Evaluative Claims within the Problem of Evil

When it comes to the problem of evil, I’ve always felt there was something wrong with someone trying to argue for a factual conclusion – that God does not exist – on the basis of moral premises. The conclusion is something that sounds an awful lot like a statement of a fact, a description of some state of affairs in the world, like a claim that ‘no human being that exists is over 10ft tall’, but the grounds for this fact are at times little more than a report of your moral feelings. Ordinarily, we wouldn’t give this kind of argument the time of day.

‘No human being that exists is over 10ft tall, because it would be really unfair to the game of basketball.’

Facts and Values

When I came to present this argument with more rigour (philosophically nit-picky though it was), I cited it as an inconsistency in J. L. Mackie’s thought. In The Miracle of Theism, he argues against moral arguments for the existence of God on the basis that they argue from values to facts, and that this gets the direction of supervenience between facts and values back to front. And yet he also offers the problem of evil as an argument against belief in God, which surely repeats the same error (if it is an error).

I stand by that argument and the other lines of argument in that article, but I don’t think it needs to be as complicated as all that. Some illustrations can make the underlying point clear.

Imagine an atheist offers a version of the argument from evil. For example, they say that there is something essentially morally wrong with God violating the ‘Pauline Principle’ that we should never do evil so that good may come. But as far as they can see, God, in creating the world with so much evil and suffering but with the purpose of everything coming good in the end, does violate this principle. The only sensible conclusion to this paradox is that God does not exist.

But imagine God appears to this atheist and sets them right: ‘Where were you when I laid the foundations of the earth?’, etc. The atheist’s original paradox remains, perhaps, and they will certainly feel confused: it remains the case that they can’t consistently believe both that it is wrong for God to do this thing and that God, the morally-perfect being, does it. But can they still maintain their original conclusion?

Of course not, because God’s existence has just been confirmed. What is their alternative, to avoid paradoxical inconsistency? They must conclude that they were mistaken in their moral beliefs. In light of new information, they change their moral minds. They align their value judgements with the facts. It turns out God can violate the Pauline Principle under certain circumstances. The atheist learns a valuable lesson and continues forwards, as a theist, with a more accurate understanding of the world.

An atheist (who is an atheist on the basis of the problem of evil) who learns that God exists must take a very different view of the moral world. Such a piece of information radically changes the shape of their moral world. The fact of God’s existence can and does and must affect their value judgements, because value judgements ought to be aligned with the facts, and especially if those value judgements are shown to be inconsistent with the facts.

But it doesn’t work the other way around. The discovery of a new moral principle cannot make it so that God does not and has never existed. That is not something that is within the power of values to do. The fact of God’s existence changes what is true or false about moral beliefs, but moral beliefs do not have the power to change the fact of God’s existence.

Logical Priority

Moral beliefs cannot change the facts; they can only change how you perceive those facts or how you feel about them. But facts can force you to change your moral beliefs.

Facts have a kind of logical priority over values: facts come first, then we try to align our value judgements with the facts. Because of this logical priority, if there is a contradiction between facts and value judgements, the value judgements must be rejected or modified, because value judgements do not have the power to change the facts; and yet we cannot remain with an inconsistent set of beliefs, so something must be done.

Mackie on Kant

When J. L. Mackie makes this point, he does so in opposition to the moral argument for theism. Many atheists are keen to reject any of the moral arguments for belief in God for the reason that you cannot argue a point about how the world is on the basis of a set of beliefs about how the world ought to be.

Just because the world ‘ought’ to be just and good does not mean that it is just and good, and certainly doesn’t imply that there must be something that makes it just and good.

Mackie’s primary target is Kant. (He does also target Newman, but he uses a different argument there.) In this, Mackie picked the high-hanging fruit; easier targets were available and would have been more appropriate for his line of argument, which does not fit quite as easily with Kant’s peculiar transcendental context.

As Mackie deals with Kant’s argument out of that peculiar context, Kant’s moral argument for belief in God is really an argument about practical consistency: It would be inconsistent to believe that you ought to uphold the moral law unless you also believed that you can uphold the moral law. Or at least, it would be practically impossible to maintain a commitment to the moral law without some assurance of goodness and justice in the end.

According to Kant, the function of human reason is a good will or reverence for the moral law, and human beings are naturally (essentially) desiring of happiness. We need to do our duty and we want to be happy. Were it to be the case that these two essential drives contradicted each other in some way, then human beings would be doomed to a tormented absurdity. We must, therefore, on pain of tormented absurdity, believe that goodness and happiness and justice will align in the end. And the only way this can be is if there is a God who can make it happen. God is a necessary postulate of practical reason.

Mackie thinks this is a bad argument because you cannot argue for a fact (even if only as a ‘postulate of practical reason’) on the basis of a set of values. This gets the direction of supervenience between facts and values back to front. Values change in response to the facts, but the facts do not change in response to values.

Facts First

The same principle is obvious beyond the context of the problem of evil. Imagine we say it is virtuous to give money to someone living on the streets. We think we do them good. This is a value judgement.

Then someone comes along and shows us that, as a matter of fact, we are doing them more harm than good because they will only use the money to buy drugs.

Say we accept this as a fact. Can we hold to our moral belief that it is virtuous to give money to someone on the street? Can we argue that we do in fact do them good by giving them money because we have decided it is the virtuous thing to do?

‘It would be wrong to give money to someone living on the street if they will only use that money to buy drugs; it is right to give money to someone living on the street; therefore, they will not use the money only to buy drugs.’ This argument runs: value, value: fact. It runs: ‘ought, ought: is’. That would get things back to front.

Our beliefs about what is good to do depend on the facts of the matter, but the facts of the matter do not depend on our beliefs about what is good to do.

Imagine someone fatally shoots someone. They point a loaded gun, pull the trigger, fire a bullet, and the bullet hits a person, killing them dead. We make a value judgement. Such a thing ought not to be done. We judge the shooter badly. We judge in response to the facts.

But if the facts change (or rather, if our understanding of the facts changes) then we change our value judgement. If we learn that the shooter had good reason to believe the gun wasn’t loaded, or that they had no way of knowing they were pointing the gun at a person, or that the bullet was meant to be a blank, we do not judge the person in the same way.

Our value judgements respond to the facts. We try to establish the facts in order to make the right value judgements. We try to align our value judgements with the facts.

But the facts are not so kind as to align with our value judgements. No matter how much you tell it that it would be wrong to hit that person, or how clearly you hold that value judgement to be true, a bullet will not alter its course. It will not misfire or change into a blank for the sake of moral consistency. The shooter cannot say ‘it would be wrong to shoot this bullet at someone; I will not do anything wrong; therefore this bullet will not shoot at someone (even if I point the loaded gun and pull the trigger)’. This argument runs: value, fact: fact. It runs ‘ought, is: is’. But that’s not how the world works.

Our beliefs about what ought to be done depend on our beliefs about what is the case, but our beliefs about what is the case cannot depend on our beliefs about what ought to be done. Our moral beliefs do not have that fact-impacting power.

The Moral Argument for Theism

But isn’t that essentially what theists are doing when they present a moral argument for the existence of God?

‘It would be wrong for the world to require goodness without reward; the world requires goodness; therefore, there must be reward’. This argument runs: value, value: fact. It runs: ‘ought, ought: is’. But just because the world ought to be a certain way doesn’t mean that it is.

The moral argument for the existence of God erroneously argues from values to facts.

Similarly erroneous would be the attempt to argue for the existence of God on the basis of the existence of objective moral values. The moral law could not exist without a divine lawgiver, or so it is claimed. If we recognise a moral law, we must recognise it as divinely-given. Setting aside the clear falsity of this claim – because there are many ways we could understand having obligations to one another without needing it to be divinely-mandated – it would remain a case of arguing for a fact on the basis of a value judgement. ‘Murder is not wrong unless God exists to command us not to murder; murder is wrong; therefore, God exists.’ Whether the first premise here is considered a fact (of an a priori nature, presumably) or a value I leave undecided, but even if we concede it as an analytic fact, this argument would run: fact, value: fact. It runs: ‘is, ought: is’.

The Argument from Evil

Is the argument from evil to the non-existence of God any different? It doesn’t seem to be: ‘It would be wrong for a good and powerful God to permit evil; evil exists; therefore, God does not exist.’ This argument runs: value, value: fact. It runs ‘ought, ought: is’.

The argument from evil to the non-existence of God argues from values to facts. If it is wrong to argue for the existence of God on the basis of values, why isn’t it also wrong to argue against the existence of God on the basis of values?

For this reason I would say that the argument from evil to the non-existence of God is fundamentally flawed, exactly as Mackie says the moral argument for the existence of God does not work. You cannot argue for a fact from a value.

Consistent Beliefs

Perhaps we could say that, in contrast to the argument from evil, the problem of evil is not about arguing for a ‘fact’ but establishing a consistent set of beliefs. Obviously I would be sympathetic to this. But in that case is the moral argument for belief in God any different? Isn’t that also about establishing a consistent set of beliefs about how to live in this world?

You need to do your duty and you want to be happy: to avoid the tormented absurdity that follows from these two being in any way contradictory, you need to find a way to make them consistent. Belief in God would make them consistent. So would Socrates’ claim that virtue is sufficient for happiness. Or else you could abandon your pious duty and try only to be as happy as you can (Epicurus can show you how). So long as these are value-based inferences from value-based claims then there is nothing obviously wrong with this.

The mistake only emerges once you try to infer a fact from these. That’s like crossing the is-ought gap but the wrong way; not inferring an ought from an is, but an is from an ought.

If Socrates were to say that he was bullet-proof, on the basis that he is a good man and ‘a good man cannot be harmed’, then we would say he has made an obvious mistake: his value judgements, however true, cannot determine the facts of the matter. Mackie would say that any moral argument for the existence of God makes the same mistake, though less obviously. I would say that the argument from evil to the non-existence of God makes the same mistake again, and the problem of evil shares this mistake if it infers a fact in an attempt to resolve an inconsistency between facts and values. You cannot infer an is from an ought.

Conclusion

What is the upshot of this? It is that any attempt to propose the problem of evil as a good reason for atheism is doomed to failure.

If you think that an assertion of God’s non-existence is describing a fact-of-the-matter, not merely an expression of my feelings or the truth as I see it, then any move from the value claims made within the problem of evil – those claims that state what ‘ought to be’ the case – to a solution that is making a claim about what ‘is’ the case, will inevitably get the direction of supervenience between facts and values back-to-front.

Postscript: Moral Necessity

As I explain in my longer-form piece on this topic, I don’t think the problem of evil is a reason for atheism: I think the problem of evil is an ethical problem that teaches us about moral necessity, among other things. And so if you look at these arguments with an eye for moral necessity, they make more sense.

Take the moral argument for theism, for example. Is this an attempt to establish the fact of God’s existence? Of course not (and especially and explicitly not for Kant).

What the moral argument for theism expresses is that the believer finds it to be morally impossible not to believe in God. This is not a statement of fact, like ‘God exists’. It is a statement about their own moral judgements or beliefs, about the limits of their moral world. For them, their moral world is defined by their belief in God: therefore, for them, there is no moral world without that belief in God. Belief in God is a foundation stone, a hinge around which their moral reasoning swings. These are expressions of value, absolute value, not of fact.

How right would it be to treat this as if it were only an intellectual matter of fact? It would miss the point to do so; it would reduce the argument to such an impoverished form that it wouldn’t be worth giving it the time of day. The conclusion of the moral argument is not that God’s existence has been demonstrated but only that the believer is firm in their moral resolve and equally firm in their hope that this resolve is not foolish.

I say the problem of evil should be understood exactly like this. It is an ethical problem that tells you about your moral limits. The conclusions of these ‘arguments’ are not facts, but values, absolute values.

Perhaps those values will, in the end, prove to be so incompatible with belief in a good and powerful God that you find such theistic belief impossible, but the impossibility will be of a distinctively moral kind. Of course you understand that it would still be metaphysically possible, on the basis of those moral reasons, that God exists.

Due to the fundamental role that value judgements play within it, the problem of evil cannot conclude with a fact-type claim. It can only conclude with a value judgement. The conclusion of the problem of evil, either ‘for’ or ‘against’, ought to be expressed as a value judgement only. It tells us nothing whatsoever about the fact of God’s existence.

The problem of evil does not tell you anything about the world because it cannot do such a thing. It can only tell you about yourself and your moral beliefs. This is why it ought to be treated as an ethical problem.

Read more: The Problem of Evil as an Ethical Problem