Angra Mainyu's blog: A Reply to Craig's Metaethical Argument (aka Craig's "Moral Argument") II (v. 7.4)

In this article, I will raise a number of objections to his Craig’s metaethical argument (aka “Moral Argument”), arguing that he’s failed to establish or provide good reasons to think that the first premise is true.

The formal argument is:

P1: If God does not exist, then objective moral values and duties do not exit.

P2: Objective moral values and duties do exist.

C: God exists.

Additionally, I will briefly address other arguments he makes usually in the same context as the metaethical argument, like an argument from moral accountability.

Before I go on, I’d like to point out that I do not claim originality. In fact, I took most of the ideas in this article from a number of different, often multiple, sources. [0]

2. Objectivity.

In his defense of a metaethical argument, Craig says that a something is objective if it does not depend on anyone’s opinions, and it’s subjective if it does. [r1]

So, that raises some questions, like:

a. What kind of ‘something’ counts?

Are we talking about objective judgments, or statements? Facts? Something else?

Craig doesn’t say, but in his defense of the metaethical argument, Craig talks about objective values and duties, as well as objective truth [r1], and implicitly about objective statements and/or propositions when he talks about what “holds true”[r2] independently of what people believe.

In a different article – not defending the metaethical argument but clearly referring to it -, Craig also talks about objective facts. [r 3 ] [1 ]

b. What kind of 'dependence' is relevant?

In the rest of this section, I will analyze the matter of objectivity in the context of Craig's metaethical argument.

2. Objective statements or judgments, and objective matters.

Let’s consider some examples of statements or judgments[2]:

S1: All combustible materials contain phlogiston.

S2: Santa Claus exists, and flies in a sleigh.

S3: Richard Dawkins believes that Yahweh exists.

Using the word ‘objective’ and other words in those statements in an ordinary sense, it’s clear to me that there is an objective fact of the matter as to whether S1 is true, there is an objective fact of the matter as to whether S2 is true, and there is an objective fact of the matter as to whether S3 is true. They’re all false.

In that ordinary sense of ‘objective’, I would also say that whether combustible materials contain phlogiston is an objective matter, so is the matter(s) of whether Santa Claus exists and flies in a sleigh, and so is the matter of whether Richard Dawkins believes that Yahweh exists.

As for Craig’s use of ‘objective’, whether all combustible materials contain phlogiston, whether Santa Claus exists, etc., are matters that do not depend on anyone’s opinion, so those matters and the corresponding statements seems to meet Craig’s requirement for objectivity as well.

What about S3?

Is S3 objective, in the sense of the term ‘objective’ used in Craig’s metaethical argument?

In a sense, whether S3 is true depends on one of Dawkins’ beliefs – namely, whether he believes that Yahweh exists.

However, that is not the relevant kind of dependence when it comes to ascertaining whether a statement, a matter, etc., is objective, in the sense in which Craig uses the word ‘objective’ in the context of the metaethical argument.

If that were the relevant kind of dependence, it’s clear that moral judgments would often not be objective in the sense of ‘objective’ used in Craig’s metaethical argument, since – for instance – we can set up scenarios in which, say, whether Bob has a moral obligation to refrain from putting some of the white substance in the sugar bowl that is on the table –and which is in fact sugar – in his father’s coffee does depend on facts such as whether, say, Bob believes that the substance in question is poison.

But there is no reason to think that Craig meant to count that kind of dependence against objectivity, and there seems to be no other sense in which S3 would depend on anyone’s beliefs.

So, it seems that in the cases of S1, S2 and S3, objectivity in the ordinary sense of the term outlined above and objectivity in the sense of the term used by Craig in his metaethical argument go together.

Still, one may ask whether Craig – at least, when it comes to the objectivity of statements – is actually using the word ‘objective’ in the same sense as the ordinary sense outlined above, or S1, S2 and S3 are just cases in which the referent of the term ‘objective’ in the ordinary sense outlined above and the referent of the term ‘objective’ in the sense of the term used by Craig in his metaethical argument, happen to overlap.

A number of his statements strongly support the hypothesis that the meaning is the same or approximately close to that, at least when it comes to the objectivity of statements, matters, etc., despite some obscurity in Craig’s formulation, since – for instance:

a. In his defense of the metaethical argument[r1], Craig makes a classification of matters between objective ones and matters of personal taste or opinion. Objective matters would be those that do not dependent on people’s opinion, whereas those that do depend on that would be subjective, and those would be matters of taste, or opinion.

b. Craig also uses such examples of non-objectivity (i.e., matters of taste or opinion) in the context of his argument against gay relations[r 3 ], where he also makes some of the main metaethical claims he makes in the context of his metaethical argument.

In that context, he’s using the term ‘objective’ in the usual sense – he does not try to define it in that context, but rather appeals to the intuitive grasp of the term.

In that context, he also opposes matters of fact to matters of taste, or matters of opinion.

Those considerations also allow us to infer how Craig classification works for matters. In that case, there would be two categories:

1. Matters of fact, or objective matters.

2. Matters of opinion, matters of taste, or subjective matters.

There might be debatable subtleties about, say, whether there are some differences between the meanings of ‘it’s a subjective matter’, ‘it’s a matter of taste’, ‘it’s a matter of opinion’, ‘there is no objective fact of the matter’, ‘there is a fact of the matter’, etc., but for now, we may establish the following equivalences, which seem to be in line with common usage of the terms, and also seem to match Craig’s usage:

E1: A matter M is objective if and only if M is a matter of fact if and only if M is not a matter of opinion.

E2: A statement or judgment S is objective if and only if there is a fact of the matter as to whether S is true if and only if there is an objective fact of the matter as to whether S is true.

Perhaps, someone might disagree with the equivalence between ‘there is a fact of the matter as to whether S is true’ and ‘there is an objective fact of the matter as to whether S is true’. I will address the issue in this context later.

2.2. Objective moral values and duties.

While the meaning of ‘objective’ in the case of statements, judgments, matters, etc., appears clear enough, the terms ‘objective values’ and ‘objective duties’ in Craig’s metaethical argument are more obscure.

When addressing the matter, Craig makes a distinction between what he calls ‘moral values’ and ‘moral duties’, and associates the first with good and bad, and the second with right and wrong. [r1]

I partially agree with Craig’s distinction, though not completely. The degree of agreement is not entirely clear. For example:

1. I do not see any difference between saying ‘Bob behaved immorally’ and ‘Bob behave in a morally bad way’. More generally, I do not see any differences in the meaning of ‘morally bad’ when applied to actions, and ‘morally wrong’.

On the other hand, ‘morally bad’ is also predicated of agents, and as such it’s a character trait, whereas ‘morally wrong’ is predicated of actions only.

Craig’s explanation of the distinction he makes is not sufficiently detailed for me to tell whether he would disagree with any of my assessments in point 1.

2. One may talk about doing something that is ‘a necessary evil’, and in that case, a person may be bringing about something bad – or evil -, but in order to prevent something worse – for instance -, and without behaving in a morally bad/morally evil way.

Craig’s explanation of the distinction he makes is not sufficiently detailed for me to tell whether he would disagree with any of my assessments in point 2.

3. Also, in that context, I do not think that ‘a necessary evil’, or ‘something bad’, means morally evil, or morally bad. And it seems probable to me, given his words, that Craig would disagree with my assessment on that matter.

4. I would make a distinction between bad things, situations, even behaviors, agents, etc., and morally bad behavior, agents, etc. For instance, a child dying of cancer is a bad situation, but I don’t know there is anything morally bad about it, as long as no moral agents are involved in causing it. Also, the behavior of some agent may be a bad thing without being morally wrong, or morally bad. For instance, a crocodile drowning a child in order to eat him is a bad thing. But the crocodile’s behavior is not morally bad, as far as I can tell. It’s not morally anything.

Craig's explanation of the distinction he makes is not sufficiently detailed for me to tell whether he would disagree with any of my assessments in point 4.

5. As for goodness, I would agree of course that there is a difference between morally good actions and morally obligatory actions. For instance, some actions involving self-sacrifice are plausibly morally good but not obligatory, but supererogatory. It seems clear that Craig accepts that distinction too.

Still, here we may stick to Craig’s distinction for the sake of the argument, since whether he’s right about that particular point makes no relevant difference in the context of the objections to his metaethical argument I’m going to raise.

However, in any case, his explanation of the distinction between good/bad and right/wrong does not elucidate what Craig means when he claims that objective moral values and duties do exist, though he provides some information that we may use to analyze his claim. On that note, we can tell the following:

1. In asserting the second premise, Craig is saying that there is an objective fact of the matter as to whether, say, the Holocaust was immoral. And the same goes for other moral matters.

2. It’s clear also that the second premise does not only entail that there is an objective fact of the matter as to whether, say, the Holocaust was immoral – and the same for other matters about right or wrong or [morally] good or bad, etc.-, but also at least that some judgments like ‘A has a moral obligation to Z’, ‘Y is morally good’, etc., are true., for some actual behavior or agent Y, and some actual agent A.

We can tell that 2. is true for the following reasons:

2.a. If the second premise only entailed that there is an objective fact of the matter as to whether behaviors are right or wrong, etc., then that would be compatible with a moral error theory. As a parallel, the statement ‘All combustible materials contains phlogiston’ is objective, but false.

2. b. The reason for the actuality requirement is that without it, then also a moral error theory would seem to be compatible with the second premise. For instance, let's say theory T1 holds that:.

2.b.1. There is an objective fact of the matter as to whether the Holocaust and other behaviors are morally wrong, whether some agent has a moral obligation, etc. However, no actual agent has any moral obligations, or is morally good or morally bad, etc.

2.b.2. Statements like ‘All immoral behaviors are immoral’, ‘If a person has a moral obligation not to lie, then that person has a moral obligation not to lie’, ‘All moral obligations are moral obligations’, ‘All entities who have at least one moral obligation have at least one moral obligation’, etc., are all true, and furthermore necessarily true.

2.b.3. It's metaphysically possible that there is some agent that has a moral obligation, but actually, no agent has any moral obligations, and given the way the world actually happens to be, there will never be any agent who will have any moral obligations and/or will be morally good or bad, etc.

So, T1 is a moral error theory. But it seems clear that Craig intended to rule out all such theories in his second premise, and that’s also clear given the arguments he gives in support of it.

So, based on the above, we can tell that the second premise of Craig’s metaethical argument at least implies the two following two conditions:

OMVD1: Statements – or judgments, or whatever one calls them – of the form ‘X is immoral’, ‘Y is morally good’, ‘A has a moral obligation to Z’, etc., are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that gay sex is immoral among humans – as Craig does [r3]-, then there is an objective fact of the matter as to whether that claim is true, and so on.

OMVD2: Some statements of the form ‘A has a moral obligation to Z’ are true, and some statements of the form ‘Y is morally good’ are true, and so on, where Y is an actual behavior of a person, or a person, A is an actual agent, etc.

The two issues are very different ones, as the phlogiston and Santa Claus examples – for instance – illustrate.

For example, S1 and S2 are objective statements, but all statements of the form ‘X contains phlogiston’ are false. Similarly, someone might maintain that OMVD1 is true, but OMVD2 is false – and so, that a moral error theory obtains, etc.

Also, for instance, to say that no statement of the form ‘X contains phlogiston’ is true for any actual X is very different from saying that, at least for some X, whether X contains phlogiston is a matter of taste, or a matter of opinion, or that there is no objective fact of the matter as to whether X contains phlogiston, etc.

I will address that distinction in the context of Craig’s claims in greater detail later.

For the moment, given the analysis so far, and despite some obscurity in his argument, we can tell that at least P2 entails OMVD1 and OMVD2.

Still, some questions remain, like:

a. Does P2 mean the same as the conjunction of OMVD1 and OMVD2?

b. If not, are P2 and the conjunction in question at least necessarily equivalent conditions?

Given how Craig argues, it seems to me that a. is plausibly true – and hence, so is b.

However, in any case, Craig’s arguments in support of P2 are essentially an appeal to human intuitions in support of OMVD1 and OMVD 2 and nothing beyond that, so if P2 implied something not implied by the conjunction of OMVD1 and OMVD2, Craig would have failed to support the premise, and so his theistic case would fail on account of that, independently of other considerations.

Moreover, there seems to be no good reason to grant anything beyond those conditions.

Given that, and since P2 entails the conjunction OMVD1 and OMVD2, we may consider the conditions necessarily equivalent. In other words, we may assume in this context that b. is true – which seems to be the case, anyway, given many of Craig’s words.

Under that understanding, let’s accept Craig’s second premise.

2.3. Objectivity and mind-dependence. Objective pain and objective fear.

As a parallel to the claim that objective moral values and duties do exist, let’s consider the claim that objective pain does exist. As in the case of OMVD1 and OMVD2, the claim that objective pain exists means the same as or is at least equivalent to the conjunction of the following two conditions:

OP1: Statements – or judgments, or whatever one calls them – of the form ‘X is in pain’ are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that Bob is in pain – Bob is a specific agent -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether Bob is in pain is not a matter of opinion, but a matter of fact.

OP2: Some statements of the form ‘A is in pain’ are true, where A is an actual agent.

Now, the claim that objective pain exists, so understood, is true.

For example, just as the matter of whether there is a cat on a specific mat over there is a matter of fact, not a matter of opinion, so is whether the cat that is on the mat – let’s say there is one, named 'Bob' – is in pain.

Also, there are some actual agents that are in pain.

Similarly, we may consider objective fear, and it's clear that objective fear does exist. For example, if Alice is a specific person, whether Alice afraid of the actual cat that is on the mat, or whether she fears tarantulas, are matters of fact, not matters of opinion.

Moreover, it's clear that some actual agents do fear some things.

So, objective fear does exist.

Now, there is a sense of 'mind-dependent' in which fear, pain, etc., are mind-dependent states.

However, the meaning of ‘objective’ (and ‘subjective’) used by Craig in the context of his metaethical argument is not related to that matter.

That said, there are two points that I would like to address here:

1. It might be argued that it’s not possible for Bob to experience pain without believing that he is experiencing pain. Does that mean that whether Bob is in pain depends on Bob's beliefs, and so it's not an objective matter?

Leaving aside the issue of whether ‘depends’ would be the correct term here, and even if it’s true that it’s not possible for Bob to experience pain without believing that he is experiencing pain, that is still not a problem for objectivity in this context. The matter of whether Bob is in pain is not a matter of opinion. It’s a matter of fact, and that is the classification that Craig uses, and the measure of objectivity as he uses the words in the context of the metaethical argument.

2. Someone might suggest that even if there is a fact of the matter as to whether the cat that is on the mat is in pain, there is no objective fact of the matter, because objectivity requires mind-independence.

As I see it, there is no semantic difference in colloquial speech between the meanings of the statements 'there is an objective fact of the matter as to whether the cat is in pain', and 'there is a fact of the matter as to whether the cat is in pain', etc., so I will use the expressions interchangeably unless otherwise specified.

However, even if I'm mistaken about that, the point would remain that whether Bob – i.e., the cat in question – is in pain is a matter of fact, not a matter of opinion, and that is the classification that Craig uses, and the measure of objectivity as he uses the words in the context of the metaethical argument. [3]

So, if it turned out that there is a difference in colloquial speech between the meaning of the statement 'there is an objective fact of the matter as to whether the cat is in pain', and the meaning of the statement 'there is a fact of the matter as to whether the cat is in pain', etc., then Craig would be using 'objective' in a different way, but in any case, by his usage of the words, it would remain the case that objective pain exists, objective fear exists, and so on.

So, in short, it seems that objectivity in the context of Craig's metaethical argument is not about mind-independence and mind-dependence – at least not unless there is a sense of the mind-independence/mind-dependence classification that matches the matter of fact/matter of opinion classification.

2.4. Objectivity and necessity.

While this section is not required in the context of Craig’s metaethical argument,– this section can be seen as a long side note -, I think the matter of necessity might be important in the context of some ontological metaethical arguments for theism, if some points are raised, so I will address it.

While objective fear and objective pain do exist, it’s clear that there are possible scenarios in which they do not. It's true that possibly, no agent experiences pain or fear.

Now, let W1 be a scenario in which there is never any entity that experiences any pain.[4 ]

If W1 were actual, then objective pain would not exist. However, there would still be a fact of the matter as to whether a specific entity E that exists in W1 is in pain. Moreover, the matter of whether an entity in W1 is in pain would not be a matter of opinion. Rather, the fact of the matter would be that no entity in W1 is in pain.

Now, let’s consider objective moral values and duties.

It seems that on Craig’s ontology, the claim that objective moral values do exist would be necessarily true – since Craig’s ontology posits that God is necessary -, but the claim that objective moral duties do exist would be contingently true, since plausibly God could refrain from creating any entities to whom he would issue any commands, so OMVD2 would not be true with regard to moral obligations, and since the second premise affirms that both objective moral values and objective moral duties exist, it seems that the second premise is, on Craig's ontology, contingently true.

Since I see no good reason to believe that a necessary agent exists, or that necessarily, there are agents, or that necessarily there are entities that are morally good, or morally bad, or have moral obligations, etc., I would reject both a claim that objective moral values necessarily exist, and a claim that objective moral duties necessarily exist.

Of course, to be clear, if W2 [4 ]is a scenario in which objective moral values and/or duties do not exist, it does not follow from that that whether some agent (say, a shark) that exists in W2 has a moral obligation, or is morally good, etc., are matters of opinion, or that statements like ‘It’s immoral for any adult human to torture infants for fun’, or ‘If an adult man were to torture infants for fun, he would be acting immorally’, would not be true in W2.

But the contingency of the claims that objective moral values and/or duties exist is not problematic.

So, in short, upon reflection, I see no problem in granting the second premise from a non-theistic perspective, but I would refrain from granting that it’s necessarily true – in any case, Craig’s own position seems committed to the contingency of the second premise, for the reasons given above.

3. The first premise.

The first premise of Craig’s metaethical argument is that if God does not exist, then objective moral values and duties do not exist.

For now, I will not address the issue of what Craig means by “God”, and focus on the consequent.

So, in order to defend the first premise, Craig more or less explicitly assumes a scenario in which God does not exist, and makes some arguments[5 ] in support of the claim that in such a scenario[6 ], objective moral values and duties do not exist. [7 ]

However, there some very different ways in which someone might argue that, in a certain scenario, objective moral values and duties do not exist. For instance, someone might potentially, and under the assumption that God does not exist:

a. Argue for the conclusion that OMVD1 is false. [8 ]

b. Argue for the conclusion that OMVD 2 is false.

c. Somehow argue for the conclusion that the conjunction of OMVD1 and OMVD2 is false, without making a specific argument for the falsity of one of them in particular.

For instance, if OMVD1 is false, then there is no objective fact of the matter as to whether, say, the Holocaust was morally wrong – or perhaps there is in that case, but not in many others; I will leave those issues aside to simplify.

On the other hand, if OMVD2 is false, it seems that there is an objective fact of the matter – namely, it’s not the case that the Holocaust was morally wrong.

So, one may wonder what Craig is arguing for.

In his defense of the metaethical argument[r1], he first asks – rhetorically – whether without God, moral issues are akin to matters of taste, like whether chocolate tastes better than vanilla. That suggests that he’s trying to say that without God OMVD1 would be false, and there would be no objective fact of the matter as to, say, whether the Holocaust was immoral – or other moral judgments -, and also that it would be a matter of taste, etc.

On the other hand, in the same context, Craig claims that without God, the terms ‘good’ and ‘evil’ would have no reference point. But that actually suggest an error theory, at least for ‘good’ and ‘evil’, since taste judgments plausibly do have referent points even if speaker-dependent, and it’s improbable that he would argue that without God, an error theory holds for good and evil, but instead right and wrong are matters of taste.

So, what is Craig arguing for?

In another post [r 3 ], Craig says that without God, those who claim that right and wrong, and good and evil, are matters of taste, would be “absolutely correct”, that the matter becomes one that depends on culture, and so on. That clearly indicates that he’s claiming that without God, OMVD1 is false.

Yet, in the same document, he claims that without God, everything is permitted. But that seems to be a claim that ‘Behavior Z is permissible’ is true for all Z, which would also imply that ‘A has a moral obligation to X’ is false, for all A, X, and so on, rather than ‘it’s permissible to me, but not to you’, etc.

So, Craig appears to be jumping between claiming that without God, a moral error theory obtains, and that without God, some sort of speaker-relativism or subjectivism[9 ] is true.

His arguments do not seem to be of much help figuring out what he’s trying to establish, either. He goes on to make a number of claims, but we’re not closer to establishing what it is that he’s trying to show would happen without God.

Still, let’s take a look at what a theist would probably need to establish if he claimed without God, OMVD1 is false, and what he would need to establish if he claimed that without God, OMVD 2 is false.

With regard to OMVD1, it seems that at the very least, in order to establish that without God, OMVD1 is false, a theist would have to show that without God, at least one of the following conditions obtains:

i. Different competent speakers of moral language mean different things when they make moral statements, even if they’re using all terms correctly.

ii. Moral statements are statements about the speaker (e.g., like ‘X is immoral’ means ‘I do not like X’).

iii. The standards each human being is using when assessing moral wrongness, goodness, etc., vary from person to person[1 0 ] in ideal conditions. [1 1 ]

Those conditions overlap to some extent, but that’s not a problem.

So, regarding condition iii, that is not the same as establishing that there is moral disagreement – which is clear -, or that different people sometimes have different theories about what the proper standards are – which is clear as well.

Rather, a theist would have to establish that, when different [actual or potential] human beings use their own sense of right and wrong without making any errors, they come up to different moral assessments sometimes – or maybe a sufficient number of times, but let’s leave that aside to simplify. [1 0 ]

For instance, as an analogy to this, it may be that Bob uses his gustatory taste and says that the chocolate cake tastes better the apple pie, whereas Alice uses hers and says the apple pie tastes better, and neither of them is making an error in the use of their sense of gustatory taste, which are the standards they use for assessing tastiness. So, under ideal conditions, the difference persists.

In order to establish that without God, OMVD1 is false, it seems to me a theist very probably would have to establish that without God, the faculty or faculties competent speakers of moral language use to make judgments of right and wrong or good and evil[1 2 ], are such that they diverge in a sense similar to what the gustatory taste example shows, or perhaps that i. or ii. obtain, for the following reason:

If the human sense of right and wrong and/or good and evil converge under ideal conditions [9 ], very probably there would be an objective fact of the matter in the sense of ‘objective’ relevant in this context, at least if people mean the same by the terms.

For instance, if there is convergence under ideal conditions, and Alice says that the Holocaust was immoral, whereas Bob says that the Holocaust was not immoral, it seems at least one of them is making – deliberately or not – a false claim, since in ideal conditions, they would both converge to one assessment – in this case, that it was immoral. It seems, in particular, that whether the Holocaust was immoral plausibly would not be a matter of opinion, or of taste. The same would apply to other moral judgments.

Perhaps, a theist might say they do not need iii., because the ‘human beings’ condition is allegedly too restrictive, and that there would be no objective moral values and duties if there are some other, non-human agents (say, aliens from another planet), with something akin to a sense of right and wrong or good and evil, but associated with different behaviors, states, etc., from the ones the human sense or senses are associated with.

However, that would have to be argued for, and that would seem to require arguing some moral semantics, since the theist in question would have to show – for instance – that the aliens’ utterances count as moral claims – rather than only something akin to them – and that are relevant to assessing whether there is an objective fact of the matter as to whether some behavior is immoral, even though in other cases – e.g., color – whatever the aliens do or say does not count when it comes to objectivity.

I will take a look at alien examples in greater detail later, since Craig uses them, even though he does not explain what he’s trying to establish with that example – i.e., is he trying to establish OMVD1, or OMVD2? And how?

For now, I will point out that Craig claims he’s not doing moral semantics, but moral ontology [r4][r5], yet establishing i., ii., iii., or even the alien variant, would seem to require arguing about the meaning of moral terms – i.e., it would precisely require arguing moral semantics.

Let’s now take a look at another option: a theist might claim that without God, OMVD 2 is false.

In order to establish that, he would have to show that without God, all statements like ‘X is immoral’, or all statements like ‘X is morally good’, etc., are untrue.

How might he argue in support of one of those conclusions?

One way would be for him to consider many hypothetical scenarios without God, and argue that in all of them the claim ‘X is immoral’ would be untrue. Clearly, that is a very heavy burden.

Indeed, if we use our sense of right and wrong intuitively, it seems clear that an adult human torturing children for fun is behaving immorally, even if we stipulate in the scenario that God does not exist. In fact, whether God exists is intuitively orthogonal. Granted, a theist might say that they have different intuitions. But why should we put aside our intuitions and accept his?

An alternative approach would be for the theist to, say, try to show that the meaning of the term ‘immoral’ is such that the claim ‘X is immoral’, where X is an actual behavior, entails the existence of certain entity that would not exist without God. A simple example would be if, say, ‘X is immoral’ meant ‘God forbids X’, but that’s not the case, so the theist would have to defend another hypothesis, if he were to take this route.

In any case, it seems that the theist would have to do some moral semantics if he tried that course of action. Also, of course, in order to establish such claims about the meaning of ‘immoral’, the theist would still need to appeal to his intuitive grasp of the terms, and so to his sense of right and wrong. But why should we follow his sense of right and wrong, instead of ours?

In the specific case of Craig, he says that in his metaethical argument, he’s not doing moral semantics, but moral ontology [r4][r5].

Is there another route?

In any case, it’s hard to see also how a theist might go about establishing results about moral ontology without doing at least some moral semantics.

That said, I will consider Craig’s specific claims and arguments in support of P1, granting that objective moral values and duties do exist, under the understanding explained above.

3. A traditional view on moral values and moral duties?

According to Craig[r2], moral values were traditionally considered “anchored” on God, and moral duties were considered to come from God’s commandments. In that context, God is understood to be the ‘Supreme Good’. Also, Craig maintains that by definition, God is the greatest conceivable being (GCB). [r5][r5b]

Now, while ‘moral duties’ is a term in English, the concept of moral duties exists in other languages as well. As for ‘moral values’, the expression is somewhat obscure, but in any case, the concepts of morally good and morally bad exist in other languages as well, and so does the concepts of justice – which Craig refers to as a moral value.

So, once we take that into consideration, it’s clear that Craig’s claim is not true as a general claim about humans, or human civilization, etc., but is limited to some cultures.

For instance, in China, people do not traditionally believe that their obligations have to do with the commands of God. In fact, they traditionally do not believe in God, in the sense in which Craig is using the word ‘God’.

That way is not particularly precise (e.g., the word 'greatest' is not so clear in that context), but it's clear enough to tell that God is not the creator of some Chinese folk religions, like Pangu or Shangdi, and isn’t anything posited by Buddhism, either, or by any religion common in China.

Also, in Japan, there is no traditional belief in the existence of God, or on any “anchoring”, etc., of goodness or of moral obligations in anyone’s nature and/or commands, or in any other agents for that matter. There are different religious beliefs, like Shinto or Buddhism, that influenced each other and had more or less acceptance in their various forms at different times, but in any case, they did not posit a God-based moral ontology, or any divine commands, and so on.

In fact, historically, there are plenty of cultures without belief in God, and without a belief that moral obligations were the commands of certain entities, even if most people in those cultures traditionally believe or believed in different entities that are sometimes called in English ‘gods’, but which do no metaethical work. Similar points apply to moral goodness.

So, it’s not the case that traditionally, among humans, moral goodness was considered to be anchored in God or in any other entity for that matter, and the same goes for moral duties. Instead, beliefs on the subject appear to vary widely from culture to culture, and sometimes within cultures. [1 3 ]

4. Naturalism, moral values and duties, pain and fear.

Craig claims that the most popular view among atheists is Naturalism, which Craig describes as “the view that science and science alone determines what exists. What exists is what our best scientific theories of the world require. If something is not required by our best scientific theories of the world, then it does not exist.”[r2]

Now, Craig does not say there whether it’s our current best scientific theories, some future theory that will plausibly be developed.

If it’s our current theories, then it’s very obvious that Naturalism so defined is false, since there are many matters of existence of objects that remain unsettled in present-day physics – for instance -, and there almost certainly are particles we do not know about yet, etc.

But moreover, even if one includes future theories, it seems clear that neither human scientists nor any post-human beings [1 4 ]will discover all of the things that exist on all of the planets in the entire universe, and in particular, that if there are alien lifeforms elsewhere in the universe, chances are some – even most – of them will never be discovered by human or post-human beings.

Still, perhaps, in his definition of '‘Naturalism’, Craig meant that Naturalism is the view that what kinds of entities exist is somehow “determined” by science alone, or something along those lines. That would raise questions about how ‘kinds’ and ‘entities’ are understood in his context, so more precision on Craig’s part would be needed, but in any case, at the very least, Naturalism so defined is probably unjustified, since there seems to be no good reason to think that human beings even or post-human beings will ever have the kind of epistemic access that it probably would require, under any reasonable understanding of Craig’s definition.

In any event, Craig’s claim that Naturalism so defined is the most popular view among atheists seems to be also unjustified.

But problems for Naturalism so defined aside, let’s assess Craig’s claim that on Naturalism, there are no objective moral values:

In support of that claim, he gives the example that one cannot find moral values in a test tube.

It’s a strange example. Science is not done only in test tubes. In any case, if one can’t find moral values in a test tube, then it seems that similarly, one can’t find fear or pain in a test tube. What would be the difference?

If Naturalism – as defined by Craig – is committed to the claims that objective pain does not exist and/or that objective fear does not exist, and so on, then we can easily tell that Naturalism so defined is false, independently of the previous considerations about alien planets or particles, and independently of metaethical considerations as well.

But perhaps, someone might claim that on that kind of Naturalism, objective pain does exist, since our best theories about, say, pain, seem to require that there is objective pain, even if we can’t find pain in a test tube. Similarly, it might be claimed that since our best theories about, say, fear, require that there is objective fear, there is objective fear on naturalism so defined.

However, in that case, one may also point out that our best theories in psychology seem to require that objective moral duties do exist. For example, based on our best theories in psychology and experiments, we can tell that objective antisocial personality disorder does exist, and such disorder is characterized as “... a pervasive pattern of disregard for, and violation of, the rights of others that begins in childhood or early adolescence and continues into adulthood”[1 5 ]

That seems to require that others have rights that can be violated in the first place. Also, it’s not the case that those rights in the definition are legal rights. Surely, people may well meet the criteria for having antisocial personality disorder even if they live in, say, a failed state, or generally a lawless land. Alternatively, a brutal ruler with antisocial personality disorder may respect the law – only he makes the laws that he likes.

But if there are rights, plausibly necessarily there are duties.

Moreover, Craig provides no good reason to think that our best scientific theories will not require objective moral values as well. For instance, studying the conditions in which children are more likely to become good persons is a perfectly valid scientific research goal [1 6], as long as we have a reasonably good epistemic access to moral truth – which Craig accepts. Granted, a theist might claim that if theism is not true, that would not be a valid research goal. But the burden would be on him. The point here is that Craig provides no good reason to think that in the future, scientists will either be theists or reject such research projects.

So, if we include psychology and/or neuropsychology, in the definition of ‘Naturalism’, then Craig has failed to establish that on Naturalism so defined, P2 is false.

On the other hand, if we exclude psychology, neuropsychology, etc. from the definition, then Naturalism as defined by Craig is committed to the non-existence of objective pain, pleasure, anger, fear, etc.; in that case, it's very obvious that Naturalism so defined is false, even if we don't consider metaethical questions.

But in any case, a non-theist need not worry about that. She may simply not adhere to Naturalism as defined by Craig – I don't adhere to it, for instance -, and point out that – for instance – she still sees no good reason to believe that God exists – I don't, for example.

5. Morality and determinism.

In another article, Craig also maintains [r 1 0] that without freedom, our choices are not “morally significant”, and that there is no freedom if determinism is true.

I would say that if we cannot make choices of our own free will, then plausibly we cannot behave immorally, and we cannot have moral obligations. But Craig does not seem to provide any good reason to suspect that if determinism is true, we cannot make choices of our own free will, or anything like that.

Even granting for the sake of the argument that the libertarian conception of free will is coherent, that does not entail that the libertarian account is true.

In fact, unless the regular concept of making choices of one’s own accord or one’s own free will – which is what plausibly is needed for morality, freedom-wise – has ontological commitments incompatible with determinism – or with causal determinism, if someone makes that difference and claims the problem is only with causal determinism -, it seems there is no difficulty in this context.

But Craig does not show that there are such ontological commitments in the regular concept of free will. Rather, he claims that we would be like puppets, etc., “controlled by the strings of sensory input and physical constitution”, but does not provide a definition of ‘physical’, or explain why we would not be making our choices freely.

In particular, causation at a lower level does not seem to deny causation at a higher level. For instance, even if there is a causal explanation in terms of particles of the event consisting in a lioness killing a wildebeest, that does not entail or suggest that she did not kill the wildebeest, or even that she did not kill the wildebeest because she was hungry. If a theist claims otherwise, the burden would be on them.

Of course, this is by no means a comprehensive analysis of compatibilist accounts of freedom. But there is no need for that here.

Craig claims, in the context of one of his defenses of the claim that morality needs theistic metaethical foundations, that freedom requires non-determinism – and a specific kind of non-determinism -, but has not shown that that is the case.

6. Color, exobiology and genetic engineering.

As part of his defense of the metaethical argument, Craig also considers some hypothetical scenarios in which aliens evolved differently,[r1 1 ] and claims that those hypothetical scenarios support his claim that without God, there are no objective moral values and duties.

I will address the specific case of objective moral values and duties later, but first, I will consider other cases, as a means of introducing parallels.

So, let’s consider the hypothesis that objective color does exist, which – mirroring the analysis in the case of objective moral values and duties – means the same as the conjunction of the following two conditions, or is at least equivalent to the conjunction of the following two conditions:

OC1: Statements – or judgments, or whatever one calls them – of the form ‘X is green’, ‘Y is red’, etc., are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that the cucumber on the table is green – pointing at a specific table -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether the cucumber on the table is green is not a matter of opinion, or a matter of taste.

OC2: Some statements of the form ‘X is green’, or ‘Y is red’, etc.., are true, where X, Y, etc., are actual objects.

Incidentally, it seems Craig agrees that objective color does exist.[r11b]

Now, we already know that, even here on Earth, different animals have different visual systems that react to different parts of the EM spectrum. For instance, there are birds that can see light that is either infrared or ultraviolet.

Now, let’s consider the following scenario:

There is a species of intelligent social alien (say, species#1) that evolved on another planet, and they have also a different visual system, which is stimulated by light in a way that is considerably different from the way in which our human visual system is stimulated.

Species#1 has a language shared by their entire species, and words that they use in a way similar to the way we use color words in our human languages. Let’s call that species#1-color language. [1 7 ]

If they came to Earth, humans with a normal visual system, and under normal conditions, would see all of them as almost entirely red – and basically the same tone of red.

However, they do not see themselves in that manner.

Instead, they classify themselves into six different categories, species#1-color wise, which is in accordance to their perception, and they also perceive each of the individuals in each of the categories to have large parts of its body of one species#1-color, other parts of another species#1-color, and so on. On the other hand, if they were to visit Earth, they would see no difference between the way ‘go’ and ‘stop’ traffic lights look, since they’re not different species#1-color.

Of course, if aliens like that do exist, it’s clear what those aliens do or believe has nothing to do with, say, whether Bob ran a red traffic light, or the light was green. In fact, whether a traffic light was green or red is a matter of fact, and it does not depend on anyone’s opinion, whether human or alien from another planet.

Now, if some of those aliens came to Earth and they were to claim, in English, that traffic lights for ‘stop’ and ‘go’ are the same color, they would be making a false statement.

And if some humans were to claim, in their language, that all individuals of species#1 are the same species#1-color, those humans would be making a false claim as well, regardless of what they believe.

On the other hand, humans would be correct in claiming, in English, that all individuals of species#1 that they encountered are almost entirely red, and essentially all the same tone of red.

The crucial point is that such aliens might make objective species#1-color statements – the description of the scenario supports that they do, so let’s say they do -, whereas humans make objective color statements. The objectivity of color statements, or the objectivity of species#1-color statements, in the sense of ‘objective’ in the context of the metaethical argument, is not affected, threatened or otherwise debunked by the existence of such different species.

Moreover, objective color does exist, and in the scenario, so does objective species#1-color.

While species#1, with those particular traits, is a hypothetical one, it seems that for all we know, there may well be elsewhere in the universe a species of intelligent, social aliens with a visual system very different from the human visual system, color-like language, true objective color-like statements, and so on.

Now, there might be some philosophical sense of ‘objective’ in which, perhaps, such scenarios provide evidence against the existence of objective color. I will not take a stance on that, but rather point out that that would not be relevant in this context, because that’s not the sense of ‘objective’ we’re talking about.

An alternative scenario would be one involving human genetic engineering.

For instance, using computers and other technologies, we can set up a camera or set of cameras that take video footage but change the way it looks to us, color-wise. We can use cameras sensitive to light in a spectrum of around, say, from 440nm to 750nm instead of the visible spectrum from 390nm to 700nm, and then set it up so that on a computer screen, the cameras’ spectrum is translated into our visible spectrum assigning color according to some rules of our choosing, so that the images on screen look to us differently from the way they would look to us if we were to look at the objects directly under some usual daylight conditions, and instead they look like the way some entities similar to us but with a different visual system that responds to a spectrum from 440nm to 750nm might see them.[1 8 ]

In this case, we would see things like that due to a combination of the cameras, the computer, etc., and our eyes – which we would use to look at the computer screen -, without any genetic engineering, but in a distant future, using genetic engineering, perhaps someone might modify a group of humans to see things in that way. [1 9 ]

Let’s say that those genetically modified humans (say, GM-humans#1) lived together in a country (or in a spaceship, etc.), and developed words similar to our color words, but that would be useful to people with their visual system. In that scenario, they could make objectively true GM-human#1-color statements, etc., as we can make objectively true color statements.

So, objective color does exist[2 0 ], and in that scenario, so does objective GM-human#1-color.

At this point, someone might perhaps wonder whether GM-humans#1, or species#1, etc., are nomologically possible. I would say that at least, we do not know that they’re not, nor would we be justified, given what we know about biology at this point, to claim that they are not. Moreover, in the case of aliens with different visual systems – not species#1, but any other species that may be used for a similar argument -, we are not even justified in saying that they are not actual.

In other words, we do not have grounds for claiming that there are no intelligent and social alien species with language and with different visual systems like the ones suggested above in the entire universe.

7. Beauty, exobiology, and Neanderthals.

In his reply to an objection to the metaethical argument, Craig claims that the evidence supports that beauty is objective. [r 1 2] In a different context, he claims that there are objective aesthetic judgments[r1 3], and clearly indicates that he also believes that some statements of the form ‘X is beautiful’ are true.

So, let’s consider the hypothesis that objective beauty does exist, which means the same as the conjunction of the following two conditions, or is at least equivalent to it.

OB1: Statements – or judgments, or whatever one calls them – of the form ‘X is beautiful’ are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that a specific landscape is beautiful, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether that landscape is beautiful is a matter of fact, not a matter of opinion.

OB2: Some statements of the form ‘X is beautiful’, are true, where X is an actual object.

One detail here is that one might wonder whether abstract objects would count in OB2. However, addressing that issue would be unnecessary for the purposes of the arguments at hand, so I will exclude them just to simplify.

That aside, personally, I think that OB2 is true, and that plausibly in most cases we encounter, there is an objective fact of the matter as to whether an object is beautiful. However, I think there are cases in which that is not so, and the proportion of cases in our experience might not be small enough for me to accept OB1, even if one accepts a usual degree of tolerance. So, I would take no stance on whether OB1 is true.

However, there is no need to get into that in this context, so let’s grant that the conjunction of OB1 and OB2 is true. So, objective beauty does exist.

Let’s consider the following scenario:

On another planet, there evolved intelligent, social aliens – say, species#2 -, which are at least as ugly as any of the animals on those lists. Also, they evolved in dark, humid cave environments, which they find particularly appealing. They find open spaces – like savannahs if they saw one -, particularly unappealing. Such aliens have language that is akin to our language about beauty in the way they use it, etc., but associated with very different things. [2 1 ]

After reflection on the scenario, it seems intuitively clear to me that whether some aliens evolved differently – species#2 or something like that[2 2] -, has no bearing on whether the truth-value of aesthetic judgments depends on what people believe, or on whether such judgments are true.

Intuitively, what aliens do or perceive does nothing to undermine our sense of beauty, or our judgments of beauty, or the belief that objective beauty does exist, etc.

We would still make objectively true judgments of beauty, and the aliens would make objectively true judgments of species#2-beauty, etc.
Granted, someone might insist that if such aliens existed, then there would be no objective beauty.

But if someone made such a claim, he would have the burden of backing it up.

Moreover, if that’s the position of someone who also maintains that objective beauty does exist, then he’s committed to a claim about exobiology ranging across the whole universe, namely the claim that aliens like species#2 do not exist. [2 2] But why should anyone believe that they have such knowledge about exobiology? And why should we believe that he does?

Let’s leave exobiology aside for now, and turn to cloning. What if someone managed to clone Neanderthals?

It’s difficult, but with sufficient technology, it might just be doable. Granted, it may be immoral, but that does not mean it’s nomologically impossible, or even that it won’t happen.

So, let’s say that someone clones some Neanderthals.

It seems clear that Neanderthals are uglier, compared to modern humans, and all other things equal. For instance, young, healthy Neanderthals are uglier than young, healthy, modern humans, etc.

In fact, all Neanderthals are quite ugly, even though some Neanderthals are uglier than other Neanderthals.

On the other hand, on average, modern humans are Neanderthal-uglier than Neanderthals. In other words, Neanderthals are more Neanderthal-beautiful than modern humans are, all other things equal – that’s almost certain.

In addition, there might be differences between beauty and Neanderthal-beauty when it comes to, say, landscapes. I take no stance on that particular issue, but in any case, none of that would seem to be a problem for objective beauty.

8. Craig’s take on beauty and evolution. More extraterrestrials, and more Neanderthals.

In an objection to the metaethical argument, the example of beauty was used in an attempt to undermine the second premise, or at least the reliability of our sense of right and wrong, etc.

In his reply, Craig makes a number of claims and implication about beauty and evolution [r12], including the implication that if our sense of beauty comes from evolution without God, then that undermines either our belief that objective beauty exists, or at least the reliability of our appreciation of it.

As the examples given in the previous section illustrate, there seems to be no good reason to believe so.

On the other hand, the scenarios given in the previous section would seem to be a serious challenge to Craig’s stance on beauty.

For instance, it seems that Craig’s views entail that if something like species#2 does exist, either there is no objective beauty, or at the very least one of the two species – i.e., species#2 or humans – has a massively unreliable sense of beauty, and there would appear to be no good reason to trust the human one.

Yet, Craig’s position is that objective beauty does exist, and that the human sense of beauty is at least generally reliable. But that seems to commit Craig to a wide-ranging claim about exobiology, namely that nothing like species#2 evolved anywhere in the universe. [2 2]

I suppose someone might suggest that if God exists, there is no undermining even if such aliens do exist, and our sense of beauty is still generally reliable. But that would not work, because as long as Craig rejects the distinction between beauty judgments and species#2-beauty judgments, then either they would have a massively unreliable sense of beauty – rather than a generally reliable sense of species#2-beauty -, or we would have a massively unreliable sense of beauty.

But that would seem to take away any good reason to think we got the right sense of beauty even assuming God existed – why think God preferred us, rather than the aliens he also created?

In any case, it seems that one way or another, Craig’s position commits him to a universe-wide exobiology claim on the matter: either a claim that nothing like species#2 exists in the whole universe, or that if they do, they have a massively unreliable sense.

But moreover, Craig claims that it would be somehow anomalous, given an evolutionary account of our sense of beauty – without God – that we would appreciate beauty in other species, not just humans, or in humans of the same sex, etc. But his claims are on evolution mistaken, for a number of reasons, such as the following ones:

1. In the case of humans of the same sex, there may well have been an advantage – i.e., on balance, in the ancestral environment, it was conducive to reproductive success – to being able to make assessments of beauty of individuals of the same sex, given that potential mates of the opposite-sex were able make them. That would allow both men and women to compare themselves with potential competitors, predict behavior of potential mates, etc.

However, that factor is surely not the whole story, even if it may have played some role.

There are more general considerations about evolution that Craig is missing, like the following point:

2. Adaptations that were on balance conducive to reproductive success in the ancestral environment often had side effects that were not relevant to fitness and sometimes, or side effects that, on their own, would have been detrimental in terms of fitness, but resulted from the same gene mutations that conferred advantages that offset those disadvantages in that environment, etc.

Also, in some cases, there may be some traits that would be on balance slightly detrimental to reproductive success today, but were not so in the past, etc.

As examples of some of the traits in some of the previous categories, we find some artificial sweeteners sweet but there was no advantage to that. It’s a consequence of adaptations to other things. Also, there was no advantage in finding lethal antifreeze tasty, though there was no disadvantage in an environment with no antifreeze, etc. The same goes for other animals. For instance, many animals find ethanol very appealing, and so on.

I don’t know all of the causes we appreciate beauty in other animals. It may have something to do with our ability to perceive good health, and perhaps a number of other traits, but that’s speculative.

However, in any case, the point here is that Craig’s objection is not warranted. More precisely, his claims are based on a misunderstanding of present-day evolutionary biology, which result in mistaken predictions about what would result from an evolutionary process that occurs by the mechanisms posited by modern science and nothing else. On that note:

3. Craig’s objection is a bold claim or implication about human biology and evolution.

Essentially, he’s implying that the evolutionary process, only by the mechanisms posited by science – or very similar ones – would not result in the mental traits that we observe in humans.

On that note, he clearly implies that if our appraisal of beauty in the opposite sex were the result of natural selection, a “young woman with a big nose and a harelip” would look as beautiful to him as a fashion model.

There are several objections one may raise to that claim, but I will focus on just one, which is in any case decisive: Craig bases his assessment on the false assumption that non-theistic evolution would result in sexual attraction being equal given equal potential to be fertile. But that isn’t how natural selection works. Surely, it's not just the number of potential offspring that counts, but many other factors, such as that potential offspring's chances in turn to reproduce, which depend also on a number of different factors, including their ability to attract mates with genes that would give them a good chance of reproductive success, and so on.

So, there are many causes for the overall, on-average attractiveness of some traits, such as the fact that they may be cues to good health, etc., or sexual selection, and a long etc.

Craig seems to be very confused about natural selection, and generally about the mechanisms of evolution proposed by present-day science.

In any case, and with respect to sexual attraction, we may, of course, also take a look at the issue of potential or actual variations from species to species. We may consider the case of, say, warthogs. They’re ugly. Yet, they do not seem to be particularly attracted to much more beautiful animals, but rather, to other ugly warthogs. The same seems to go for a ll other ugly animals. Of course, that’s not remotely surprising given evolution by the mechanisms accepted by present-day science, or very similar ones.

That aside, we may consider another example that poses a serious challenge to Craig’s views on beauty and evolution: Neanderthals. What if someone were to bring them back, by means of cloning?

Then, one may consider the statement:

HN: Adult, healthy modern humans are more beautiful than adult, healthy Neanderthals.

Since objective beauty does exist – as Craig maintains, and as I concede for the sake of the argument -, there is an objective fact of the matter as to whether HN is true. Since – as Craig implicitly indicates – our sense of beauty is generally reliable, chances are that our sense of beauty is correct and that HN is true.

But let’s say that we ask Neanderthals. Assuming no distinction between beauty and Neanderthal-beauty – a distinction that Craig’s position seems incompatible with -, they’re almost certainly going to make the wrong assessment, namely that HN is false. But why should one think that Neanderthals had such a defective sense of beauty? How do we know we didn’t get the defective sense in the first place?

Perhaps, someone might claim that Neanderthals would not be able to learn a sufficiently sophisticated language and would not understand the question. But that would be a claim about Neanderthal biology that would be in need of defending. Moreover, it’s hard to see how that would even matter.

Also, someone might suggest that there are [or were, depending on the case] beautiful modern humans, and beautiful Neanderthals, but not comparable, beauty-wise. But the problems is that our sense of beauty tells us that Neanderthals were comparable with modern humans, and they were uglier, even assuming in both cases young and healthy individuals.

In light of the previous considerations, we can tell that Craig's stances on beauty, evolution, and the relations between them are all mistaken.

9. Morality, evolution, and evolutionary counterfactuals.

In support of the first premise, Craig argues against what he derogatorily calls ‘herd morality’ - humans never lived in herds, and neither did their ape ancestors, but never mind that -, and makes a number of claims: [r1 1 ]

In particular, he claims that if “moral values are just the product of biological and social evolution”, those “really aren’t moral values”, but “simply conditions under which the human species will flourish”.

Leaving aside derogatory words like ‘just’, ‘simply’ and several others he uses in that context, like ‘herd morality’, etc., one may reply that:

1.a. The word ‘social’ seems to indicate something other than biological evolution, and might suggest that the view he’s arguing against supports some form of cultural relativism. That does not have to be the case, since a non-theist may well hold that the sense or right and wrong and good and evil[1 2 ] is the product of biological evolution, even if it needs some environment to develop – like all of our traits do -, and even if the ancestral environment that led to their evolution included prominently a social environment.

1.b. The sense of right and wrong and of good and evil[1 2 ] would not be on such evolutionary accounts “conditions in which the human species will flourish”, but rather, the result of an adaptation or combination of adaptations that were on balance conducive to reproductive success in the ancestral environment. While, on those accounts, humans will develop a certain language – like moral language – due to some features of human biology – and, perhaps, some aliens would develop a different language -, evolution only plays a role in the etiology of those features, and so only indirectly in the development of the language.

But none of the above tells us that our judgments like ‘the Holocaust was immoral’ would not be objective judgments, or that all judgments like ‘X is immoral’, for an actual X, would be false, on such a non-theistic evolutionary view.

1.c. It is true that some accounts of this kind might also posit that there's always be a moral obligation to behave in a way conducive to human flourishing. Those accounts would be mistaken, as we can tell by means of setting up a number of scenarios in thought experiments in which it's clearly not morally obligatory to behave in a way conducive to human flourishing, and even scenarios in which it's morally obligatory to behave in ways not conducive to human flourishing at all.

However, it is not the case that, in general, according to these kinds of evolutionary accounts – which actually use evolution as a source of our mental faculties, but not as a means of setting up moral obligations, which depend on mental faculties, however we got to have them -, there would always be a moral obligation to behave in a way conducive to human flourishing.

1.d. Craig claims that a sociopath who rejects what he calls the ‘herd morality’, would not be doing anything morally wrong on that view. But for that matter, someone might say that if our ‘herd color’ evolved, then there would be nothing red about ‘stop’ traffic lights on a similar non-theistic evolutionary view, or that on such a view, there would be nothing ugly about warthogs if our 'herd beauty' evolved. Upon reflection, I see no good reason to believe either Craig’s claim about the sociopath, or the parallel claims about color or ugliness, or that the mistaken accounts that posit that there is always a moral obligation to act in a way conducive to human flourishing have any problem with moral objectivity. Rather, the problem with such accounts is that that they make a false universal first-order ethical claim. For that matter, theories involving ethical and metaethical claims that posit that there is always a moral obligation to believe that Jesus is lord and savior, at least for adult humans who have read the Gospel and some others, also make a false universal first-order ethical claim, but do not have any problem with objectivity.

Also, in addition to the above, Craig maintains that if we could rewind evolution, then a different entity with a considerably different set of moral values and duties might have resulted instead of humans, and if so, neither their values nor ours would be right. However, Craig does not do anything to establish his claims. For instance:

2.a. Those entities would have had to resolve problems similar to the ones our ancestors faced, so maybe there would not be such variation after all, if the initial conditions are the same. [23]

2.b. Craig has not explained why, assuming that there might be such differences if we could rewind evolution, that would provide any evidence in support of the claim that without God, objective moral values and duties do not exist.

For example, we may consider the same argument about rewinding evolution in the case of color vision. But as the relevantly similar example of species#1 illustrates, that would provide no evidence against the existence of objective color without God. Craig provides no argument in this context in support of the claim that it would make a difference in the case of morality; rather, he seems to assume that the example of rewinding evolution helps his case.

10. Morality and exobiology I.

Craig also considers some examples of aliens from the Andromeda galaxy who come to Earth and attack, either raping and killing humans, or eating humans. [2 4 ]

Now, Craig has not shown that evolution would not be convergent in the case of morality, at least for social entities. But we may grant that there is such difference.

So, let’s say for the sake of the argument that the aliens in question – say, species#3 – evolved very differently, but they do have a sense more or less similar to our sense of right and wrong, and to our sense of good and bad[1 2 ]; those faculties are similar to ours in the way using them normally feels to them, in the role that they play in discussions about what to do in their society, etc., but associated with different behaviors and/or situations, entities, etc.

In particular, their sense of species#3-right and species#3-wrong does not associate any negative evaluation with their actions of invading the Earth and eating humans.

So, let’s say that they invade the Earth and eat humans. [2 5 ]

Speaking of those scenarios, Craig claims:

Craig:

“I think this extraterrestrial illustration is a very powerful illustration that, in the absence of God, human morality isn’t objective. It has no more claim to be objective than some extraterrestrial alien morality.”

Before I address those claims, and on a terminological note, in the cases of color, and beauty, I used the terms “color” - rather than “human color” - as opposed to “species#1-color”, and “beauty” - rather than “human beauty” - as opposed to “species#2-beauty”, respectively. Similarly, I will here use “morality” - rather than “human morality”, which Craig uses-, as opposed to “species#3-morality”[1 7 ], respectively. But that’s merely terminology.

So, in the excerpt quoted above, Craig makes at least two claims.

1. A claim that the extraterrestrial examples show that, without God, objective moral values and duties do not exist.

2. A claim that without God, human morality has no more claim to be objective than some extraterrestrial morality – with the terminology I introduce above, that would be morality and species#3-morality in the example in question, respectively.

What to make of those claims?

In the case of claim 1., Craig does not explain whether that alien example is supposed to show that OMVD1 is not true without God, or that OMVD2 is not true.

But in any event, Craig seems to provide no reason to suspect that objective moral values and duties would not exist, in the sense of ‘objective’ used by Craig in his metaethical argument, just because of the alien example.

To be clear, I’m not claiming that there is no technical sense of ‘objective’ used in philosophy, in which the alien example would not show that morality would not be objective. But the relevant point here is that the alien examples do not show that without God, say, whether the Holocaust was morally wrong has anything to do with what anyone – human or alien – believes and so, in the sense of ‘objective’ relevant to Craig’s metaethical argument.

If someone claimed otherwise, they would have to argue that point. But it’s not illustrated by the example, as Craig presents it.

As for claim 2., assuming sufficient similarity between our moral sense and species#3-moral sense in terms of how they feel it, work, etc. - except that associated with different behaviors, situations, etc. -, then claim 2. seems to be plausibly true, just as species#1-color and species#2-beauty are also objective. But that does nothing to establish the first premise.

Granted, someone might claim that the case of morality is relevantly different from the case of color and/or from the case of beauty, etc., but that would have to be argued for.

In particular, someone might claim that the difference is that judgments of beauty, and also moral judgments, are evaluative judgments, but color judgments are not. However, in that case, they would have to explain in what sense and/or how there is an evaluative component that is relevant to the matter of objectivity, in the sense in which Craig uses the word ‘objective’. The burden would be on the claimant.

Also, someone might claim – for instance – that the meaning of moral terms is such that moral judgments have ontological commitments incompatible with the existence of something like species#3-morality, and so that if such species, or a relevantly similar one, existed, then all judgments of the form ‘X is immoral’, ‘X is morally good’, etc., would be objective but all false, or some other variant.

However, that would be a matter of moral semantics that would have to be argued for as well. As it stands, upon reflection, I see no good reason to think that that is so.

On that note, let’s consider the following scenario:

Millions of years into the future, humans or post-human beings[1 4 ] make contact with intelligent social aliens. Humans or post-humans are more advanced, so even though the aliens turn out to be hostile, they can be handled.

Now, after studying the aliens, [post]-humans figure out that the aliens do not have a sense of right and wrong and good or bad but something more or less similar to them in the way outlined earlier.

I doubt that millions of years into the future, [post]-human beings would believe that God exists, but regardless of what they believe on that matter, it seems intuitively clear that they would not and should not consider their findings about the aliens evidence supporting the conclusion that the Holocaust was not immoral.

In fact, the immorality of the Holocaust and exobiology seem to be orthogonal matters, regardless of theistic belief.

At least, upon reflection, I do not see any good reason to believe otherwise, and Craig does not seem to provide any.

All that aside, and as an alternative objection, a theist might raise the issue of genetic engineering.

What if someone made genetically modified humans, say GM-humans#3, with a slightly different sense or senses, instead of a sense of right and wrong, or good or evil?

They wouldn’t be something completely alien like species#3, but something very similar to humans, but not quite the same.

A problem I see with that scenario is that such modifications might result in conflicts between different parts of the mind, unless perhaps the changes are massive. But leaving that aside, I will address the matter of entities that are very similar to humans but not quite the same in the section dedicated to Moral Twin Earth scenarios. The genetic engineering cases, assuming they’re nomologically possible, would be handled in the same fashion.

11. Morality and exobiology II. Some objections.

In the previous section, I argued that Craig’s argument from extraterrestrials fails to support his conclusion. In this section, I will consider some objections, and offer some replies, but before I go on, and to be clear, I do not claim that Craig would raise the same objections or similar ones.

Objection 11.1:

If something like species#3 existed, invaded Earth and ate humans, they would clearly be acting immorally. Even science fiction examples illustrate that aliens that attack like that are behaving immorally. In other words, they ought to refrain from eating humans. But the previous analysis indicates that, without God, they would not be acting immorally, contradicting our moral intuitions.

Reply:

1. The previous analysis is not committed to the question of whether species#3 aliens who invade the Earth and attack humans would be acting immorally. It’s compatible with at least the following two views:

1.a. The aliens are acting immorally, but they do not have a sense that allows them to know that. They are not acting species#3-immorally, though.

1.b. The aliens are not be acting immorally. In fact, nothing that they do is immoral. Some of the things they do might be species#3-immoral, but invading Earth and eating humans are not among them.

I do not think that 1.a. is the right view, though, but other non-theists might not have any views on the matter, and that does not seem to be a problem either way.

Of course, even if 1.b. is the proper interpretation – which I find plausibly correct -, that does not mean that the situation consisting in aliens of species#3 eating humans is not a bad situation. In the sense in which, say, the situations consisting in crocodiles eating young children or young children slowly dying of cancer are bad situations – though not morally bad, as far as I can tell, if no moral agent is responsible -, we may properly say the event or situation consisting of species#3 aliens eating humans is a bad situation or event.

On the other hand, in the scenario, plausibly the event or situation consisting in aliens of species#3 eating humans is not a species#3-bad situation of event – assuming they have something akin to that good/bad distinction but associated with different situations, etc. -, and perhaps eating humans is even a species#3-good situation or event, though that would not imply that eating humans is species#3-morally-obligatory.

2. The claim that it’s clear that those aliens would be acting immorally is implausible. In fact, while the previous analysis is not committed to the question of whether species#3 aliens who invade the Earth and eat humans would be acting immorally, after reflection, it seems plausible that they wouldn’t be acting immorally. [2 6 ] They do not have a sense of right and wrong that would tell them it’s immoral, even under ideal conditions. They do not seem to have the right kind of mind to behave immorally, as far as I can tell.

As for science fiction aliens, they’re often made in the writers’ image, so to speak.

In other words, even though those fictional agents [fictionally] evolved on different planets, where they evolved is not the issue, but rather, what kind of mind they have.

For that matter, many of those fictional aliens feel sexual attraction towards humans, because writers usually make them with human-like minds, even though in case of contact we should expect aliens not to be – normally, at least – sexually attracted towards humans.

On the other hand, someone may have come up with some fictional advanced aliens that do not have moral obligations – some fictions may not be specific enough to determine that.

This is not to say that real aliens would never have moral obligations. If the universe is sufficiently big, there might be aliens that evolved with minds like ours, including a sense of right and wrong and good or evil – rather than the species#3 analogue, or anything like it.

3. That aside, according to Craig, God exists and a creature’s moral obligations are God’s commands. If so, then if species#3 evolved on some planet – guided by God, perhaps -, then it seems that those aliens would not have a moral obligation not to eat humans, since they do not have a sense of right and wrong that tells them not to eat humans, and then plausibly God would have not commanded them not to eat humans.

So, it seems that even on Craig’s theory, if those aliens existed, in any case they would not have a moral obligation not to eat humans. But then, why would it be a problem for non-theists if, on non-theism, those aliens plausibly would not have a moral obligation not to eat humans?

Granted, a theist might claim that such aliens or relevantly similar ones are metaphysically impossible, but that in particular entails that those aliens, or any relevantly similar aliens, do not exist anywhere in the universe, which a wide-ranging claim about exobiology, encompassing the entire universe. Why should we believe that a theist has such knowledge of exobiology?

Objection 11.2:

On that kind of evolutionary account, who is to say that it’s not morally obligatory for individuals of species#3 or a relevantly similar species to invade the Earth and eat humans because they like the way humans taste?

Reply:

We can tell by our sense of right and wrong that that would not be a moral obligation for them, or for anyone. That would not be morality, and we can tell because we’re the ones with the sense of right and wrong, not aliens of species#3. Those aliens have a sense of species#3-right and species#3-wrong, but that’s a different thing, like species#1-color vision is not color vision, and a sense of species#2-beauty is not a sense of beauty.

That aside, it’s not at all clear to me that natural selection in our universe would favor something as different from our moral sense as what is proposed here – though almost certainly it would if the universe has infinitely many planets.

I’m just granting that for the sake of the argument. But even if they did have a species#3-moral-obligation to invade the Earth and eat humans, that is not a moral obligation to do so.

For example, let’s consider the following scenario as an analogy:

Let’s say – which is very plausible – that humans have a species-wide very similar perception of colors – i.e., no inverted color spectrum, etc. Let’s also stipulate that there is an intelligent social alien species, species#4, which has a species-wide perception of species#4-color[1 4 ], and that in particular, they normally perceive objects that emit light in wavelengths between 320nm and 370nm in a way very similar to the way in which humans normally perceive objects that emit light in wavelengths between 520nm and 570nm, in other words, like we perceive objects that emit green light. Something similar goes for reflective properties – i.e., objects that reflect some wavelengths under such-and-such conditions, instead of objects that emits them.

Let O4 be an object that emits light in wavelengths between 335nm and 342nm, and no other wavelengths.

In that scenario, O4 emits species#4-green light, but it does not emit green light. Species#4-green is not green at all. And similarly, having a species#3-moral-obligation is not the same as having a moral obligation, just as, say, being species#4-green is not the same as being green

Objection 11.3

On that kind of evolutionary account, let’s say that some species#3 aliens have a species#3-moral-obligation to invade the Earth and eat humans. Should they do it?

Reply:

It’s not the case that they morally should do it.

However, it may well be that they rationally should attack and eat humans – it depends on the case.

It might be that they have reasons to invade the Earth and eat humans, but have no reasons not to – even if it is immoral of them to attack (which seems implausible to me for the reasons I’ve been explaining, but let’s say so), morality does not give them any reasons.

Also, there might be some term in their language – let’s call it ‘species#3-should’ – that is not a synonym of any term that we use, but which plays a function in their language somewhat similar to the function of the moral “should” among us, and maybe it is the case that they species#3-should invade, etc.

However, just as ‘species#4-green’ does not mean the same as ‘green’, and ‘species#2-beautiful’ does not mean the same as ‘beautiful’, also ‘species#3-should’ does not mean the same as ‘should’, in the moral sense of ‘should’, nor does it mean the same as any term in English or in any other human language. But to be clear, when I say ‘species#3-should’, I'm not saying that the meaning of the term has a built-in reference to the species. Rather, I’m using ‘species#3' merely to indicate that it’s a term used by aliens of that particular species, and which has no synonyms in English or other human languages.

Also, as I pointed out earlier, I’m assuming at this particular point and only for the sake of the argument that there might be that kind of huge divergence between different species, even if they’re all intelligent social species that evolved without any designer. Whether that is actually the case is an exobiology question whose answer I don’t know, so I’m taking no stance. If a theist is taking a stance, he should defend his exobiology claim.

Objection 11.4.

There can be no such thing as ‘species#3-should’. If there were a term ‘should’ in the alien language that plays a function in their language somewhat similar to the function of ‘should’ in the moral sense of ‘should’ among humans, then it actually means the same. The same for other terms.

Reply:

Going by my intuitive grasp of the terms, it seems very implausible to me that the meaning of our moral terms is like that.

However, assuming for the sake of the argument that the terms would mean the same as this rejoinder holds – which I reject -, and also assuming for the sake of the argument that there is or there might be such huge variations between species – a matter on which I take no stance -, and further assuming for the sake of the argument that species#3 actually evolved, then there would seem to be two distinct possibilities:

a. Species#3 aliens plausibly would have a sense of right and wrong generally reliable as a means of ascertaining the moral obligations of individuals of their species, whereas we humans have a sense of right and wrong generally reliable as a means of ascertaining the moral obligations of humans.

This might result in moral obligations to eat us, etc., which would be counterintuitive.

b. A moral error theory is true.

But I have no good reason to believe our moral terms are like that at all.

On the other hand, a theist raising objections 11.3 and 11.4 would seem to be committed to a position that entails the disjunction of the following options:

c. Aliens like that do not exist.

d. Aliens like that are have a vastly unreliable sense of right and wrong.

e. Aliens like that have a moral obligation to eat us if they find us in a situation like the one outlined above, or similar obligations.

However, c. is a statement about exobiology ranging across the whole universe, and there is no evidence for that; e. is very counterintuitive, and d. seems unjustified – why would their moral sense be the flawed one? -, and also has the following problem if the theist supports Divine Command Theory: If the moral obligations of those aliens are God’s commands, how does God give them commands if they have such a vastly unreliable sense of right and wrong?

As I mentioned, I see no good reason to think that our moral language is like that. On the contrary, I think it’s definitely not like that at all, and the aliens in question would be talking about something else.

Objection 11.5:

Beings with a human level of intelligence or superior but with no moral obligations are metaphysically impossible, with the possible exception of God, who does not have any moral obligations because an entity’s moral obligations are God’s commands, and God is not giving binding commands to himself.

Reply:

1. Leaving aside the odd exception for God and the odd theistic metaethics, the scenario involving species#3 seems to me plausibly a case in which some beings with a human level of intelligence or superior would not have moral obligations, though I need not take a stance here. A claim that either they’re metaphysically impossible or that they would have moral obligations seems to place a burden on the claimant.

2. Let’s consider the following scenario:

On a different planet, another alien species – say, species#5 – evolved. Those aliens are very advanced, and eventually, by means of a combination genetic engineering and artificial intelligence, some reckless individuals of that species make an ultra-intelligent cyborg, even far more intelligent than species#5 aliens. As it turns out, the cyborg – which is not a social being, does not care about any other being, and does not even have anything akin to the sense of right and wrong – wipes out species#5 for its own benefit, and in accordance to its values.

It then goes on to use the resources of that planetary system, and starts working towards becoming more powerful, even if that will lead it to attack and destroy other intelligent beings in other planetary systems in the future.

In that scenario, it seems even more intuitive to me that the cyborg does not have any moral obligations. Plausibly, nothing the cyborg does is morally wrong, or obligatory, or praiseworthy, or morally anything. This does not mean that the cyborg does not have reasons for acting, or that it’s not the case that it should rationally do X – for some X.

But it’s not the case that the cyborg morally ought to do X or not to do X, for any X.

That would be my assessment, but I do not need to claim that in this context. Instead, it’s enough to point out that a claim that either such a cyborg is impossible or it would have moral obligations would seem to place a burden on the claimant.

All that said, as far as I can tell a non-theist need not take a stance on whether such aliens, or the cyborg, would have moral obligations. If the cyborg did have moral obligations, it seems plausible that it often rationally ought to do what it morally ought not to do – and in any case, it would not care -, but as before, this does not seem to be a problem. The problem seems to be for those claiming otherwise.

Objection 11.6:

If our sense of right and wrong comes from evolution without God, then there are no moral obligations. How could a mindless process impose a moral obligation?

Reply:

That is a different kind of question, not directly related to the part of Craig’s argument I addressed above, but that aside:

1. The evolutionary process would not be imposing anything except in a figurative sense. Moral agents are agents with some kind of mind, however they came to exist, whether by evolution or by some other means.

In other words. as long as the evolutionary processes or any other process results in minds of the right kind, those entities might be morally good, bad, behave immorally – and hence, have moral obligations, etc. [2 7 ]

In fact, this is intuitively clear, and it illustrates another problem with Craig’s metaethical argument, namely that it has a conclusion that flies on the face of moral intuitions.

For instance, we may set up a scenario in which God does not exist, and Jack kills children just for entertainment. Intuitively, it’s clear that in that scenario, Jack would be acting immorally. The presence of God is orthogonal to it. I will get back to this point later.

2. While a theist might claim that the evolutionary process would not result in minds, or in the right kind of mind, that’s a claim that he would have to argue for.

Objection 11.7:

There cannot be such thing as species#3-morality. There is only morality. If species like species#3 evolved on different planets, then at most one intelligent social species would have the true morality, and all of the rest would be morally confused by a vastly flawed sense of right and wrong, and of good and evil. But then, what would be the reason for believing that our faculties are the reliable ones, rather than the faculties of some aliens?

Reply:

As in previous cases, whoever made such a claim would have to argue for it. As it stands, after reflection, I see no good reason to believe so.

In fact, it seems highly counterintuitive, as the contact example indicates.

Moreover, also as before, someone who maintains that we do have a reasonably reliable moral sense yet raises objection 11.7, is committed to a claim that there are no aliens relevantly similar to species#3 in the whole universe.

12. Morality and exobiology III. Moral Twin Earth.

I argued in previous sections that Craig does not establish that his alien examples support P1.

However, there are other arguments a theist might make, also involving entities with something similar to a moral sense in some ways, but not quite the same.

On that note, I think an interesting case to consider would be some of the Moral Twin Earth examples, and in particular the scenarios presented by Horgan and Timmons in papers like “New Wave Moral Realism Meets Moral Twin Earth”[r14], or “Analytical Functionalism Meets Moral Twin Earth”. [r15]

Even though Horgan and Timmons use the examples in a very different context, the scenarios raise the issue of aliens with different faculties, so I think they are plausibly relevant in the context of Craig’s metaethical argument for theism. I will focus on the examples in the second paper. [r15]

So, the authors explain that some of the differences in the “mature folk morality” of humans and Twin Earthers respectively, result from some species-wide psychological differences between Earthers (i.e., humans) and Twin Earthers. [r15]

The reference to a “mature folk morality” is made in the context of their reply to analytical moral functionalism – which a non-theist need not endorse; personally, I don’t -, but essentially the relevant point here is that some human moral judgments would differ from some moral or moral-like judgments of Twin Earthers, as a result of their different species-wide psychological makeup, and even upon careful reflection.

When presenting these scenarios, Horgan and Timmons consider two hypotheses, which essentially can be put as follows:

1. Twin-Earth-moral terms cannot be properly translated into moral terms. Their meaning is different. Twin Earthers and humans would be talking past each other if they did not realize that. The judgments made by Twin Earthers are Twin-Earther-moral judgments, which are moral-like judgments, but not actually moral judgments.

2. There is no difference in meaning, and translation between Twin Earth moral terms and human moral terms is appropriate. Twin Earthers make moral judgments, and disagreements between Earthers and Twin Earthers are genuine moral disagreements.

Horgan and Timmons claim that “by far the more natural and plausible mode of description, when one considers the Moral Twin Earth scenario, is the second”. [r15]

In defense of that claim, the authors say that reflection does not result in hermeneutical pressure to interpret the terms ‘good’ and ‘right’ used by Twin Earthers as non-translatable by their Earth counterparts in English.

I have to say that my intuitions are that the by far more natural and plausible mode of description, when one considers the Moral Twin Earth scenario, is the first, and that the differences here are not in belief or theory, but in fact in meaning.

Also, the authors suggest, as an example in support of their claim – as always, when arguing against analytic moral functionalism [r15] – that even after recognizing the difference in their respective mature folk moralities, members of each group (i.e., humans and Twin Earthers) would consider that debating with the other group in a way akin to the way humans debate moral issues with each other would be appropriate, rather than silly. But if the differences were in meaning rather than in theory or belief, the members of those groups would encounter such debate rather silly, according to the authors.

In order to analyze this claim, we needn’t assume Analytical Moral Functionalism, but only that there is a mature folk morality and a mature folk Twin-Earther-morality, which differ as a result of species-wide psychological differences between humans and Twin Earthers.

Essentially, what this means is that the human sense of right and wrong, and good or evil, and the Twin Earther mental faculties that resemble them, are similar but not the same, and are actually associated with different classes of behaviors, so the difference in some cases would persist even under ideal reflection.

Then, when it comes to cases in which it’s known that the reason Twin Earthers who speak Twin Earth English accept [or reject] a judgment in Twin English but humans reject [or accept, respectively] an orthographically identical judgment in English does stem from some species-wide psychological differences between the human moral sense and the Twin Earther similar counterpart, it would strike me as very silly to debate, unless there is a good reason, as in the following scenario:

Alice is human, and Bob is human, whereas Tom is a Twin Earther. It might be that, under some circumstances, Tom does not have a Twin-Earther-moral-obligation not to beat Bob, and he actually has some motivation to beat Bob. However, it might also be that in those circumstances, it’s either morally obligatory or at least morally praiseworthy for Alice to try to protect Bob from Tom even at the expense of confusing Tom, and she reckons the only way she has a shot at protecting Bob is by persuading Tom that it would be Twin-Earth-immoral for him to beat Bob – even though it would not be so, and Alice knows it or at least suspects so -, and she properly reckons Tom probably can be confounded by means of some clever Twin-Earther-moral-like pseudo-argumentation.

So, in a situation like that, it seems to me it would not be silly for Alice to engage in such a debate, though it would be extremely silly for Tom – unless, perhaps, Tom also has some reason to engage Alice, attempting to confuse her.

However, if the intent is not to confuse one’s interlocutor – which may be in some cases morally justified for humans, or Twin-Earther-morally-justified for Twin Earthers -, as I said I would find such debate very silly, given that it’s known that upon ideal reflection, one’s interlocutor will not agree with a judgment in their language that is orthographically identical to the judgment one makes in one’s language.

Of course, if the human moral sense and the Twin Earthers’ Twin-Earther-moral sense are extremely similar, and the vast majority of cases of divergence in orthographically identical judgments result from errors committed by one or more of the humans or the Twin Earthers involved, rather than from species-wide psychological differences, then unless there are good reasons to think that a particular divergence results from species-wide psychological differences, it may well make sense to debate, at least at first. But that’s not the issue.

So, while different people will assess the matter by their own intuitions, of course, in my intuitive assessment, the most plausible answer to the Twin Earth scenario by far is:

Answer A1:

A1.i.. Twin Earthers are talking about Twin-Earther-right and Twin-Earther-wrong, etc., rather than about right and wrong.

A1.ii. Twin Earthers have no moral obligations, though they have Twin-Earther-moral-obligations.

Another answer to the Twin Earth scenario might be:

Answer A2:

A2.i. Twin Earthers have moral obligations.

A2.ii. Twin Earthers have a sense of right and wrong generally reliable as a means of assessing what’s right or wrong for a Twin Earther – which is psychologically different from a human being – to do, whereas humans have a sense of right and wrong generally reliable as a means of assessing what’s right or wrong for a human being to do.

A2.iii. While there is apparent disagreement on the morality of behavior in superficially similar scenarios that do not factor in the species-based differences in the psychologies of the individuals involved in those scenarios, there is no disagreement upon reflection on what the moral obligations of individuals of each species are.

In fact, upon reflection, a human should generally rely on the normal sense of right and wrong of Twin Earthers in order to assess the moral obligations of Twin Earthers, and vice versa, unless they have other means, like sufficiently powerful computers that can do the job as well.

Answer A1 holds that 1. is true, but it’s different from the way in which Horgan and Timmons seem to analyze that possibility, in the sense that (by A1.ii.) humans would upon reflection converge – if they make no mistakes – to the assessment that Twin Earthers do not have any moral obligations, though they have a usually considerably close analogue.

Answer A2 is that the meaning of the terms is the same, but denies that there is disagreement in the mature folk moralities, so this is implicitly denying the scenario that Horgan and Timmons propose, by saying that the folk moralities would not be different because they would factor in the species-based differences in the psychological makeup of the agents, and that will yield different verdicts about what an agent morally ought to do.

However, I think A2 is vulnerable to a number of potential difficulties, like:

a. Someone might suggest that there are situations in which the death of some humans is a worse result than the death of the same number of Twin Earthers because of some species-specific traits of human psychology – all other things equal.

If so, then if a human is in a position to save only one group of humans or one group of Twin Earthers, perhaps she has a moral obligation to save the humans, bringing about the better result – i.e., the group of humans survive. But would the Twin Earthers have a moral obligation to bring about the worse result, namely that the Twin Earthers survive? Or should they save the humans?

There are potential answers to this:

One option – the best by far I think – is that if 2 is true, the survival of the group of humans is neither better nor worse than the survival of the group of Twin Earthers just because of their species. Another potential answer – but I think much weaker assuming 2 – would be to separate human-good and Twin-Earth-good (even if not right and wrong), and say that Twin Earthers would have a moral obligation in that case to bring about the worse result, but not the Twin-Earth-worse result.

b. What if it’s not Twin Earthers but far more different aliens, like species#3?

In the earlier example of species#3, it’s not stipulated that they have a species#3-moral-obligation to invade the Earth and eat humans. But perhaps, the condition of a species#3-moral-obligation to do so might be added, whereas it’s not at all plausible that a moral obligation to attack the Earth and eat humans, might be added. So, it’s intuitively clear that in the case of species#3, they’re talking about different things, and species#3-morality is not the same as morality.

It might be argued that that suggests that the similarities between humans and Twin Earthers in the Twin Earth scenario give a mistaken impression – to some humans, at least – that the meaning of the orthographically identical terms is the same, even if it’s not.

A potential reply in defense of A2 here would be that Twin Earthers may be sufficiently similar to humans to have moral obligations, etc., even though they have somewhat different moral obligations sometimes even in superficially identical situations – i.e., situations that appear identical without factoring in some distinct features of human and Twin Earther psychology -, but on the other hand, species#3 aliens are too different from humans, and do not have moral obligations, but species#3-moral obligations, etc.

An alternative reply in defense of A2 might be that it applies to species#3 as well, and that perhaps they do have moral obligations like invade the Earth and eat us – even if they’re not at any risk -, or things like that; but that seems even more implausible.

c. On a different note, it might be stipulated that correct Twin Earth morality assessments do not take into consideration the psychological differences between humans and Twin Earther.

A potential reply here is that unless Twin Earther mature folk morality converges to taking into account the psychological differences between humans and Twin Earthers, then it’s not morality, but Twin-Earth-morality, and we would be in the case of answer A1.[2 8 ]

Intuitively, as I said, I find A1 a lot more plausible.

Still, in any case, neither answer seems to provide anything that could support a metaethical theistic argument.

13. Ontological foundation or grounding.

According to Craig, moral values are grounded in God’s character, and moral duties are also ontologically grounded in God because God’s commands constitute our duties. Without God – Craig maintains – there would be no such foundation or grounding.

In this section, I will assess the matter, and raise some objections.

13.1. Craig’s foundational conditions.

In different documents, Craig makes a number of statements that we may use to approach his concept of ontological foundation, and also infer some of the conditions he claims or implies are required for an ontological foundation of moral values and duties.

On that note, we have the following pieces of information:

a. Craig rejects Kagan’s ‘ideal observer’ proposal[r17] on different grounds, one of which is that that “non-realities” cannot be the ontological foundations of actual things. So, in particular, this rules out ideal observers and the like as foundations of objective moral values and duties.

b. In a different post explaining part of his metaethical argument, Craig claims that some qualities are good because they’re found in God’s nature. [r18]

c. Craig rejects abstracts objects as a foundation of morality, or generally it seems as foundations of any concrete things, given his take on abstracta.[r1 9 ]

d. In one of his books[r20], Craig says that the kind of grounding (i.e., foundation) he's talking about is what Mark Murphy calls “informative identification” [r21], and Craig proposes an informative identification of moral values with God’s character – or at least some aspects of it -, and moral duties with God’s commands.

As illustrations of what kind of foundational account an informative identification is, Murphy gives the examples of the identification of water with H2O, and of heat with molecular motion.

e. In his debate with Sam Harris[r5], Craig makes it clear that when he's asking about the foundation about objective moral values and duties, one of the things he's asking is what makes those actions good and evil, right or wrong.

f. Furthermore, Craig goes on to provide an example: he says that moral values are determined by the paradigm of God’s character in a manner akin to the way in which in the past, a meter was defined in terms of a paradigmatic meter bar.[r1 9 ][r20][2 9 ]

We have enough information to list at least some of the conditions Craig requires as an ontological foundation of objective moral values and duties.

So, given the previous points, it seems that according to Craig, if X is an ontological foundation of objective moral values (OFOMV), then the following conditions – which may overlap – obtain:

OFOMV1: X is not an abstract object.

OFOMV2: X is actual and not some sort of ideal observer.

OFOMV3: To be morally good is non-trivially but informatively identified with X, or at least with some states or aspects of X, in a way akin to the way in which water is identified with H2O, or heat is identified with molecular motion in 'informative identification' accounts of water and heat, respectively.

OFOMV4: Things (people, behavior, etc.) are morally good to the extent to which they are so because they stand in certain relation to X, or at least to some aspect or state of X, etc. In other words, that is what makes them morally good.

OFOMV5: [somewhat tentative; Craig is not clear on this] X (or at least some aspect, or state of X, etc.) provides a paradigm, a measure based on which moral goodness is determined, in such a way that nothing would be morally good if X did not exist. Also, this paradigm is in some way akin to the paradigmatic meter bar in the past.

There might be more requirements in Craig’s view, but at least those appear to be at least some of the requirements for an ontological foundation of objective moral values, although in the case of condition 5 – i.e., OFOMV5 -, and while it seems probable based on Craig's statements and argumentation that that is one of the conditions he demands, the evidence is less strong than it is in the case of the other conditions. But Craig does not specify which conditions he maintains an ontological foundation meets or must meet, we're only able to infer his conditions from claims and arguments he makes in different works, debates, etc., and the evidence appears to be insufficient to fully settle the matter. So, given that condition 5 seems to probably be one of his conditions, I will keep it, but also consider the alternative interpretation that he does not require 5 as a condition.

That aside, Craig makes corresponding demands in the case of an ontological foundation of objective moral duties, so there seem to be conditions OFOMD1-5 as well, corresponding to OFOMV1-5, and which I will not list for the sake of brevity.

So, some key questions are:

1. Should we expect that objective moral values have an ontological foundation, in the sense of informative identification?[3 0 ]

1': Should we expect that objective moral duties have an ontological foundation, in the sense of informative identification? [30]

2. Assuming that the answer to 1. is affirmative, should we expect that the ontological foundation of moral values meets conditions OFOMV1-5? (or 1-4, in case Craig does not require condition 5)

2'. Assuming that the answer to 1'. is affirmative, should we expect that the ontological foundation of moral duties meets conditions OFOMD1-5 – which are conditions corresponding to OFOMV1-5? (or 1-4, in case Craig does not require condition 5)

3.a. Assuming that the answer to 1. is affirmative, is there generally a burden on non-theists to present a theory of an ontological foundation of moral values, in the sense of informative identification?

3a'. Assuming that the answer to 1'. is affirmative, is there generally a burden on non-theists to present a theory of an ontological foundation of moral duties, in the sense of informative identification?

I will address these questions below, beginning in section 13.3, but first, I will raise some objections to Craig's own foundational account, in the following subsection.

On a terminological note, from now on, when I speak of an ontological foundation, I'm talking about a foundation in the sense of informative identification[30], unless otherwise specified, since that is the sense of foundation that Craig is talking about.

13.2. Craig’s theistic foundational account, and Craig's first premise.

In this subsection, I will first raise an objection to Craig's first premise, and I will argue that we have good grounds not just to reject it as unsupported, but to conclude it's actually false.

After that, I will raise an o bjection Craig's foundational account – and relevantly similar metaethical theories -, based on a more general hypothesis about ontological accounts in terms of informative identification, which I find plausible.

Later, I will raise another objection to Craig's foundational account, based on a circularity issue.

But I go on, I'd like to point out that the rest of the arguments in this essay are independent of any of the arguments in this subsection, and are also sufficient to properly reject Craig's metaethical argument – one of my goals is to provide mutually independent grounds to reject Craig's metaethical argument, each of them sufficient.

13.2.1. Duplicate minds.

Let's consider the following scenarios:

Scenario S3:

Dick#3 is a human serial killer who enjoys killing his victims in a particularly slow and painful manner. He does all of that for fun, and he has dozens of victims. God exists. [3 1 ]

Scenario S4:

Dick#4 is an agent whose mind is like Dick#3's mind is in S3 in the sense that Dick#4 perceives, remembers, desires, feels, generally subjectively experiences, believes, intends and chooses in S4, just the same as Dick#3 perceives, remembers, desires, feels, generally subjectively experiences, believes, intends, and chooses in scenario S3.[32] That includes, in particular, making the same free choices, and for the same reasons. For example, if Dick#3 in S3 freely chooses to torture a little girl for fun, cut her limbs off and watch her bleed to death, Dick#4 makes the same choice in scenario S4. And they both feel the same amount of pleasure in doing so.

However, Dick#4 has no victims. He never inflicted any pain or any suffering on anyone. He believes he has dozens of victims, he has experienced the same as Dick#3 has experienced, but he has no victims. Instead, in scenario S4 there is a very powerful creature – say Joker#4 -, who trapped Dick#4 – who was still planning to attack his first intended victim – in a sort of holodeck or similar environment, in which Dick#4 experiences what looks to him like interacting with people, torturing them, etc., but those aren't real people: they're just constructs of Joker#4's mind and/or a device made by Joker#4, which do not experience any suffering, pain, etc.

In S4, God exists.

Scenario S5.

Dick#5 is an agent whose mind is like Dick#3's mind is in S3 in the sense described above, who has victims who suffer the same as Dick#3's victims in S3 – their minds are also exact duplicates of those of Dick#3's victims -, etc.

God does not exist.

Scenario S6.

Dick#6 is an agent whose mind is like Dick#3's mind is in S3, in the sense described above. However, Dick#6 has no victims. He never inflicted pain or any suffering on anyone. He believes he has dozens of victims, he has experienced the same as Dick#3 has experienced, but he has no victims. Instead, in scenario S6 there is a very powerful entity – say Joker#6 -, who trapped Dick#6 – who was still planning to attack his first intended victim – in a sort of holodeck or similar environment, in which Dick#6 experiences what looks to him like interacting with people, torturing them, etc., but those aren't real people: they're just constructs of Joker#6's mind and/or a device made by Joker#6, which do not experience any suffering, pain, etc.

In S6, God does not exist.

If we make now intuitive assessments of the respective Dicks and their behavior, it's clear that:

a. In S3, Dick#3 is a morally bad person.

b. In S4, Dick#4 is a morally bad person, and exactly as morally bad as Dick#3 is in S3.

c. In S5, Dick#5 is a morally bad person, and exactly as morally bad as Dick#3 is in S3.

d. In S6, Dick#6 is a morally bad person, and exactly as morally bad as Dick#3 is in S3.

e. In S3, Dick#3 is behaving immorally when he's torturing and/or killing his victims for fun.

f. In S4, Dick#4 is behaving immorally when he experiences as if he's torturing and/or killing victims for fun. Furthermore, he's is behaving exactly as immorally as Dick#3 is behaving in S3, in the corresponding cases.

g. In S5, Dick#5 is behaving immorally when he's torturing and/or killing his victims for fun. Furthermore, he's behaving exactly as immorally as Dick#3 is behaving in S3, in the corresponding cases.

h. In S6, Dick#6 is behaving immorally when he experiences as if he's torturing and/or killing victims for fun. Furthermore, he’s behaving exactly as immorally as Dick#3 is behaving in the corresponding cases.

More generally, we are able to establish by reflection on our intuitive grasp of the relevant moral concepts[3 3 ] and by means of assessing scenarios like S3-S6 and others, that whenever two agents in different scenarios are such that their minds are alike in the sense described above, then one of them is morally good or bad if and only if the other one is – and to the same extent -; one of them behaves immorally if and only if the other one does and to the same extent, etc., and generally they have the same moral properties regardless of any other factors, and in particular regardless of whether certain other agents – be it God or even victims in the case of immoral behavior – exist in the scenarios in question.

Based on that, it seems clear that it's extremely counterintuitive the claim that if God did not exist, then objective moral values and duties would not exist. In other words, the first premise of Craig’s metaethical argument flies on the fact of our intuitions, and indeed our grasp of moral concepts.

Now, in a reply to an objection raised by Swinburne, Craig claims that on his account, God exists necessarily, so scenarios in which God does not exist are impossible, and that would allegedly block Swinburne’s objection. [r 2 2 ]

Without taking a stance on whether Craig’s reply works against Swinburne’s objection, we may ask here: Would that reply, or a similar one, work against the objection to the first premise that I'm raising here, which is somewhat similar but not in terms of possible worlds?

It seems clear that it would not, since what we’re assessing is whether if God did not exist, objective moral values and duties would exist, and in order to assess that matter, we need to assess scenarios in which God does not exist, regardless of whether such scenarios are possible.

In fact, Craig himself gives arguments in support of the first premise of his metaethical argument that are based on hypothetical scenarios in which God does not exist, and makes some assessments in those scenarios, claiming that in those cases, objective moral values and duties would not exist. However, in addition to not giving any good reasons in support of such claims – which is what I argue in much of this essay -, Craig fails to include in his assessment of the hypothetical scenarios in which God does not exist the fact that his conclusion that in those cases, objective moral values and duties would not exist, flies on the face of our intuitions. For instance, in support of the second premise, Craig uses the example of the Holocaust, appealing to moral intuitions and/or our grasp of moral concepts to support the claim that the Holocaust was immoral and would have been immoral even if all humans believed otherwise. There is no problem with that, but what Craig does not point out is the fact that the same intuitions and/or grasp of the concepts also support the conclusion that the Holocaust was immoral, and would have been immoral even if humans were the only persons that actually exist.

Scenarios like S3-S6, and the assessments I made above, also make use of some scenarios in which God does not exist, and those assessments strongly support the conclusion that it’s not the case that if God did not exist, objective moral values and duties would not exist.

At this point, someone might object to that assessment and claim that even though our moral intuitions constitute some evidence in favor of the view that in some scenarios in which God does not exist, objective moral values and duties do exist, upon reflection and when also factoring in the argumentation provided by Craig as evidence, the overall evidence supports the claim that in those scenarios, objective moral values and duties do not exist, and generally the contention that if God did not exist, objective moral values and duties would not exist.

However, that’s not even remotely plausible. Purely for example, our assessment that, say, a person who tortures children for pleasure every single day is a morally evil person is extremely intuitively plausible. On the other hand, some of the implicit or explicit hypotheses on which Craig bases his arguments in support of the first premise of the metaethical argument are remarkably weak. Again, purely for example, even before considering any of the arguments that I will give below, Craig's assumption that objective moral values and duties have or at least plausibly have an ontological foundation that meets conditio ns 1-5 – or even 1-4, in case Craig does not assume condition 5 – is not nearly as plausible as our assessment about the moral evilness of the sadistic torturer in question.

Hence, the considerations made above do constitute very strong evidence against the first premise of Craig's metaethical argument, that is, very strong evidence that the premise in question is false. [3 3 ]

Let us now turn to the other issue: Do scenarios like S3-S6 and/or considerations similar to those made above, also work against Craig’s theistic foundational account, and not merely his first premise?

It seems to me that they plausibly do, for the reasons I will explain below.

13.2.2. Informative identification accounts and conceptual problems.

Let's stipulate that:

1. O(X, Y) is a hypothesis that X is identified with Y, for some X and Y. In other words, O(X, Y) is a foundational account of X in terms of Y, in the sense of informative identification.

2. S(X, Y, r) is a hypothetical scenario, defined by some conditions C(S(X, Y, r)), and which are not contradictory. In other words, the conjunction of the conditions that define the scenario does not entail a contradiction. [r is an index number]

3. E(X, r) is a known, actually true hypothesis asserting that some concrete particular E in the actual world is (or was) X, or has (or had) X if X is a property. This condition is not scenario-dependent. It's about something in the actual world, not actual from the perspective of some hypothetical scenario.

For example, given that we know that Pol Pot was a morally evil person, we may pick E(moral evilness, 1) to be 'Pol Pot was morally evil', or equivalently 'Pol Pot had the property moral evilness'.

Another example: Given that I know that the liquid that is in the bottle on my desk is water, I may pick E(water, 8) to be 'The liquid in the bottle on my desk is water'[3 4 ]. And since we know that the liquid that fills the Pacific Ocean is water, we may pick E(water, 74) to be 'The liquid that fills the Pacific Ocean is water.' [3 4 ]

We say that condition N(O(X, Y)) obtains if the following conditions obtain for some r:

a(X, Y, r). From C(S(X, Y, r))&E(X, r), using the concept of X, it follows that there is/was some concrete particular Z in S(X, Y, r) such that Z is/was/has/had X in S(X, Y, r) but it's not the case that Z is (or was/has/had) Y in S(X, Y, r).

b(X, Y, r). From C(S(X, Y, r))&E(X, r), using the concept of X, no contradiction follows.

c(X, Y, r). From C(S(X, Y, r))&E(X, r), without using the concept of X, it does not follow that there is/was some concrete particular Z in S(X, Y, r) such that Z is/was/has/had X in S(X, Y, r) but it's not the case that Z is/was/has/had Y in S(X, Y, r).

d(X, Y, r). From C(S(X, Y, r))&E(X, r), even using the concept of X, it does not follow that there is/was some concrete particular Z in the actual world such that Z is/was/has/had X in the actual world, but it's not the case that Z is/was/has/had Y in the actual world.

Under those definitions, the general hypothesis in question is as follows:

C1: If N (O(X, Y)) obtains, then O(X, Y) is not a true foundational account of X in terms of Y, in the sense of informative identification.

I think C1 is plausible because if N (O(X, Y)) obtains, identifying X with Y would seem to be in conflict with our understanding of the concept of X.

In other cases of ontological accounts in the sense of informative identification, like the hypothesis that water is H2O, or the hypothesis that heat is molecular motion, conceptual analysis is insufficient to establish the identities in question – we also need empirical findings -, but on the other hand, the hypotheses in question do not seem to conflict with our grasp of the relevant concept.

Below, I will show that if C1 is true, Craig's foundational account is false. But before that, let's assess an objection:

Objection 13.2.2.1

i. C1 is implausible because it includes scenarios S(X, Y, r) that are metaphysically impossible.

ii. A hypothesis C2 just like C2 but restricted to metaphysically possible scenarios is true.

iii. Any scenario in which God does not exist is metaphysically impossible if God exists.

Reply:

Point iii. is usually accepted in my experience, but I do not see why it should be, under a GCB conception of God. At least, that would have to be argued for, and a mere claim that necessary existence is a great-making property would not seem persuasive in my view. Also, defining the term 'God' in modal terms (instead of conceivability terms, in order to avoid the issue of whether conceivability is coextensive with metaphysical possibility) in a way such that point iii. is guaranteed, may well give rise to coherence difficulties – apart from any coherence issues with concept of the GCB itself, some of which I will briefly comment on later.

Still, let's grant iii. for the sake of the argument, and let's say that the GCB is such that if God actually exists, God exists necessarily.

Then, point i. still seems very implausible, since the conflict between the allegedly informative identification between X and Y and our grasp of the relevant concepts seems to remain, regardless of whether S(X, Y, r) is possible.

Moreover, there seems to be no good reason why the [assumed] necessity of God if God exists should immunize theistic hypotheses of informative identification from challenges based on our grasp of the relevant concepts, while any other account of informative identification that does not posit an allegedly necessary being would be subject to such challenges.

On that note, and for that matter, someone might say that to be kind is to resemble God. Should we accept that there is no way of challenging such an account on the basis on our grasp of the concept of kindness and some hypothetical scenarios just because God [we assume] is necessary if he exists?

That seems very implausible.

Generally, it seems very plausible to me that we may properly challenge any account of informative identification on the basis of our grasp of the relevant concepts. C1 is an attempt to capture somewhat more precisely at least one of the ways in which we may raise some of those challenges based on our grasp of the relevant concepts.

Now, C1 seems plausible as far as I can tell, but even if C1 were to be rejected for some reason, I would still intuitively reckon that hypotheses of informative identification may properly be challenged on the basis of the relevant concepts even if not on the basis of C1. Still, given that C1 appears plausible and I see no good reason to reject it, I will make an argument against Craig's foundational account of moral goodness based on C1.

13.2.2.1. Goodness without God.

So, let's argue against Craig's ontological account based on C1.

Craig holds that moral values are grounded in God's nature, and God is the paradigm of goodness. [r 4] He also identifies moral obligations with God's commands.

So, on his account, it seems that to be morally good is to resemble God in some way, and to be morally obligatory is to be commanded by God. In other words, moral goodness would be identified with some unspecified resemblance to God, and moral duties with God's commands.
Let's assume at least for now that the concept of resemblance to God is not contradictory – else, Craig's ontological account fails just because of that.

Let's also assume that in a given scenario, nothing resembles God in the relevant sense unless God exists, since that seems to be Craig's position – else, it seems that resemblance to God would not require the existence of God, and the account would be compatible with non-theism plus the truth of ordinary moral assessments, like the assessment that it's immoral for a person to kill people for fun.

Let X1 be moral goodness, and let Y1 be resemblance to God (in some sense of resemblance; it works for any sense). Let E(X1, 1) be "Giorgio Perlasca behaved in a morally good way when he made some choices to save people from the Nazis"[r 23 ] In case someone objects to this particular example for whatever reason, any known actual example of morally good behavior by a human being will do.

Let's consider O(X1, Y1), and define scenario S(X1, Y1 1) by conditions C(S(X1, Y1, 1)), which are as follows:

C(S(X1, Y1, 1))-1: God does not exist.

C(S(X1, Y1, 1))-2: There was entity Giorgio#1 in S(X1, Y1, 1) such that the mind of Giorgio#1 was like the mind of Giorgio Perlasca in the actual world in the sense that Giorgio#1 believed, remembered, perceived, desired, felt, generally subjectively experienced, intended, chose, etc., in S(X1, Y1, 1), the same as Giorgio Perlasca believed, remembered, perceived, desired, felt, generally subjectively experienced, intended, chose, etc., in the real world. That includes in particular making the same free choices in the corresponding cases, and for the same reasons. [32]

Now, from C(S(X1, Y1, 1)) (i.e., the conjunction of the conditions), plus E(X1, 1), and our concept of moral goodness, no contradiction seems to follow, but it follows that Giorgio#1 behaved in a morally good way in S(X1, Y1, 1) when he made some choices to save people from the Nazis. In particular, it follows that Giorgio#1 had at least some moral goodness in S(X1, Y1, 1), but had no resemblance to God in S(X1, Y1, 1), since God does not exist in S(X1, Y1, 1).

On the other hand, we are not able to reach that conclusion only from C(S(X1, Y1, 1)) plus E(X1, 1), but without using the concept of moral goodness. Nor are we able to conclude, from C(S(X1, Y1, 1)) plus E(X1, 1), even using the concept of moral goodness, that there is some agent in the actual world that had some moral goodness but has no resemblance to God.

Granted, someone might say that there might be some hidden entailments somewhere, but there is no good reason to assume so.

So, it seems that N(O(X1, Y1)) obtains, and so if C1 obtains – which seems plausible -, then Craig's foundational account is false.

13.2.2.2. Water without H2O?

In this subsection, I will address a potential objection to the objection to Craig's metaethical account that I raise above based on a parallel with the account of water and H2O.

So, let X2 be water, and Y2 be H2O, and let's try to construct some S(X2, Y2, 2) based on which we would conclude that N(O(X 2 , Y 2)) obtains, and so if C1 obtains, then the account of water in terms of H2O is false.

Let E(X2, 2) be "The liquid that fills the Atlantic Ocean is water.", which is true oil spills notwithstanding[34]. The choice of the Atlantic Ocean, of course, is of no importance, as long as the claim meets the requirement of point 3. above.

So, how might someone go about arguing that N(O(X 2 , Y 2)) obtains?

Mirroring the construction I made above, someone might try the to define C(S(X2, Y2, 2)) as the conjunction of the following conditions:

C(S(X2, Y2, 2))-1: Oxygen does not exist.

C(S(X2, Y2, 2))-2: There is a liquid L2 such that L2 has in S(X2, Y2, 2) the same physical properties as the liquid that fills the Atlantic Ocean – or, if needed, the same physical properties as a specific sample from that the liquid in the Atlantic Ocean that is water.

However:

i. If being composed of H2O – i.e., two hydrogen atoms and one oxygen atom – is a physical property, and the definition of C(S(X2, Y2, 2))-2 is to be understood as using the actual world as a means to choose the properties of L2 in S(X2, Y2, 2), then from the conditions C(X2, Y2, 2) it follows that oxygen does exist in S(X2, Y2, 2) and also that it's not the case that oxygen exists in S(X2, Y2, 2). So, S(X2, Y2, 2) is contradictory, and so it may not be used to support N(O(X 2 , Y 2)).

ii. If being composed of H2O – i.e., two hydrogen atoms and one oxygen atom – is a physical property, and the definition of C(S(X2, Y2, 2))-2 is not to be understood as in i. above, then it's unclear how to construe it. There are different potential variants. But moreover, in any case, if S(X2, Y2, 2) is not contradictory, then it remains the case that from C(S(X2, Y2, 2)), it follows that L2 is not composed of H2O in S(X2, Y2, 2) – since oxygen does not exist in the scenario -, and since the physical properties of L2 in S(X2, Y2, 2) are the same as the physical properties of the relevant liquid in the actual world, the liquid in question is not composed of H2O. From that and E(X2, 2), it follows that the liquid that fills the Atlantic Ocean, in the actual world, is water but is not H2O. But then, N(O(X 2 , Y 2)) is not supported by S(X2, Y2, 2), because condition d .(X2, Y2, 2) does not obtain.

iii. If being composed of H2O is not a physical property, then anyone raising this objection would need to specify what they mean by 'physical property', or otherwise define the properties they want to assign to L2, so that the matter may be assessed. Personally, I've not found a variant that works; i.e., I've not found any way of showing that N(O(X 2 , Y 2)) obtains, by defining the relevant properties in any way.

Granted, someone might try to define different scenarios, or perhaps try to show that C1 entails that some other generally accepted informative identification – i.e., other than 'water is H2O' - would be false, but that would have to be defended. Granted, also, it's not an easy matter, but so far, I've not been able to figure out any way of showing that C1 would prove too much if true, and it remains plausible for the reasons I gave above that C1 is in fact true.

13.2.3. Circularity.

As I mentioned earlier, Craig holds that moral values are grounded in God's nature, and God is the paradigm of goodness. [r 4]

So, on his account, it seems that to be morally good is to resemble God in some way.

Also, Craig defines the term 'God' as "the greatest conceivable being". [r5][r5b][r2 4 ]

That definition seems to assume a shared concept of greatness that may be used to define the term 'God'. That seems doubtful to me.

Purely for example, I would be inclined to say that Alice is a greater person than Bob if and only if Alice is morally better than Bob. In other words, at least when it comes to greatness as a person, that seems to be about moral goodness as I would use the words. If Bob and Alice are equally morally good, I wouldn't say that Alice is a greater person than Bob just because she is, say, more powerful or knowledgeable than Bob, all other things equal. I might in those cases say that Alice has greater power or knowledge, but not that she's a greater person.

Now, if there is a shared concept of greatness and if greatness depends, when it comes to personal beings, only on moral goodness – and maximal greatness is thus equivalent to moral perfection -, then there might be for all we know plenty of possible personal beings who are maximally great. Moreover, in that case, greatness does not entail properties like being particularly powerful.

Yet, Craig and many other theist philosophers appear to understand greatness very differently, talking about several great-making properties, even if sometimes Craig considers it difficult to figure out some of those. [r2 4 ] While it's true that, as Craig points out[r2 4 ], the mere fact that people disagree on which properties are great-making does not imply that there is no fact of the matter, this specific example would seem to show more than mere disagreement about whether a certain property is great-making, but seems to reveal different concepts, since he's clearly not only talking about moral goodness, and I would only be talking about that as I grasp the concept.

Granted, someone might say that ontological greatness, even when applied to humans, is not the same as being a greater person. But I do not seem to have a concept of ontological greatness; I understand the idea of being a greater person, or the idea of being greater with respect to some other property other than personhood – like knowledge or power -, but that's it.

If there is no such shared concept of greatness, it seems that Craig's concept of God ought to be rejected, and then the same goes for his foundational metaethical account.

Granted, someone might say that I have failed to grasp the concept, or raise some other objection. I remain skeptical about the concept of a 'greatest conceivable being', but let's let that pass, and let's assume for the sake of the argument that there is such shared concept of greatness, and that it's not limited to moral goodness.

Then, there seems to be the following problem with his foundational account:

a. In his debate with Sam Harris, Craig claims that it follows from the very concept of God as the greatest conceivable being that God is morally good [r5b], and that someone who understands the concept of God realizes that asking why God is good is like asking why bachelors are unmarried. Thus, according to this, the expression 'God is morally good' is analytical, and the analyticity in question is apparently transparent.

b. Additionally, Craig claims that to be [morally] good is a great-making property [r25][r5b], and uncontroversially so.[r26]

c. Furthermore, when addressing a 'perfect island' objection to the ontological argument [r26], he gives an account of what the idea of the most perfect island would be – which he rejects, of course -, and that would be an island that exists in every metaphysically – or broadly logically – possible world and has all of the properties that are great-making with respect to islands.

It seems that the idea of God is, according to Craig, that of a being that has all of the great-making properties to a maximal degree and in every metaphysically possible world.

Given that conception, it seems that Craig's foundational account of moral goodness, in the sense of informative identification[30], is as follows:

G1: To be morally good is to resemble, in some relevant way, a being who exists in every metaphysically possible world and who has all of the great-making properties to a maximally possible degree in every possible world. [35]

But there seems to a vicious circularity problem. For example, let's say that someone proposes an ontological account in the sense of informative identification as follows:

B1: To be an unmarried man is to resemble, in some relevant way, the unique bachelor who exists at every possible world.

That would not be an informative identification. In addition to other obvious problems with B1, there seems to be a vicious circularity problem, since B1 is transparently analytically equivalent to:

B1': To be an unmarried man is to resemble, in some relevant way, the unique unmarried man who exists at every possible world.

The vicious circularity problem that I'm focusing on is independent from other problems with the account, like the non-existence of such necessary bachelor. For example, the following two accounts also would be viciously circular, as a proposed informative identification.

B2: To be an unmarried man is to be a bachelor.

B2': To be an unmarried man is to be an unmarried man.

Neither B2 nor B2' provides any kind of informative identification, at least not to someone who understands the meaning of the words in the sentences – though at least they have the advantage of being true, unlike B1 and B1'.

Now, going by Craig's claims, G1 seems to be transparently analytically equivalent to:

G2: To be morally good is to resemble in some relevant way a being who exists in every metaphysically possible world and who is morally good to a maximal degree in every possible world, and who also has all of the other great-making properties to a maximally possible degree in every possible world. [35]

So, in addition to other problems, Craig's foundational account seems to suffer from a vicious circularity problem.

That said, I will from now on leave aside the arguments given in this subsection, and will continue to raise objections to Craig's metaethical argument, based on other, independent grounds.

13.3. Should we expect that objective moral values and/or duties have an ontological foundation, in the sense of informative identification? [30]

Craig claims that explanations – and, presumably, ontological foundations – have to stop somewhere [r1 8], so it's clear that his own claims entail that some things are objective and have no ontological foundation in terms of informative identification – because the explanations stop somewhere. So, why should we expect an ontological foundation in the sense of informative identification in the case of, say, moral goodness? Should we expect an ontological foundation of, say, pain? How about, say, happiness? Pleasure? Fear? Kindness? Cruelty? Rules of vulture pecking order? Rules of wolf pecking order?

In any event, Craig does not provide any good reason to suspect that such foundation is required in the particular case of objective moral values and/or duties.

Is it that supervenient properties or states need an ontological foundation, perhaps? Why?

But if that is, and following the examples of water and heat, it seems that, say, moral goodness would plausibly be identified with the mental states on which it supervenes, so for example, moral obligations would supervene on a list of states of mind (including knowledge, intentions, etc.).

In any case, Craig has not met the burden of explaining why we should think that there is such a foundation.

Still, even assuming for the sake of the argument that there is an ontological foundation of objective moral values and there is an ontological foundation of objective moral duties, the questions of whether we should expect that such foundations meet conditions O FO MV1-5 and OFOMD1-5 – or even conditions 1-4, just in case Craig does not require condition 5 - (i.e., questions 2 and 2'), remain.

Craig has not given us any good reasons to think so.

But let's consider some other example, and assess whether the accounts in question – or rather, the foundations they posit – meet Craig's conditions, or whether we should expect that the foundations in other cases meet those conditions. [3 6 ]

13.4. Heat and Craig’s foundational conditions.

Let's consider the account of heat in terms of molecular motion. Heat is identified with molecular motion, but molecular motion does not meet all of Craig's condition OFOMV1-5.[3 6 ]

In particular, the fact is that there is no paradigm of heat, so molecular motion does not meet condition 5.

Someone might suggest that some past cases of molecular motion are the paradigmatic object, but that seems absurdly implausible. How would a past state of molecular motion be a paradigm of molecular motion?

Moreover, that's clearly not what the account of heat in terms of molecular motion actually holds.

Perhaps, someone might suggest that having a paradigmatic example of a molecule would suffice, but that does not work, either. How would a molecule be a paradigm of molecular motion?

Moreover, there does not seem to exist a paradigmatic example of molecules, either – not that that would seem to suffice. Would a water molecule be a paradigmatic example? Or maybe a methane molecule? It seems extremely plausible that neither of them would be.

Trying to go into subatomic particles, etc., would not be of help, either, and for similar reasons.

In short, molecular motion does not meet condition 5. But then, why should we expect that the ontological foundation of objective moral values and/or objective moral duties meet condition 5, if there is such a foundation?

The burden would be on the defender of condition 5.

Perhaps, someone might insist that heat is identified with actual molecular motion, present or past, or perhaps future too depending on the theory of time. However, that too does not give us a paradigm of heat, so condition 5 still would not obtain.

So, in any case, molecular motion does not meet condition 5. Still, perhaps Craig does not demand condition 5.

In any case, condition 5 is not met, so let’s leave condition 5 aside, and let’s focus on the other conditions. Does molecular motion meet all of the other conditions?

If it does not, then we may ask: why should we expect that an ontological foundation of objective moral values meet such conditions, if there is such a foundation?

But if molecular motion does meet condition s 1-4, then in particular the account of heat in terms of molecular motion is identifying actual cases of heat with actual cases of molecular motion – not with some abstract ideal molecular motion – and apparently also heat in [some] counterfactual scenarios with molecular motion in those counterfactual scenarios.

However, in that case, and mirroring the case of heat and molecular motion, perhaps actual cases of moral goodness and obligations may be properly identified with some actual mental states or traits, including perhaps cases of actual intent, choices, etc., describable by non-moral concepts, and similarly for counterfactuals. More precisely, moral goodness would be identified with the property of having such-and-such mental states, traits, and to have a moral obligation would be to have such-and-such states, etc.

That account would seem not to require God, and Craig provides no good reason to rule this out.

At this point, someone might ask things like: 'But how would moral goodness just attach itself to some actual mental states or traits, etc.?', suggesting that there is a problem with that suggested account. But for that matter, one may similarly ask: 'But how would heat just attach itself to molecular motion?'.

The problem with that kind of questions is that, on the heat account, heat just is molecular motion. And in the suggestion in the moral case, to be morally good just is to have some actual mental states/properties – including intent, choices, etc.

In brief:

a. Molecular motion does not meet condition 5, which Craig probably demands – though that's not entirely clear, as I mentioned earlier.

b. If molecular motion fails to meet one or more of the other conditions, then why would an ontological foundation of objective moral values, if there is one, meet all of those conditions?

A similar question may be asked with regard to objective moral duties.

Craig provides no good reason to think it's required in either case.

c. If molecular motion meets condition s 1-4, then in particular the account of heat in terms of molecular motion is identifying actual cases of heat with actual cases of molecular motion, not with some abstract ideal molecular motion, and also heat in [some] counterfactual scenarios with molecular motion in those counterfactual scenarios.

But in that case, mirroring the case of heat and molecular motion, perhaps actual objective moral goodness is identified with some actual mental states, traits, etc. - and a similar result holds for moral obligations, and for counterfactuals.

At this point, a defender of Craig's argument might say that those accounts do not work, because [allegedly] some properties are valued and others are not, and then, moral goodness, as a valued property, is not the same as any property that can be entirely described by non-moral concepts, which is not valued. A somewhat similar objection might be raised in the case of moral obligations. I will address the matters later.

13.5. Water and Craig’s foundational conditions.

Let's now turn to the account of water in terms of H2O. Water is identified with H2O, but H2O does not meet all of Craig's condition OFOMV1-5.[3 6 ]

At least, H2O does not meet condition 5, since there is no concrete paradigm of H2O, no paradigmatic molecule.

Now, someone might suggest that water molecules are all interchangeable because somehow they have some identical intrinsic properties – or something like that -, so any of them will do as a paradigm. However, even leaving aside isotopes, the problem of defining 'intrinsic' and other issues, that does not seem to work, either, since:

I. A paradigm apparently has to determine what water is. But a liquid is not water in virtue of being like some specific, actual molecule.

II. The account of water in terms of H2O does not depend on whether all molecules of water have identical intrinsic properties. In fact, the account would not fail if, in the future, we were to find out that, say, subatomic particles have some differences – and then, so do H2O molecules -, which are not detectable by the experiments carried out so far but require far more advanced technology and resources. So, even assuming for the sake of the argument that condition 5 would somehow be met if there are intrinsically identical particles, the point is that we would have a satisfactory account of water in terms of informative identification even if the particles were not intrinsically identical, and condition 5 were not met.

So, in short, plausibly condition 5 is not met in the case of H2O, either, but moreover, the account seems to work regardless of whether it's met.

Since condition 5 is plausibly not met in the case of H2O, why should we expect that the ontological foundation or foundations of objective moral values and/or duties meet condition 5?

Moreover, if the account of water in terms of H2O would be satisfactory regardless of whether condition 5 is met, why should a proposed account of objective moral values and duties in terms of informative identification that does not meet condition 5 be rejected just because of it?

The burden would be on the defender of condition 5 .

Also if someone suggested an ideal molecule of water, or some Platonic form, that would be in conflict with either condition 2 and/or condition 1; moreover, if that were acceptable in the case of water, why not in the moral case?

In any case, the fact is that H2O does not meet condition 5. Still, perhaps Craig does not demand condition 5.

Regardless, condition 5 is not met, so let's leave condition 5 aside, and let's focus on the other conditions. Does H2O motion meet all of the other conditions?

If it does not, then we may ask: why should we expect that an ontological foundation of objective moral values meet such conditions, if there is such a foundation?

But if H2O does meet condition s 1-4, then in particular the account of water in terms of H2O is identifying actual water with actual H2O and apparently also water in [some] counterfactual scenarios with H2O in those counterfactual scenarios, in which cases the condition s 1-4 should be understood relative to those scenarios.

However, in that case, and mirroring the case of water and H2O, perhaps actual cases of moral goodness and obligations may be properly identified with some actual mental states or traits, including perhaps cases of actual intent, choices, etc., describable by non-moral concepts, and similarly for counterfactuals. More precisely, moral goodness would be identified with the property of having such-and-such mental states, traits, and to have a moral obligation would be to have such-and-such states, etc. That account would seem not to require God, and Craig provides no good reason to rule this out.

The problem with that kind of questions is that, on the water account, water just is H2O. And in the suggestion in the moral case, to be morally good just is to have some actual mental states/properties – including intent, choices, etc.

In brief:

a. H2O does not meet condition 5, which Craig probably demands – though that's not entirely clear, as I mentioned earlier.

b. If H2O fails to meet one or more of the other conditions, then why would an ontological foundation of objective moral values, if there is one, meet all of those conditions?

A similar question may be asked with regard to objective moral duties.

Craig provides no good reason to think it's required in either case.

c. If H2O meets condition s 1-4, then in particular the account of water in terms of H2O is identifying actual water with actual H2O, and also water in [some] counterfactual scenarios with H2O in those counterfactual scenarios.

But in that case, mirroring the case of water and H2O perhaps actual objective moral goodness is identified with some actual mental states, traits, etc. - and a similar result holds for moral obligations, and for counterfactuals.

13.6. Sick birds and Craig’s foundational conditions.

Let’s consider the hypothesis that objective bird illness does exist, and which means or is at least equivalent to the following conditions.

OBI1: Statements – or judgments, or whatever one calls them – of the form ‘X is ill’, where X is a bird, are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that Bob the bird is ill – Bob is a specific bird -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether Bob is ill is matter of fact, not a matter of opinion.

OBI2: Some statements of the form ‘X is ill’, are true, where X is an actual bird.

So, objective bird illness does exist.

So, we may ask about the ontological foundation of bird illness. But perhaps, someone might say that illness is just the absence of health, and as such it doesn't require an ontological foundation. But then, we may ask about the ontological foundation of bird health.

It seems that it does not meet condition 5, one way or another. There seems to be no paradigmatic case of an ill or a healthy bird, for instance.

Granted, someone might claim that proper function requires God, or something like that, and then claim that the ontological foundation of bird health is also God, or God's choices, or something along those lines. But that would have to be argued for. [37]Without that, there is no good reason to believe that if objective bird illness or health has an ontological foundation, the foundation meets co ndition 5, and it seems implausible.

But if the ontological foundation in question – assuming there is one – fails to meet condition 5, what about the other conditions?

As in the case of water and heat, we may consider two cases:

1. If the ontological foundation of bird illness or health – whatever it might be, assuming there is one – fails to meet one or more of the conditions 1-4, why would a foundation of objective moral values or duties – if there is one – to meet all of them?

2. If the ontological foundation of bird illness or health – whatever it might be, assuming there is one – meets conditions 1-4 but not condition 5, it seems that the foundation is plausibly actual states or events [or hypothetical ones in counterfactual scenarios], and then a parallel for moral goodness and moral duties seems to be available, as described in the cases of water and heat.

13.7. Green stuff and Craig’s foundational conditions.

As we saw earlier, objective color does exist; here, to simplify, let's consider objective green only.

The hypothesis that objective green – or greenness, if one prefers – does exist, means the same as the conjunction of the following two conditions, or is at least equivalent to it.

OG1: Statements – or judgments, or whatever one calls them – of the form ‘X is green’, are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that the cucumber on the table is green, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether the cucumber on the table is green is not a matter of opinion, or a matter of taste. It's a matter of fact.

OG2: Some statements of the form ‘X is green’ are true, where X is an actual object.

So, objective green exists. But is there an ontological foundation of objective green, matching Craig's conditions? [3 6 ]

It seems plausible that there is not, for the following reasons:

a. Since objective green existed long before there were any humans, any ontological foundation that is some actual humans, or some aspect or state of some actual humans, would not match condition 2.

b. Ideal things and/or abstract objects are not an option, due to conditions 1 and 2.

c. Specific, actual photons in the past, or present, very plausibly would not meet condition 5.

d. Specific, actual objects with certain reflective properties in the past or present would plausibly not meet condition 5, either.

In particular, on an account in terms of wavelengths, light, etc. - which is more or less common, I think -, the ontological foundation of objective green would not meet condition 5.

As for the other conditions, it seems to meet them as long as the account identifies actual instances of green with actual instances of some reflective/emitting properties, etc., and makes similar identifications in [some] counterfactual scenarios. However, in that case, a similar account may be offered in the cases of objective moral goodness and objective moral duties, as explained in the analysis of the cases of water and heat, and Craig has provided no good reason to block that account.

At this point, a defender of Craig's argument might say that those accounts do not work, because [allegedly] some properties are valued and others are not, and then, moral goodness, as a valued property, is not the same as any property that can be entirely described by non-moral concepts, which is not valued. Similarly, he might say that obligations are either valued or normative, whereas properties describable by non-moral concepts entirely, are not. I will address the matter later.

13.8. Objective kindness and Craig’s foundational conditions.

Let's consider the hypothesis that objective kindness does exist, which means the same or is at least equivalent to the following two conditions:

OK1: Statements – or judgments, or whatever one calls them – of the form ‘A is kind’, are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that Alice is a kind person – Alice is a specific person -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether Alice is a kind person is not a matter of opinion, or a matter of taste.

OK2: Some statements of the form ‘A is kind', or 'A is a kind person', or 'A is a kind agent', or 'A is behaving in a kind manner', or 'A behaved in a kind manner', where A is an actual agent, or generally statements asserting that some actual entity A or some actual behavior B is kind, are true.

Then, since some humans are kind persons, we can tell that objective kindness does exist. Granted, someone might suggest that no human is a kind person. But that seems absurd. Under a usual understanding of the concept of kindness, some humans are obviously kind people, even if clearly not maximally kind. Moreover, it's enough that sometimes some humans are kind to other humans sometimes in order for objective kindness to exist.

So, we can tell, based on our observation of human behavior that objective kindness does exist.

We may then consider questions such as:

a. Does objective kindness have an ontological foundation, in the relevant sense, so that there is a proper ontological account of kindness in terms of informative identification?

b. If the answer to a. is affirmative, does the ontological foundation of objective kindness meet all of Craig's conditions?

Before assessing those questions, let's point out – just in case, considering some potential objections – that kindness is akin to moral goodness in the sense that it supervenes on mental properties of an agent described by concepts other than the concept of kindness.

For example, let's say that A13 is an agent that perceives, remembers, desires, feels, generally subjectively experiences, believes, intends and chooses in scenario S13 the same as A14 perceives, remembers, desires, feels, generally subjectively experiences, believes, intends and chooses in scenario S14. That includes their making the same free choices in the respective cases, and for the same reasons.

Then, it’s clear that A13 is a kind person if and only if A14 is a kind person. Moreover, the two are kind to the same degree – or unkind to the same degree, if they happen to be unkind persons. In other words, if their minds are a match in all of that, then they’re also a match with respect to kindness.

Let’s now address questions a. and b.:

a. Does objective kindness have an ontological foundation, in the relevant sense, so that there is a proper ontological account of kindness in terms of informative identification?

I don’t know. I wonder: Why would objective kindness have such an ontological foundation?

And if objective kindness does not have an ontological foundation, why would objective moral goodness have one?

But let’s assume for now that objective kindness does have an ontological foundation in the relevant sense, and let’s turn to the second question.

b. If the answer to a. is affirmative, does the ontological foundation of objective kindness meet all of Craig's conditions?

Again, I do not see why it would.

In particular, let’s consider the fifth condition – though, granted, it's not certain that Craig requires it.

That condition, in the case of moral goodness, is the following condition:

OFOMV5: X (or at least some aspect, or state of X, etc.) provides a paradigm, a measure based on which moral goodness is determined, in such a way that nothing would be morally good if X did not exist. Also, this paradigm is in some way akin to the paradigmatic meter bar in the past.

In the case of kindness, the condition would be:

OFOK5: X (or at least some aspect, or state of X, etc.) provides a paradigm, a measure based on which kindness is determined, in such a way that nothing would be kind if X did not exist. Also, this paradigm is in some way akin to the paradigmatic meter bar in the past.

But what might such be a paradigm of kindness?

At this point, someone might claim that the ontological foundation of objective kindness is God, that to be kind is to resemble God in some way, and that that ontological foundation meets all of condition 5. But why should we believe or even suspect that? How many ontological foundations is God supposed to be? Should we also believe that if God did not exist, objective kindness would not exist?

All of that would have to be defended by the claimant.

Alternatively, someone might claim that there is some relevant difference between objective kindness and objective moral goodness, so that the former plausibly has no ontological foundation or no ontological foundation that meets Craig's conditions, but the latter plausibly does. But that would have to be argued for.

Now, Craig's metaethical argument is an ontological metaethical argument, not a semantic one, and Craig does not seem to press open question arguments, at least not in that context and to the best of my knowledge. But perhaps, someone might suggest something along those lines, claiming that there is a difference between moral goodness and kindness that is relevant when it comes to whether there are ontological foundations, or ontological foundations of a certain kind, and which is based on semantically open vs. closed questions.

However, it's not clear that there is such difference. For example, the question 'I know that Dick inflicts excruciating pain on children every day, purely for pleasure, but is he a morally good person?' appears no more open to me than the question 'I know that Dick inflicts excruciating pain on children every day, purely for pleasure, but is he a kind person?'

Granted, someone might claim – say – that only the first question is open, or that both questions above are open, but that that is only a particular case when it comes to kindness, whereas other, relevant questions about kindness are closed, but relevant questions remain open in the case of moral goodness.

But the burden would be on the claimant.

As it is, it seems that the ontological foundation of objective kindness, if there is one, does not meet condition 5. Perhaps, Craig does not demand that a foundation meets condition 5. But in any case, let's consider the other fo ur conditions:

1. If objective kindness has no ontological foundation, why would objective moral goodness have one?

2. If objective kindness has an ontological foundation that fails to meet one or more of the conditions 1-4, why would objective moral goodness have one that does?

3. If objective kindness has an ontological foundation that meets conditions 1-4, since it does not meet condition 5, plausibly it would be some combination of mental states or traits describable by concepts other than the concept of kindness, or moral concepts. But then, as in the case of water and heat, a similar account may be offered in the case of moral value, and of moral duty.

13.9. Objective cruelty and Craig’s foundational conditions.

Let's assess the hypothesis that objective cruelty exists, which means or is at least equivalent to the following two conditions:

OCR1: Statements – or judgments, or whatever one calls them – of the form ‘A is cruel’, are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that Dick is a cruel person – Dick is a specific agent -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether Dick is cruel is not a matter of opinion, or a matter of taste.

OCR2: Some statements of the form ‘A is cruel', or 'A is a cruel', or 'A is a cruel agent', or 'A is behaving in a cruel manner', are true, where A is an actual thing.

Now, it seems obviously true that objective cruelty does exist.

Before we go on, and in order to preclude some potential objections, let's establish first that cruelty is not an absence of kindness, or an absence of love, or both, or a similar absence.

To see that, let's consider the following scenario:

Scenario S12:

In the Andromeda galaxy, advanced aliens made an enormously intelligent cyborg, Zantarly, believing it would help them develop new technologies more quickly. But they made a serious mistake during the design process. As a result, they ended up with a being who cares about no one's interests but its own, and who is also willing to use any resources it finds to its advantage and with no regard for any other agent. Zantarly destroyed its creators by the most efficient method it found, and took over their planet. After that, it went to other planets, and then to other planetary systems, consuming resources and killing trillions of complex living organisms, just to use the resources for its goals, which are basically controlling more space without rivals, surviving for as long as it can without any suffering, and a number of means-to-ends goals.

Now, Zantarly has no kindness whatsoever. It also has no love, or similar feelings. It does care about itself, but for no other agent. The lives or suffering of others isn't even a consideration for Zantarly.

On the other hand, Zantarly is not interested in causing anyone any suffering for the sake of it, either. Zantarly simply does not care at all about the suffering of others, one way or another, and so it causes a lot of suffering, but always goes for the most efficient method of extermination, and it's usually very fast, compared with most other alternatives.

On a different galaxy, very distant and very similar to the Andromeda galaxy, there is another cyborg, Zindur, who is as intelligent as Zantarly, and equally advanced scientifically and technologically. Like Zantarly, Zindur is completely devoid of kindness, love, and any similar traits. It values itself just as Zantarly values itself, but that is all.

Also, like Zantarly in the Andromeda galaxy, Zindur travels through the galaxy where it resides, destroying other beings and consuming resources, in similar numbers and amounts as Zantarly, respectively.

However, there is a the following difference between the two: in most cases, when it finds civilizations, Zindur does not choose a fast method of extermination. Instead, it keeps a good number of individuals alive for the purposes of torturing them for pleasure, and it now has a collection of billions of intelligent, social and usually kind beings; those beings are tortured by Zindur for their entire lives, only for the purpose of entertainment, and with no hope of escape.

In scenario S12, Zindur is more cruel than Zantarly, even though they both equally lack all love, kindness, and any similar property. Also, similarly, in scenario S12, Zindur’s behavior is overall more cruel than Zantarly’s behavior, even though both Zindur’s and Zantarly’s behaviors are completely devoid of any love or kindness.

Hence, cruelty is not just the absence of one of those properties, or any similar one, since in that case Zindur and Zantarly would be equally cruel, and so would be their behaviors.

Now, let’s point out – just in case – that cruelty supervenes on mental properties described by concepts other than the concept of cruelty.

For example, let’s say that A15 is an agent that perceives, remembers, desires, feels, generally subjectively experiences, believes, intends and chooses in scenario S15 the same as A16 perceives, remembers, desires, feels, generally subjectively experiences, believes, intends and chooses in scenario S16. That includes their making the same free choices in the respective cases, and for the same reasons.

Then, it’s clear that A15 is a cruel person if and only if A16 is a cruel person. Moreover, the two are cruel to the same degree, if they are cruel. In other words, if their minds are a match in all of that, then they’re also a match with respect to cruelty.

Let’s now ask:

a. Does objective cruelty have an ontological foundation, in the relevant sense, so that there is a proper ontological account of cruelty in terms of informative identification?

I don’t know. I wonder: Why would objective cruelty have such an ontological foundation?

And if objective cruelty does not have an ontological foundation, why would objective moral goodness have one?

But let’s assume for now that objective cruelty does have an ontological foundation in the relevant sense, and let's turn to another question:

b. If the answer to a. is affirmative, does the ontological foundation of objective cruelty meet all of Craig’s conditions?

It seems clear that that is not the case. In particular, the fifth condition is probably not met – though, granted, it's not certain that Craig requires it.

That condition, in the case of moral goodness, is the following condition:

In the case of kindness, the condition would be:

OFOCR5: X (or at least some aspect, or state of X, etc.) provides a paradigm, a measure based on which cruelty is determined, in such a way that nothing would be cruel if X did not exist. Also, this paradigm is in some way akin to the paradigmatic meter bar in the past.

But what might be a paradigm of cruelty?

At this point, someone might claim that the ontological foundation of objective cruelty is God, that to be kind is to be dissimilar to God in some way, and that that ontological foundation meets condition 5. But why should we believe or even suspect that? How many ontological foundations is God supposed to be?

Incidentally, it seems that the God option is not available to Craig in the case of cruelty, for the following reason: in his defense of his metaethical argument, Craig maintains that if God did not exist, rape would be cruel but not morally wrong. [r 27 ] For all of the reasons I'm giving, I do not think Craig has provided any good reasons to even suspect that rape would not be morally wrong if God did not exist, but that aside, the fact that Craig says that if God did not exist, rape would still be cruel, indicates that Craig does not believe that the ontological foundation of objective cruelty is based on God.

Perhaps, someone say argue that there is a difference between objective moral goodness and objective cruelty that is relevant in this context, and so that only objective moral goodness has an ontological foundation meeting all of Craig's conditions, including condition 5. But the burden would be on the person making that claim. Regarding the issue of open vs. closed questions, the case of cruelty is handled in a manner similar to the case of objective kindness.

Perhaps, though, Craig does not require condition 5. But in any case, if objective cruelty has an ontological foundation, it does not meet condition 5, so let's consider the other conditions.

Here too, the matter is handled as in the case of objective kindness.

13.10. Objective greatness.

I already raised an objection to Craig’s foundational account based on the idea of greatness. But in this part of the essay, I’m leaving aside for the sake of the argument anything in subsection 13.2, in order to give mutually independent arguments against Craig’s metaethical argument, so I will address the matter again, this time asking for an ontological foundation of objective greatness.

Let's assess the hypothesis that objective greatness exists, which means or is at least equivalent to the following two conditions:

OGR1: Statements – or judgments, or whatever one calls them – of the form ‘A is great’, are objective, in the ordinary sense of the term ‘objective’ mentioned above. For instance, if someone claims that Alice is a great person – Alice is a specific agent -, then there is an objective fact of the matter as to whether that claim is true, and so on. To use Craig’s own distinction, whether Alice is great is not a matter of opinion, or a matter of taste.

OGR2: Some statements of the form ‘A is great', are true, where A is an actual thing.

Does objective greatness exist?

Craig is committed to an affirmative answer, since his conception of God is the GCB conception – i.e., Craig defines the term 'God' in that fashion – so God is by definition the greatest conceivable being -, and claims God exists. [r5][r5b][r25]

I already raised some objections to that definition, but I’m leaving that aside here as well.

So, we may ask someone defending Craig’s arguments: what is the ontological foundation of objective greatness, in the sense of informative identification? [30]

Now, someone might suggest that God is the ontological foundation of objective greatness, and that to be great is to resemble God. However, the term 'God' is defined as 'The greatest conceivable being', so a hypothesis that to be great is to resemble God means the same as a hypothesis that to be great is to resemble the greatest conceivable being. Surely, that would no an informative identification. In fact, such an account would be viciously circular. Of course, to say that to be the greatest conceivable being is to have all great-making properties to a maximal degree or something along those lines would not resolve the problem: the circularity would remain, and would be just as vicious.

Granted, someone might say that the ball has to stop somewhere, and there is no informative identification account of greatness. But if there is no such account in the case of greatness on Craig's views, then why should we accept his claim that there is such account in the case of moral goodness?

13.11. Valued properties?

Earlier, I suggested a potential account of moral goodness in terms of mental states and/or traits – including, perhaps, decision, intentions, etc. - describable in non-moral concepts, mirroring the cases of water/H2O and heat/molecular motion, and also suggested one potential objection in terms of value.

To be clear, I’m merely considering this potential alternative, but I’m not suggesting Craig would raise this particular objection.

So, according to this potential objection, the properties in question cannot be identical because one of them – moral goodness – is [in some sense] a valued property, whereas the properties of having some mental traits, states, etc., describable by means of non-moral concepts, are not valued properties.

But here, we may ask some questions, like:

Is there a difference between valued and non-valued properties?

If so, what is for a property to be valued? More precisely, in what sense is moral goodness – allegedly – a valued property?

We may consider two alternatives:

1. If to say that a property is valued is to merely say that it’s a property picked by the concept of moral goodness, or moral badness, or generally some moral concept, etc., then as long as properties such as greenness or heat are picked by non-color or non-heat concepts respectively – like some concepts involving wavelengths, etc., or the concept of molecular motion -, there is no good reason to think that the same can’t apply to moral properties.

2. If to say that a property is valued is to say that it’s valued by some agent or another – say, normal humans -, that also does not constitute a problem for a proposed identification between moral goodness and the properties of having some mental traits, states, etc., describable by means of non-moral concepts.

So, let’s say that it’s neither 1. nor 2. Then, the question remains: in which sense of ‘valued’ might an objector claim that the property of moral goodness is a valued property?

The burden to explain himself would be on the person raising this objection. Without such clarification, it's not clear how to make sense of it. For instance, 'to value' is a verb. Agents value. There are properties that are valued by some agents, and for example in the case of normal humans, usually or always kindness and moral goodness are positively valued by such agents, whereas cruelty and moral badness are negatively valued by such agents. That’s comprehensible. But given that the objection is not about what agents value, and not about alternative 1. above, either, then what is it about?

That aside, I will consider some potential replies, which might depend on how the person raising the objection I'm considering in this subsection explains the claim of a valued property.

For instance, we may consider options:

Is kindness a positively valued property, and cruelty a negatively valued one?

And what about illness or healthiness? Cowardice and bravery? Aren't those valued properties?

So, let's consider kindness and cruelty. Craig himself says that rape would still be cruel if God did not exist, even if he claims it would not be morally wrong [r27]. Perhaps, other theists might claim otherwise, but of course that would have to be argued for. Without that, there is no good reason to suspect that if God did not exist, there would be no kind people and no cruel people.

So, let's first let's first assume that cruelty [kindness] is a valued property – whatever that means.

Then, if cruelty [kindness] has no ontological foundation, then there are valued properties with no ontological foundation. If cruelty [kindness] has an ontological foundation in terms of informative identification, then either cruelty is informatively identified with a non-valued property, or it's informatively identified with a valued property. If it's informatively identified with a valued property, we repeat the process.

In the end, the alternatives are infinite regress of informative identifications of valued properties not grounded in God, a valued property with no ontological foundation, or a valued property that is informatively identified with a non-valued property.

If there is a valued property identified with a non-valued property, that blocks the objection I'm considering in this subsection. On the other hand, if there is infinite regress of informative identification of valued properties not grounded in God, or a valued property with no ontological foundation, we may suggest the same might be the case for moral properties. Either way, the metaethical argument seems to remain blocked.

So, what if cruelty and kindness are not valued properties? We may run a similar argument with properties like bravery or cowardice, healthiness or illness. But what if none of them are valued?

Someone might even claim that cruelty, kindness, etc., are valued properties because God exists, but they wouldn't be valued properties if he did not, even if they were instantiated.

So, as before, the key question here is: what does it mean for a property to be valued, in the sense of this objection?

The burden to explain what they mean by 'valued property' would be on the person making that claim, and as I pointed out earlier, it seems difficult to make sense of a claim of that sort without such clarification.

Still, another potential reply to this objection – as before, depending on how the objector construes the term – is based on scenarios involving alien minds, like scena rio 12, or an earlier cyborg scenario, or similar scenarios.

As I mentioned earlier, normal humans, such agents [at least usually] positively value kindness and moral goodness, and negatively value cruelty and moral badness.

But some of the aliens in those hypothetical scenarios do not value kindness positively ever, and do not value cruelty negatively either, except, perhaps, to the extent that someone might be cruel towards them. But even then, what they may actually value negatively in that case is not the cruelty itself – which depends on the mind of the agent being cruel only, regardless of results -, but rather the pain they might experience as a result.

Moreover, some of those aliens place no value whatsoever in moral goodness, or moral badness.

Granted, those aliens are hypothetical. But they're not contradictory, and furthermore, in cases this makes a difference, they seem to be metaphysically possible. In fact, for all we know, some aliens relevantly similar to those might even be actual. We don't really know one way or another.

The point I'm trying to make is that the positive or negative evaluation of the properties seems to be on the mind of the agents making the evaluation. Else, should those agents consider also that they are valued properties based on what they value, and ask for the ontological grounding of those properties?

For example, let's consider the following scenario:

Scenario 18:

There is a planet where some intelligent, social aliens evolved, say species#18. They have a common language for their entire species, and part of their language resembles moral language and is based on something that resembles human moral intuitions. However, there is no perfect match. Even under no error, their sense of species#18-moral-goodness and the human sense of moral goodness diverge in some cases.

They also have language similar to language of obligations, and more or less similar intuitions.

Should species#18 philosophers conclude that there is a necessary being who is the ontological foundation of species#18-moral-goodness, which would be something other than moral goodness, so that the being in question is not God, but species#18-God?

Granted, a defender of Craig's metaethical argument may simply say that aliens like that are metaphysically impossible, but in particular that involves a claim about exobiology encompassing the entire universe. Why should we believe such a claim? It seems humans do not have at this point the knowledge to make such a claim in a justified manner.

Alternatively, he might say that either they or us would be mistaken about moral goodness, rather than there being something like species#18-goodness. But that is a problem as well. Furthermore, there might aliens that are far more different and that have nothing particularly similar to a sense of moral goodness, even though they too would value some things, and have some language associated with some of those evaluations. Should we conclude that those aliens would be massively in error, or that we would be? If there are many different species with pairwise different evaluations, etc., should we conclude that nearly all of those aliens are in error? That too would seem to be an unjustified claim about exobiology.

Moreover, what if our successors in a distant future where to contact aliens like that?

Intuitively, it seems clear to me that the humans or post-humans[14] in question should not doubt that Hitler was a morally bad person regardless of how some extraterrestrials value things and/or because of the kind of language they use when making those evaluations. Nor should they conclude that those aliens are massively mistaken when making claims in their own alien language.

My point here is that, in addition to being hard to make sense of, the 'valued properties' objection may well reveal a commitment on the part of the defender of the metaethical argument to a number of unjustified claims, or at least claims he ought to defend. The burden to explain and defend his stance would be on him.

13.12. Normativity.

In addition to objection based on value, a defender of Craig's metaethical argument might raise an objection based on moral obligations and normativity, and contend that the property of being morally obligatory is a normative property – in some sense of ‘normative’' -, and [allegedly] normative properties cannot be identical to non-normative properties.

As in case of the value objection, here too one may ask for clarification about the relevant sense of ‘normative’, ask why being commanded by God is normative in the relevant sense, and so on.

For instance, let’s say for the sake of the argument and for now that – as the objection contends -, the property of being morally obligatory is a normative property in some sense. So, here one may ask: is the property of being commanded by God also normative?

If it’s not, then Craig’s account identifies a normative property with a non-normative one.

If it is, then why is that the case? In other words, why do God’s commands have the property of being normative in the relevant sense. And what is that sense, anyway?

In his debate with Sam Harris, Craig contrasts commands given by a random stranger, and commands given by a competent authority. [r5] So, someone might suggest that God's commands are normative because they were issued by a competent authority. But why is God a competent authority?

Someone might suggest that the property of being normative in the relevant sense is the property of being commanded by a competent authority. But then, why should we accept that the property of being morally obligatory has the property of being normative, if the property of being normative is the property of being commanded by a competent authority?

Perhaps, a defender of the metaethical argument might just say that explanations have to stop somewhere, or something along those lines. But the point here is that the person raising a ‘normative’ objection ought to explain what he means by ‘normative’ in that context, and generally argue in support of the objection.

In addition to the above, here one one may also construct scenarios involving aliens, and:

a. Ask whether those aliens should conclude that their species#18-moral obligations are constituted by the commands of a being other than God.

b. Point out that it’s intuitively clear that if, in the distant future, humans were to find aliens who make evaluations and normative judgments very different from those made by humans, have language associated to their evaluations and normative judgments, etc., those future humans shouldn’t just because of that suspect that Hitler did not have a moral obligation not to commit the Holocaust – he had that obligation regardless of what happens with aliens on other planets.

Nor should those humans believe that those aliens would be massively mistaken when they make those judgments in their own alien language.

c. Generally raise a number of issues similar to those raised in the previous subsection,

In order to assess the matter in greater detail, further clarification on the part of the objector would be required. But the previous considerations indicate that the ‘normativity’ objection does not appear promising for someone defending Craig’s metaethical argument.

Still, perhaps a defender of the argument might suggest that somehow morality is rationally binding, in the sense that for any moral agent, it’s always irrational to behave immorally, and that somehow that is a problem for non-theists.

However, whether it’s irrational to behave in a certain way seems to depend on the value structure of the agents in question. If it is the case that it’s always irrational for humans to behave immorally, that would seem to be a matter of human psychology. But even in the limited case of humans or very similar beings, it appears very implausible that it’s always irrational to behave immorally.

To see this, let’s consider the following scenarios, one involving a non-modified human and one involving a genetically modified human.

Scenario 19:

Let’s say that Joseph is a brutal dictator. He engaged in mass murder, torture, theft, etc., all for personal gain. He’s also a psychopath, and couldn’t care less about morality.

Regardless of whether sometimes Joseph behaved irrationally in the process of getting to where he is, a question here is what would be rational for him to do now that he is in that position, and in particular, whether he rationally should continue to suppress the expression of ideas that probably would be a threat to the continuation of his regime, even by means of torture, murder, etc.

It seems to me that if he rationally reckons that the moment he loses his grip on power, some people seeking retribution will kill him – and we may further stipulate that he does -, then given what he values, he rationally ought to continue suppressing the expression of ideas in that manner, even though he morally ought not to do so. At the very least, it would not be irrational on his part to continue to do that, even though it would clearly be immoral.

Scenario 20:

Let’s say that in a distant future, a small group of people are victorious after a global war using all sorts of weapons of mass destruction, like nuclear bombs, biological weapons, very advanced killer robots, etc. They take control of the planet, and the few other survivors have little technology left. Later, in an internal struggle, one of the victors, Jack, uses robots to carry out a successful coup against the others, and kills them all.

So, Jack controls the world. Jack sends his robots to search and destroy any piece of advanced technology left in the hands of anyone else or abandoned, and to kill anyone with knowledge of science and/or technology.

Five centuries later, Jack – who does not get old, due to genetic engineering – rules the world unopposed, and all other people do not know about the war, have no knowledge of history beyond a couple of generations, and know nothing about modern science.

Jack – who is a psychopath and does not care about right or wrong – forces people to worship him, and goes around doing as he pleases to others; given his army of robots and his advanced technology in general, there is no hope of resistance.

In that scenario, it’s obvious that Jack behaves immorally by keeping anyone else under his thumb. But there seems to be nothing irrational about that behavior on his part.

So, the suggestion that for any moral agent, it’s always irrational o behave immorally seems very implausible, intuitively.

All that aside, there are further problems for the defender of Craig’s metaethical argument; on that note, in the next subsection I will consider the issue of whether non-theists have any significant burden on this particular issue.

13.13. Non-theism, ontological foundations, and burden.

Let's address now questions 3 a, and 3a', assuming for the sake of the argument and that the answers to 1 and 1' are both affirmative, and even that the answers to 2 and 2' are also affirmative – even though Craig has provided no good reason whatsoever to think so.

It seems implausible that she would have such a burden, as the following parallels illustrate:

I. Let's take a look at one of the examples of an informative identification given by Murphy: heat is molecular motion. [3 8 ]

Through the vast majority of human history, that account was not available.

Yet, that did not place any burden on someone rejecting previous theories of heat to provide an account of her own. In fact, a person rejecting previous theories of heat could have properly said that she did not know what heat was, and that a lot more research was needed in order to then develop a foundational account.

In particular, if someone had posited that heat was God's causing certain sensation of people, and had challenged others to come up with an alternative foundational account, that would not have resulted in any burden on the people being challenged in that fashion, and the lack of an alternative account would not have provided any non-negligible evidence in support of theism, even if a refutation of the theistic account had not been available.

II. Let's take a look at another example of an informative identification, given by Murphy: water is H2O. [3 9 ]

Through the vast majority of human history, that account was not available.

Yet, that did not place any burden on someone rejecting previous theories of water to provide an account of her own. In fact, a person rejecting previous theories of water could have properly said that she did not know what heat was, and that a lot more research was needed in order to then develop a philosophical account.

III. Let's consider an example Craig himself gives in the context of his debate with Sam Harris [r5]: light and darkness. As Craig points out in that debate, humans knew how to use the word 'light' (or synonyms, one might add) before they knew the nature of light, and they could also tell the difference between light and darkness long before they knew the physics of light. But that's not what matters in the context of this section. Rather, in this context we may add that for most of the history of humanity, the ontological foundation of light, in terms of informative identification, was not known. It was only discovered after a lot of scientific progress.

Points similar to I., II, and III can be made for a number other cases of informative identification accounts, plausibly for most of them. But if, in all those cases, there was no burden, why should there be such a burden on a non-theist in the case of objective moral values and duties, which is plausibly far more complicated than the cases of water or heat?

After all, non-theism is not a philosophical stance that claims to have an account of objective moral values and duties in the sense of informative identification, or to be capable of explaining it in those terms, or for that matter in any other sense. Granted, some not-theists might make one of such claims, perhaps by adhering to a general philosophical stance that makes one of those claims. But that does not create a general burden on other non-theists, but only on those making such claims.

So, given the previous points, the answers to 3.a and 3.a' are plausibly both negative, at least for non-theists in general – even philosophers -, or even non-theists familiar with Craig's argument, or even non-theists familiar with other theistic metaethical arguments. Some non-theists might have a burden because of other claims they make, and which go beyond their rejection or theism, but that's another matter.

Moreover, in all of those cases (i.e., water, heat, color, light, etc.), it took a very long time a lot of scientific progress was required before a correct or approximately correct philosophical foundational account was developed.

Given all of that, if someone claims that unless a plausible foundational account compatible with non-theism is given, we should based on that and the second premise of Craig's argument reckon that God exists, they ought to argue in support of that claim.

14. Evolution and self-sacrifice, and generally human behavior.

A theist might object to a non-theistic evolutionary accounts on the ground of acts of heroism, sometimes resulting in the death of the person making the sacrifice. More generally, a theist might claim that an evolutionary process without God would not lead to entities with the psychological make up of humans.

This kind of objection is not a metaethical challenge, though, but it’s a challenge to any scientific understanding of human evolution that does not involve God or any other creator, but instead involves the processes posited by present-day science, with perhaps some minor adjustments.

An example of this kind of objection is Craig’s challenge to a biological account of our sense of beauty. I addressed that particular example earlier, but a general account of the evolution of human traits is a very complicated scientific question, and surely beyond the scope of this article.

So, I will just point out that if a theist claimed that the evolutionary process without God would not result in the kind of being that we humans are, the burden is on him to make his case.

As it stands, the evidence for evolution by the means posited by present-day science is very solid, so one should expect only minor adjustments in the future, rather than a massive change like the introduction of a designer guiding evolution.

That said, a non-theist may also argue that if God existed, we would not observe what we observe today. While it would be beyond the scope of this article to make any such arguments, we may mention – purely for example – that she may make an argument from suffering and/or from moral evil, or an argument from the existence of moral agents with flawed senses of right and wrong, etc.

15. Moral obligations, legal obligations and commands.

In his debate with Sam Harris[r5], Craig claims that moral obligations and prohibitions result from commands issued by a competent authority.

Apparently as a parallel, he points out that if a police officers tells one to pull over, one is legally obligated to do so, whereas if a random stranger tells one to pull over, one is not legally obligated to do so.

Then, Craig goes on to claim that on atheism, since there is no authority to give moral commands or impose moral prohibitions, then there are no moral obligations.

However, his assertions about moral obligations are not backed by any arguments he makes, nor by his example of the police officer.

In particular, the case of the police officer is obviously one of legal obligation, not a moral obligation. Granted, in many cases – but not all -, a person did not have a moral obligation to pull over before the police officer gave the order, but she does have such an obligation after that.

However, the fact that in some cases moral obligations result from commands does not even suggest that all moral obligations are like that. Purely for example, let's consider the following scenario:

Alice works as a lifeguard, but is on vacation abroad. She's taking a walk by a lake, enjoying the view, when she sees that a child is on a small raft alone, and then falls to the lake, and is seriously struggling, clearly in immediate danger of drowning. The child is about 10 meters from the shore, and 20 meters from Alice's position.

The waters are perfectly calm, and Alice correctly reckons she can jump into the lake, swim towards him, etc., and she has a reasonably good shot at getting there in time and saving his life, with no non-negligible risks to herself. She sees no other people around.

In that situation, Alice did not have any moral obligation to jump into the lake until she saw what happened, but then she did.

Granted, we may say that Alice already had the general moral obligation to jump into a lake if a situation like that involving the child were to happen. But the same applies to the case of the police officer – i.e., one usually has a moral obligation to obey commands by police officers in such contexts.

Also, arguably Alice would have had no moral obligation to jump if – say – she hadn't witnessed what happened to Bob, without any fault on her part. But also similarly, if a person does not see the police officer's indication, without any fault on her part, then she has no moral obligation to pull over, either.

In case Alice or the other person are at fault, the matter is more debatable, but in any case, there is no difference that is relevant to the matter at hand, namely regarding how moral obligations arise.

So, it seems that to the extent to which a moral obligation to pull over resulted from the police officer's command, a moral obligation also resulted from the child's falling into the water.

There are of course, plenty of other examples of moral obligations that one may consider, but based on our experiences, as well as hypothetical scenarios, there appears to be no indication whatsoever of any general connection between moral obligation and commands, let alone evidence that commands are a requirement for moral obligation.

Craig does not seem to give any evidence in support of the idea that moral obligations require a commander. If he's trying to appeal to some intuitions, I will say it's extremely counterintuitive that moral obligations would require an authority. For example, we may set up the following scenario:

God does not exist, and Bob is the only human left alive. There is a cat on the mat.

Then, there is no commander, authority or command, but intuitively, it's clear that Bob would be acting immorally if he were to pour gasoline over the cat and set her on fire just for fun. But that entails that in the scenario, Bob has a moral obligation not to behave in that manner.

Moreover, instead of a cat, we may consider, say, a child, Mary – who also isn't a moral commander, and in any case is asleep and has issued no commands -, setting up the scenario so that Bob and Mary are the only humans left alive, and God does not exist. In that case, it's also clear that Bob has a moral obligation not to set Mary on fire for fun.

So, Craig's claim goes against strong moral intuitions [3 3 ], and there seems to be no good reason to believe it's true.

In another defense of his metaethical argument[r9], and also in his debate with Richard Taylor[r 2 8], Craig quotes Taylor[r2 9], apparently as a means of supporting his contention about moral obligations and commands. [r2 9]

In some of those quotes, Taylor says that questions about right and wrong were never answered outside religion, and also that the concept of moral obligation is unintelligible without God. [r2 9]

However, Taylor's claims that questions about moral right and wrong have never been answered outside religion" is false unless one construes the word 'religion' in a way so broad that it encompasses philosophies like – purely for example – Confucianism.

Moreover, regardless of how one construes the word 'religion', as I pointed out earlier, historically, there are/were plenty of cultures that have no concept of God, and even if they have/had beliefs in some superhuman entities, they do/did not hold that moral obligations come from their commands.

Yet, those people people did not seem to have any trouble understanding the concept of moral obligation, and they surely did not understand themselves to be making claims about the commands of any entity, let alone of God. Also, as before, we seem to have no difficulty understanding the claim that in the scenarios I constructed earlier in this section, Bob has moral obligations. Nor do we seem to have any particular difficulty understanding the use of moral terms in, say, Richard Swinburne's claim that "God has a moral obligation to make himself known".[r 30 ]

To be clear, Craig is not making a semantic argument. But the point is that regardless of what Craig is trying to achieve by quoting Taylor, the claims in question are at best unfounded, and given the previous considerations, we can tell also false.

16. Prudential reasons vs. moral reasons?

While not strictly part of the metaethical argument, in the context of his defense of that argument, and on the issue of prudential reasons, Craig challenges atheists as follows: [r2]

Craig:

“Prudence and morality are often at odds, and the question for the atheist would be: why act morally rather than out of self-interest? Why not just act out of prudence?”

There are a number of replies available to the atheist, but for instance:

1. Plausibly, a human being is a kind of being one of whose goals is to be a good or a better person – as a result of evolution, perhaps -, so it would always be favorable to some of a human's interests not to behave immorally, at least under a broad conception of interest, and with the potential exception of a very small percentage of pathological cases.

Now, Craig is using 'self-interest' in a way that excludes a person's interest in doing what's right, simply because it's the right thing to do, and/or in order to be a good person. That usage is common – in fact, probably the most common -, so that’s not a problem, but we ought to keep in mind what's meant by 'self-interest’; despite the wording, it's not the only interest people normally have.

2. It’s not clear that morality and prudence would be often at odds. Even if we exclude by stipulation all moral considerations from ‘prudence’, humans tend to be less happy if they feel guilty, so that’s a non-moral reason not to behave immorally, since immoral behavior tends to lead to guilt.

That said, that would still treat not behaving immorally as a means to an end, which is worse than trying to be good because it’s good, doing the right thing because it’s right, etc.

3. Let’s say that there is a case in which, say, a psychopathic dictator is pondering whether to behave immorally to get what he wants. Someone might ask: Why shouldn’t he act immorally?

In the moral sense of ‘should’, tautologically he shouldn’t act immorally.

But some of his interests would surely be served by acting immorally.
Would it be irrational on his part to behave immorally?

It's a matter of human psychology, it seems to me.

Perhaps, human minds are such that their deepest values, even if obscured due to errors and irrationality, are always in line with morality, and so it's irrational even for that dictator, all things considered, to behave immorally.

But maybe that is not always so. Maybe not all humans are like that, and the dictator is one of very few abnormal humans who sometimes might act immorally without acting irrationally. While I need not take a stance here, I would say that after reflecting on some hypothetical scenarios, it seems plausible that it’s possibly rational for some humans to behave immorally.

For example, let’s consider the following scenario:

In a distant, dystopian future, Jack, a psychopathic human – and so, a person -, has become the ruler of the planet. He uses robots to rule over everyone else, and no one else has access to advanced technology. No other human has the power to be any threat to him, and he enjoys – among other things – hunting humans for fun, and there are no threats from extraterrestrials or any other non-human beings.

In that scenario, Jack obviously has a moral obligation not to hunt humans for fun, he behaves immorally when he hunts humans for fun, and equivalently he ought to not hunt humans for fun in the moral sense of ‘ought’. However, intuitively it seems to me that Jack does not behave irrationally if he hunts humans for fun, or equivalently it is not the case that Jack ought to not hunt humans for fun, in the sense of ‘ought’ of practical rationality.

Moreover, in some other scenarios in which, say, a psychopathic dictator properly reckons that he would be at much greater risk of being ousted, tried and probably eventually executed if he stops oppressing people, murdering peaceful dissidents, etc., then it seems plausible that he rationally ought to continue engaging in such behaviors, even though clearly he morally ought not to continue engaging in such behaviors – whether he behaved irrationally on his way to becoming such a dictator is another matter.

At any rate, the matter of prudence and morality does not provide any good reason to think that the first premise of Craig’s metaethical argument is true, either, or generally to believe that God exists.

17. Accountability

Craig raises the possibility that someone can get away with evil, and uses that as what he calls a “powerful practical argument”[r 10] for theism. This is not part of the metaethical argument, since Craig does not claim that it supports the premises, as far as I know.

However, he makes this argument in the context of a defense of the claim that morality needs theistic metaethical foundations, so I will address it.

So, Craig maintains that if there is no God, there is no moral accountability, and that given death with no afterlife, it really does not matter how we live. [r 10]

However, even without any kind of afterlife, our choices affect our future, and the future of others; they can cause happiness, suffering, etc., to us and/or to other people, and of course that kind of thing normally matters to us and to many others. Our choices might not matter to some aliens from another planet, or to some creator, but that does not change the fact that they matter to many people.

As for accountability, there is in many cases moral accountability even if God does not exist, like the case of bank robbers going to prison. So, there does not need for an afterlife in order for justice to be done in many cases.

There isn't always accountability, but there is in many cases. Granted, sometimes there is not, and that’s that. But the fact that there is not always accountability without God does not provide evidence of the existence of God.

Now, it’s true that Craig does not attempt to use the practical argument alone. Instead, he proposes to use practical arguments to “back up or motivate” the acceptance of what he believes are sound theoretical arguments. However, that would still be epistemically improper, since a person would still be assigning greater probability to theism due to considerations that are not evidence for theism.

That aside, Craig maintains that lack of belief in God weakens moral motivation.

Even if that were true, it would provide no evidence for the existence of God, and so it would still be epistemically improper to assign a higher probability to the hypothesis that God exists than one would otherwise assign on the basis of a concern that one’s moral motivation would be weakened if one assigned a lower probability to that hypothesis. So, the argument fails. But there is another problem, namely the fact that despite Craig’s explicit recognition that there is no reason to think that atheists can lead “what we normally characterize as good and decent lives”, Craig’s motivational argument implicitly holds that, at least generally, not believing in God will tend to make a person’s moral character morally worse.

However, Craig has not given any good grounds to believe that.
The claim of a weakening of a moral motivation has not been established at all.

But additionally, and for example, those who adopt theistic beliefs usually do not just adopt some sort of unspecified theism, but some version of Christianity or Islam, with all the baggage of false beliefs – including false moral beliefs -, attached to them. Now, false moral beliefs – regardless of their source, religious or not – tend to lead people to behave immorally believing that they're doing the right thing. As an example that illustrates the case of false moral beliefs that result from religion and that lead to behaviors that are usually[4 0 ] immoral – like, say, publicly accusing people who are not doing anything wrong of behaving immorally -, one may point to Craig’s false moral claims against gay relations [r3], or the claim usually made by many Christians that it’s immoral not to believe that God exists and/or to believe that he does not exist – a claim that would be false even if God existed. [4 1 ]

Granted, not all Christians accept those particular false moral beliefs – i.e., in the cases of gay relations and unbelief -, but the point is that usually, theism leads to religions with many false moral beliefs, and many of the people who believe in them promote those beliefs, if not publicly by means of telling their children that those beliefs are true, etc. Given their religious source, changing those moral beliefs through reason may be even more difficult than it would otherwise be.

Granted, it might be argued that perhaps not-theists are on average more inclined to adopt other false ideologies, and that they too would have false moral beliefs as a result, etc., so the issues would then be questions like which ideologies are on average worse, etc. But the point is that Craig has not provided any good reasons to suspect that not believing that God exists, on average, leads a person to be morally worse than she would otherwise be.

18. Conclusions.

Craig’s argumentation fails to provide any good reasons to think that the first premise of his main metaethical argument is true. Moreover, some of the previous considerations lead us to the conclusion that it's false, as long as we take it to mean the same as the conditional if God didn't exist, objective moral values and duties would not exist, which is what Craig meant and so what is relevant in this context. [4 2][r 9] - though that result is not needed to properly reject Craig's metaethical argument.

Other metaethical or ethical arguments for the existence of God that Craig makes in the same context of one of the defenses of his main metaethical argument[r 10], also fail to establish the intended conclusion, or even to provide any support for it.

Notes:

[0]

A poster at www.freeratio.org who posts under the nickname ‘Bomb#20’ is one of the sources of some of the key ideas I use in the arguments above, but I do not claim that he would agree with everything I say above, since the article is a combination of ideas from different sources plus my own additions.

Another source are some of Eliezer Yudkowsky’s posts at www.lesswrong.com, though I do not claim that he would agree with the views I express in this article, either.

[1]

Elsewhere, Craig also talks about objective truth [r 6][r 7 ], objectively true claims [r 8], etc. In those cases, he does not try to define or explain ‘objective’, so it seems probable that he’s using the work ‘objective’ in an ordinary sense of the term. Some of the claims he makes seem to confirm that, though there is no need to go any further into that here.

[2] I'm not making a distinction between statements and judgments. Terminology varies, but it seems that the examples I give would qualify as statements and judgments in Craig's terminology.

[3]

Incidentally, Craig also claims that objectivity is a requirement for disagreement and error[r1], and surely there can be disagreement about whether, say, the cat in question is in pain, some people might erroneously believe that the cat is not in pain, etc. The same goes for fear.

Still, this particular claim is neither required nor decisive when it comes to understanding what Craig means by 'objective', and further, it seems the claim is not entirely correct.

For example, in the case of errors, let's consider the following scenario:

Alice tastes a cake, finds it yummy, and comes to believe that the cake is yummy. But as it turns out, the part that she tasted was a small portion with a composition different from that of the rest of the cake. Later, when she's eating another portion, she says to herself 'this isn't yummy at all: it tastes funny', and realizes that she was mistaken in her belief that the cake was yummy.

But Mary says that she tasted several portions and found them yummy, so Alice gives Mary the rest of the portion she found non-yummy, and Mary finds it yummy too.

In that scenario, plausibly whether the cake – or one of its portions – is yummy is a matter of taste, not a matter of fact. Yet, Alice was in error.

That said, while Craig's claim on this point seems to be exaggerated, the claims, arguments, attitude, etc., and generally behavior of the parties involved in actual moral disagreements usually show that they do understand themselves to be talking about a matter of fact, not about a matter of opinion, in the cases in which they're being sincere – which is also often the case.

That does not mean that they believe that there is always a fact of the matter, even in every conceivable hypothetical scenario, about whether a person [morally] ought to behave in such-and-such way, etc. Many people believe so – perhaps even most people – but that is not required to make sense of the kind of behavior displayed in specific cases of disagreement by people who are being sincere. For example, a belief that in all or at least nearly all cases that we might actually encounter there is a fact of the matter as to whether a person should do such-and-such thing would do, though there is no need to argue that point here.

On the other hand, I see no evidence that the sincere behaviors of people in usual cases of moral disagreement indicate that they implicitly believe in the mind-independence of moral duties and/or values, at least if 'mind independence' is understood in such a way that it would entail anything beyond what's entailed by there being a fact of the matter on the issue they're disputing.

[4]

The scenario may be understood as a possible world, if one accepts talk of possible worlds.

Else, one may consider the matter in terms one prefers, like scenarios, etc.

That aside, and just in case the definition is not clear, I’d like to clarify that in c onditio n OP2 , an agent is actual with respect to the world or scenario under consideration – i.e., that’s how the condition of actuality should be understood. The same goes for similar conditions in the case of morality, beauty, color, etc.

[5]

I’m talking about arguments in the sense of ‘arguing a case’ here, not formal arguments.

[6]

Craig also believes that a scenario in which God does not exist is metaphysically impossible. However, since arguing from metaphysically impossible hypothetical scenarios – or scenarios that one suspects and/or believes are metaphysically impossible – is not in general improper, that does not seem to provide a good ground for challenging Craig’s metaethical argument.

[7]

It might be argued that, in this context, the proper way to argue is that in such a scenario, they would not exist, rather than do not exist. I’m going by Craig’s most common formulation of the second premise, and I actually don’t think that this particular issue is a problem for his argument.

As I see it, whether use ‘would’ or ‘do’ in that context is a matter of writing style – as long as one’s clear about what one is talking about; I maintain that Craig is often obscure, but I don’t think that that particular issue is one of the factors that make his metaethical argument obscure.

So, I will not raise any objections to the use of ‘do not’, etc.

In any case, Craig also used the subjunctive formulation in his debate with Stephen Law.

[8]

An alternative would be to argue that, sometimes there is no objective fact of the matter as to whether some behavior is immoral, and sometimes there is. Would that mean that there are objective moral duties, or that there aren’t?

I will not address such variants, to simplify, but a deeper, more comprehensive analysis of certain metaethical issues would require just that.

[9]

I’m using the term ‘subjectivism’ in the sense of ‘subjective’ that Craig seems to be using, namely that it’s a matter of taste, etc.; depending on terminology, it seems to me such views might be classified as speaker-relativism as well.

[10]

If slight variations would only result in non-objectivity in very specific cases, ‘almost always objective’ may be properly called ‘objective’, and in any case, lack of objectivity is extremely rare cases may not be so problematic, so arguably the theist would have to show that there would significant variations without God, where ‘significant’ is sometimes difficult to evaluate, but it may plausibly be evaluated intuitively in most cases.

For instance, if there is some very slight variation in the color perception of people with normal color vision and as a result, on some very rare occasions, there is no objective fact of the matter as to, say, whether an object is pure green or bluish green, that would not entail that objective color does not exist, if we allow for some tolerance – just as we normally would say that an object is blue even if 1/1000000 of its surface is not blue, we may say that color is objective even if there is no objective fact of the matter in 1/1000000 of cases we encounter.

Still, a full analysis of this matter would exceed the scope of this article, so I will leave that aside.

[11]

By ‘ideal conditions’ I mean conditions in which the person has all the information about non-moral matters needed use her faculties and make a moral assessment of a given situation, and in which she is being rational and her faculties are not failing.

[12]

I’m not taking a stance here on whether there is a sense of right and wrong and of good and bad/evil, or a sense of right and wrong plus a sense of good and bad/evil, or some other combination of faculties.

In fact, I may sometimes talk about a sense of right and wrong, etc., to simplify the arguments, but essentially the same arguments, objections, etc., are compatible with the view that our faculty to tell right from wrong, etc., is actually the combined result of two or more mental systems, etc., which may even be the result of more general cognitive capabilities.

Those are complicated issues of moral psychology on which I’m taking no stance here.

[13]

Some of those religions that do not have any belief in the existence of God involve many false moral beliefs – of course, believers usually disagree -. For that matter, so does Christianity – of course, Christians usually disagree -, and other religions that do hold that God exists.

But in any case, the question here is not whether people in those cultures had or have many false first-order moral beliefs resulting from their religions, but whether in general humans traditionally had metaethical beliefs according to which moral duties were in one way or another based on the commands of God, goodness based on the nature of God, etc.

That is certainly not the case. In fact, even in the United States, or in Europe, Divine Command Theory is traditionally rejected by many Christians, so it's quite debatable whether Craig's claim about how moral obligations have been traditionally understood is true even there. But even granting it's true in the United States and/or Europe, it's certainly not true about human beings or human cultures in general.

[14]

By ‘post-human’, I mean any descendants of humans that have significantly changed, whether via the usual mechanisms by which our ancestors evolved into us, or by means of genetic engineering and/or integration of machines into their bodies. I'm making no claim here that there will be such beings, or that there will not, or about whether the actions leading to them will be immoral, if they were to happen.

[15]

https://en.wikipedia.org/wiki/Psychopath

[16]

Someone might ask: How would scientists measure that a person is violating other people’s rights, or that a person is morally good, etc.?

But for that matter, someone might ask how scientists would measure that a person is ill, or that an object is, say, green. Granted, in the case of green objects, someone might try to measure wavelengths, etc., and in the case of illnesses, there are lists of illnesses, symptoms, etc.

However, neither detecting illness nor detecting green – or greenness, if one prefers – was beyond the capabilities of scientists before they had such lists, and even the lists themselves were developed by means including using human color vision and a human intuitive understanding of illnesses in the first place.

The same goes for many other cases.

[17]

When I speak of ‘species#1-color’, ‘species#1-color language’, etc., I use that terminology to distinguish the visual system and the language associated with it from their human counterparts, but I do not mean to suggest that that’s the only species in the universe that has that visual system, or words associated with those specific wavelengths.

In fact, if the universe if large enough even if finite, there may well be perhaps many species with very similar or even almost identical visual systems, including many with a visual system like the human one, and similar language as well.

Similarly, my use of ‘species#2-beauty’ and ‘species#3-morality’ is meant to distinguish those species from humans in the matters relevant to the arguments under consideration. I do not mean to imply that their faculties, or their language, would be unique in the universe. The same goes, of course, for humans and morality. Other intelligent social species might have morality as well, even if not all of them did – a matter on which I’m not taking a stance here.

[18]

Of course, some other entities with a visual range from 440nm to 750nm might see things very differently. The example is just one possibility.

[19]

That raises moral issues depending on how it’s done, but that’s not the matter at hand.

[20]

Actually, there seems to be some slight divergence in the way people with normal color vision see some colors, which suggests that, in rare cases, there is no objective fact of the matter on color issues – e.g., there might not be an objective fact of the matter as to whether some object is pure green or bluish green, in some rare cases.

However, nearly always there is an objective fact of the matter as to what color an object is – i.e., in nearly all cases we encounter, that is so.

So, it seems to me that, speaking colloquially, it would be true that color is objective, since there seems to be some tolerance (e.g., we properly say that a car’s door is green, or made of metal, even if it’s not 100% sop, but 99.9999% so).

Still, this point is not crucial to the argumentation. We may reformulate the hypothesis so that if color statements are nearly always objective in the relevant sense and some statements like ‘X is green’, etc., are true, objective color exists.

Alternatively, and to simplify, we may assume that it’s not the case that there is a very small degree of divergence between the color vision of people with normal color vision, and as a consequence in the way people use color terms. While that probably wouldn’t be true, the point would remain that the objectivity of color would not be affected by the presence of such aliens, etc.; in fact, for that matter, we may consider different species of aliens with different visual systems in our scenarios, stipulating that there is no divergence at all within each of the alien species in terms of color-like perception, and the points would be the same, etc.

[21]

It’s compatible with the scenario – so, one might split it into sub-scenarios if one needs more examples, but I think that’s not required here – that in addition to the difference in what they find appealing, etc., their visual representation of their environment, if they have one, is considerably different from ours, and it’s also compatible that it’s not so different. But there is no need for such details for the purpose of the point I’m trying to make.

[22]

In this context, when I say something “like” species#2, etc., I mean something similar in a sense that would be relevant in the context of these argument. For instance, that they evolved in dark caves is not relevant, as long as there are great differences between what they find appealing which is associated with their judgments of species#2-beauty, and what humans find appealing and is associated with judgments of beauty.

[23]

Assuming non-determinism. Of course, on determinism, the result would have been the same no matter how many times one rewinds it, as long as the initial conditions are the same, though it would change in general with slightly different initial conditions.

It's not clear whether Craig is assuming non-determinism, or the ‘might’ is epistemic and there is no such assumption. But this is not crucial, so we may leave that aside. In fact, we may even assume non-determinism for the sake of the argument – non-determinism is Craig's position, regardless of whether he's implicitly assuming it in this particular context.

[24]

The Andromeda galaxy is perhaps too far away even for advanced aliens – I’m not entirely sure, though-, so personally I would prefer to set up the aliens to evolve on another planet in the Milky Way, but clearly galaxy of origin is irrelevant to the metaethical argument.

Also, rape is extremely improbable in that scenario (Why would they want to rape humans in the first place?), and eating humans is also quite improbable (why not a more efficient way of getting food, even if humans happened to be edible to them?), but let’s say for the sake of the argument – and to pick one of Craig’s examples – that they want to eat humans, and let’s further say that they want to eat humans because they like the way humans taste – or for any other reason suitable for the purposes of the argument – and so they invade the Earth and do so.

[25]

I make no claim as to whether there are aliens similar to that, or even whether they are nomologically possible. I’m just assuming for the sake of the argument that such aliens evolved.

[26]

Granted, someone might reply that they have different intuitions. But if that is the case, then what we a clash of intuitions, but that does not give me good reasons to accept their claim. Other people will of course make their own assessments.

In any case, if a theist claims that this would be problematic for non-theism, he would have the burden of showing why or how.

[27]

On that note, we may properly and truthfully say, in hypothetical scenarios involving aliens written as moral agents – not as species#3-moral agents -, that some of them behaved immorally in those scenarios, etc..

Also, we may properly make moral assessments in the case of the beings posited by different religions, as long as the religions in question state or entail that such beings are moral agents – e.g., a way of entailing it would be to claim that an entity does morally praiseworthy actions, or is morally good, etc.

[28]

There are a number of issues that might be raised at this point, but addressing them would be beyond the scope of this essay.

[29]

There is a difference here, namely the fact the term ‘meter’, was once defined in terms of the meter bar, whereas moral terms are not defined in terms of the term 'God', let alone in terms of an actual person. Furthermore, even if the judgment 'God is morally good' is, on Craig's understanding of the term 'God', analytical, also the term 'God' is not defined in terms of any actual person. But I'll leave that aside, at least for now.

[30]

More precisely, the informative identification would be the central part of an account of objective moral values/duties, and the ontological foundation would be the thing that moral values/duties would be identified with, or [some of] whose states/aspects/etc. moral values/duties would be identified with. But I don't think this slight notational abuse is a problem, so I will talk about 'ontological foundation in the sense of informative identification' and perhaps similar expressions, keeping this in mind.

[31]

My position is that a serial killer would not exist in a scenario in which God existed, and neither would a being who intends to be a serial killer, etc. But I'm leaving aside that for the sake of the argument, since I'm raising a challenge to Craig's theistic metaethical account that does not require an argument from evil against theism.

[32]

Some of the mental states or events on the list overlap, given the general "generally subjectively experiences", or "generally subjectively experienced", but I prefer to list some specific ones, in order to be more specific.

[33]

Granted, someone might say that their grasp of the concepts and/or intuitions say otherwise. But that kind of claim might always be raised. I invite readers to make their own assessments.

[34]

While the liquid in question is not pure water, the claim that it's water is true, using the words in a usual, colloquial sense (which includes some tolerance), and while picking a more precise expression would be doable if needed, it would seem to be needlessly long in this context.

[35]

Someone might argue that the 'greatest conceivable being' definition requires having a combination of great-making properties that achieves a maximal degree of greatness in every possible world, but perhaps does not require having all great-making properties to a maximal degree in case that that is not possible for some reason.

In that case, one may suitably modify the definition given that maximal moral goodness is required anyway, and the circularity objection remains.

[36]

For the sake of brevity, I will just talk about whether a foundation meets Craig's conditions, meaning the relevantly corresponding conditions.

For instance, in the case of heat, the condition corresponding to condition 5 – in case Craig requires condition OMOMV5 for moral goodness – would be a paradigmatic object or paradigmatic case of molecular motion which determines heat in a way similar to which, according to Craig, there is a paradigm that determines goodness.

[37]

C1 might be used to argue against any theistic accounts of proper function, but I'm leaving all of the arguments in subsection 13.2 aside in other subsections, for the sake of the argument, because I aim to give independent and sufficient reasons for rejecting Craig's argument.

[38]

Molecules are in turn identifiable with some interactions between atoms and/or configurations of atoms, which in turn are identifiable with some interactions between subatomic particles and/or configurations of subatomic particles, and for all we know, the subatomic particles we know of might not be the bottom. In fact, there might not even be a bottom – what if it's turtles all the way down?

But let's leave all of that aside, to simplify, and since I don't need to focus on that issue.

[39]

In the case of H2O, that's a description of a molecule at the level of atoms, but an identification with interactions of and/or configurations of subatomic particles is available. Moreover, for all we know, the subatomic particles we know of might not be the bottom. In fact, there might not even be a bottom – what if it's turtles all the way down? But let's leave all of that aside, to simplify, and since I don't need to focus on that issue.

[40]

I say “usually” because one can easily construct scenarios in which that’s not immoral; for example, there are scenarios in which someone claims that gay sex is always immoral because someone else is making a serious and sufficiently significant threat if she does not make that claim, and she's not doing anything immoral by making that claim – she's just under a serious threat.

But that's not Craig's situation, who in addition has had decades to assess his moral beliefs, yet continues to promote the false moral beliefs that non-belief in God is always immoral, that gay sex is always immoral, etc. - 'always' at least for mentally competent human adults, that is.

[41]

Of course, Christians usually disagree with my moral assessment on that and many other matters involving Christian-based moral beliefs. On a related note, a link to that a moral case against Christianity can be found here.

[42]

If we take the first premise as a material conditional, in order to show it's false we would have to show that God does not exist, which is beyond the scope of this essay. But Craig did not intend it as a material conditional. Rather, Craig intended the first premise to mean "If God did not exist, then objective moral values and duties would not exist". In fact, that is the formulation he chose in his debate with Stephen Law.[r 9 ] Moreover, Craig says that he thinks that the first premise is a true counterpossible, and that it means that if God did not exist, then objective moral values and duties would not exist. [r9b]

References/sources:

[r1]

http://www.reasonablefaith.org/defenders-2-podcast/transcript/s4-19

[r2]

http://www.reasonablefaith.org/defenders-2-podcast/transcript/s4-20

[r3]

http://www.reasonablefaith.org/a-christian-perspective-on-homosexuality

[r4]

http://www.reasonablefaith.org/the-plausibility-of-grounding-moral-values-in-god

http://www.reasonablefaith.org/does-theistic-ethics-derive-an-ought-from-an-is

[r5]

http://www.reasonablefaith.org/is-the-foundation-of-morality-natural-or-supernatural-the-craig-harris

[r5b]

http://www.reasonablefaith.org/defining-god

[r6]

http://www.reasonablefaith.org/do-we-live-in-a-post-modern-society

[r7]

http://www.reasonablefaith.org/are-there-objective-truths-about-god

[r8]

http://www.reasonablefaith.org/the-revolution-in-anglo-american-philosophy

[r9]

http://www.reasonablefaith.org/does-god-exist-the-craig-law-debate

[r9b]

http://www.reasonablefaith.org/formulating-the-moral-argument

[r10]

http://www.reasonablefaith.org/the-indispensability-of-theological-meta-ethical-foundations-for-morality

http://www.leaderu.com/offices/billcraig/docs/meta-eth.html

[r11]

http://www.reasonablefaith.org/defenders-2-podcast/transcript/s4-21

[r11b]

http://www.rfmedia.org/RF_audio_video/Defender_podcast/20040801MoralArgumentPart3.mp3

[r12]

http://www.reasonablefaith.org/our-grasp-of-objective-moral-values

[r13]

http://www.reasonablefaith.org/hawkings-curious-objections-to-divine-creation

[r14]

http://philpapers.org/rec/HORNWM

Horgan and Timmons, “New Wave Moral Realism Meets Moral Twin Earth”; Journal of Philosophical Research 16:447-465 (1991)

[r15]

http://philpapers.org/rec/HORAMF

Horgan and Timmons, “Analytical Functionalism Meets Moral Twin Earth”, in “Minds, Ethics, and Conditionals: Themes from the Philosophy of Frank Jackson”, edited by Ian Ravenscroft. Oxford University Press (2009).

Link:

http://thorgan.faculty.arizona.edu/sites/thorgan.faculty.arizona.edu/files/Analytical%20Moral%20Functionalism%20Meets%20Moral%20Twin%20Earth.pdf

[r16]

https://en.wikipedia.org/wiki/Visible_spectrum

[r17]

http://www.reasonablefaith.org/contemporary-moral-arguments

[r18]

http://www.reasonablefaith.org/moral-argument-for-god

[r19]

http://www.reasonablefaith.org/classifying-immaterial-objects

[r20]

William Lane Craig, “The Most Gruesome of Guests”, in “Is Goodness Without God Enough: A Debate on Faith, Secularism, and Ethics”, edited by Robert. Garcia and Nathan King.

[r21]

Mark Murphy, “Theism, Atheism, and the Explanation of Moral Value”, in “Is Goodness Without God Enough: A Debate on Faith, Secularism, and Ethics”, edited by Robert. Garcia and Nathan King.

[r22]

http://www.reasonablefaith.org/can-god-ground-necessary-moral-truths

[r23]

https://en.wikipedia.org/wiki/Giorgio_Perlasca

[r24]

http://www.reasonablefaith.org/perfect-being-theology

[r25]

http://www.reasonablefaith.org/defenders-2-podcast/transcript/s4-23

[r26]

http://www.reasonablefaith.org/defenders-2-podcast/transcript/s4-24

[r27]

http://www.reasonablefaith.org/the-euthyphro-dilemma-once-more

[r28]

http://www.leaderu.com/offices/billcraig/docs/craig-taylor0.html

http://www.leaderu.com/offices/billcraig/docs/craig-taylor2.html

[r29]

Richard Taylor, "Ethics, Faith, and Reason". Prentice-Hall, Englewood Cliffs, NJ: 1985.

[r30]

Swinburne, Richard "The Existence of God", Second Edition. Clarendon Press Oxford. Page 130.

Angra Mainyu's blog

Pages

Friday, June 7, 2013

A Reply to Craig's Metaethical Argument (aka Craig's "Moral Argument") II (v. 7.4)