premise 2 - there are objective moral values and duties - even if our final positions appear to be consequentialist, relativist or universalist.

Now we arrive at premise 1.  Craig can (and does) argue that “objective values” are things that flow naturally from the existence of a God to form and enforce them.  He does this by basically by arguing the reverse, in two separate ways.

For his theist audience, Craig appears to argue that if there is a God with all the necessary characteristics, then one would have a sound basis for objective values.  This is true and even I would agree with it, and I assume that even such vaunted figures as the Four Horsemen of Atheism would too.  We do so because there is an if-then.  At this point the unwary will be labouring under the misapprehension that premise 1 is true (when really a quite different premise is true, one with both instances of "not" removed) and will have already accepted that premise 2 is true, and is then presented with 3 which does indeed follow logically, if you’ve accepted 1 and 2.

For his atheist (or uncommitted) audience, Craig appears to be arguing something slightly different due to the curious wording of the premise.  He explains this in The New Atheism and Five Arguments for God:

In a pluralistic age, people are afraid of imposing their values on someone else. So premise 1 seems correct to them. Moral values and duties are not objective realities (that is, valid and binding independent of human opinion) but are merely subjective opinions ingrained into us by biological evolution and social conditioning. 

He makes the assumption that all atheists are moral relativists (or at least moral univeralists).   This is an unchallenged assumption.

Another assumption which seems to go unchallenged is hidden under the bland wording of premise 1. Unless you redefine “objective moral values and duties” to mean “the moral values and duties as laid down by God”, there’s no reason to believe that the premise is necessarily true. If you don’t require necessary truth for premise 1 or you are not challenged on it (and you accept premise 2), you can argue whatever you like:

1. If the Grand Pixie does not exist, objective moral values and duties do not exist.
2. Objective moral values and duties do exist.
3. Therefore, the Grand Pixie exists.

Noting the form of this argument, and the apparent rules which allow you to make unsupported assertions in premise 1, we could present the Kim Kardashian proof (I couldn’t think of anything less meaningful than Kim Kardashian, and I did try):

1. If God does not exist, then Kim Kardashian does not exist.
2. Kim Kardashian does exist.
3. Therefore, God exists.

This is, in fact, a better argument because premise 2 is now less contentious – sadly very few people appear to be unaware of the existence of Kim Kardashian, and those who are can follow the link provided to ascertain whether she exists or not, whereas there are schools of thought which deny the existence of “objective moral values and duties”, either totally or in part.  Theists could hardly deny the truth of premise 1, right?  They would have to agree that if there were no God, then nothing would exist, including Kim Kardashian.  

Admittedly, in the atheist camp, the argument is just as contentious with respect to the original first premise, but I do think that the scale of improvement in the second premise more than compensates for any lack of improvement in the first.