The Problems of Evil and Poor Design

I think the problem of evil can be made most forcefully when amalgamated with an argument from poor design. This is essentially the move Paul Draper makes with his Bayesian argument from evil. Draper notes that pain and pleasure serve a biological role that makes sense on a hypothesis of indifference (and is arguably a prediction of evolutionary naturalism), but is incredibly surprising under the hypothesis of a morally conscious creator.

The idea is that some features of the human design are morally significant, so that a morally concerned designer would have reason to discard or, at least, revise such design. But on naturalism, the forces of evolution are indifferent to the moral significance of the design it produces.

For another example consider how, compared to most other animals, giving birth is incredibly difficult and even dangerous for humans. On evolutionary naturalism the explanation is obvious: there were very powerful evolutionary pressures on our ancestors to have narrow hips to facilitate our bipedal mobility, and very powerful evolutionary pressures for humans to have big heads to facilitate our big brains and great intelligence. But both these traits together obviously mean a lot of trouble for birth. A morally conscious designer would see that this will cause a lot of unnecessary death and suffering and design things differently, but evolution is blind to these things as long as they don't impede reproductive fitness.

We then have two competing hypotheses: an anthropic origin by evolutionary naturalism, or by a morally conscious creator. And since each entails the negation of the other, we have strong reason to think there is no morally conscious creator.

A Moral Argument Against Theism

It's commonly thought by theists that, in the absence of God, one is faced with difficult ethical or meta-ethical problems. Within the atheistic world view, morality itself, moral knowledge, moral motivation and so on are thought to be on shaky ground. Such beliefs are the basis of moral arguments for theism. Here then is a similar argument in the reverse: the existence of God, along with three indubitable and commonly held ethical beliefs, leads to a vexing contradiction.

1. It is wrong to treat people as a means to an end, rather than as an end in of themselves.
2. God (a perfect being who can do no wrong) allows innocent children to suffer and die for some greater end.
3. Therefore, that greater end must be greater for the child: it must be in his own best interests to suffer and die.
4. We should always aim to do what is in the best interests of others, and avoid doing what isn't in others best interests.
5.  Therefore, we should not attempt to prevent the unbearable suffering or untimely death of innocent children.

But of all things, the unbearable suffering and untimely death of children seems to be something we should always aim to prevent. To deny this is absurd, so one of the premises must be rejected. Premise (1) is a statement of Kantian ethics, a very popular normative ethical theory. It seems too obviously true to reject. Premise (2) is true if God exists at all. To reject this would be to reject theism. Premise (4) seems to follow from our very a priori concept of a moral should. These are the only three premises.

What makes this argument interesting is that (1), (4) and ~(5) are almost impossible to reject. If anything could be taken as self-evident, surely these statements should be. God's existence, on the other hand, is not self-evident. Even if theism is very well evidenced, it's surely not as well supported by our intuition as these three other statements. Therefore, we must reject theism in favour of these three induibiable ethical statements.

Divine Nature Theory and the Parsimony Problem

Divine nature theorists believe that goodness is grounded in a likeness to God's essential nature. As they would say, it is good to be loving and kind and just because God is loving and kind and just. In other words, the property goodness reduces to being a member of the set of properties that God holds necessarily. But one might wonder, why do we need God at all? Wouldn't it be simpler to say that goodness reduces directly to that collection of properties, without referencing God's nature? Why shouldn't we rather say that what it is to be good is just to be loving or to be kind or to be just and so on? This revised theory would seem to have all the same explanatory power as divine nature theory, but would be a great deal more parsimonious for not requiring the existence of God.

I once suggested this to a friend, and he responded by saying that in such a case there would be nothing all good things had in common. But this seems odd: surely for there to be such a thing as goodness, there must be something all good things have in common. Intuitively, he thought, there must be something that brings all these properties together (and in the darkness binds them). At the time I was unaware of the classic counter examples: species being an obvious case of a bundle property, being made up of a collection of properties which do not necessarily all share something in common. There are no necessary and sufficient conditions for being, for example, a lion. By analogy, we might think there need not be necessary and sufficient conditions for being good.

On top of this I now think there are two even stronger objections to this line of reasoning:

First of all, divine nature theory itself suffers from the same problem my friend had with treating goodness as a bundle property. Why is this? Because being like God is a bundle property. There is nothing all of God's essential attributes have in common, other than being those attributes God necessarily instantiates. The only difference is that we've given this collection a name ("God's essential nature") and ground goodness in a likeness to it. Theists may suggest that on greatest being theology, the divine nature theorist could think that all God's essential attributes have in common their being great-making. But greatness is no less metaphysically queer and demanding of explanation than goodness, so this move would only push the need for grounding a step back without really explaining anything substantial.

Secondly, there is something all good things have in common without God after all: their normativity. If something is good, then there is reason to conduct oneself towards it certain ways. There is reason to desire good things, to perform good actions, and to praise the good behavior of others. Furthermore, most would say this follows from our very concept of goodness itself. We therefore need not talk about God's essential nature to explain what it is to be good.

Divine Nature Theory: Why goodness is not a likeness to God

Divine nature theory is what I call the common theistic meta-ethical opinion that goodness is grounded in a likeness to God's essential nature. God is essentially loving and kind and just, which is taken to explain why it is good to be loving and kind and just and so on.

But there is an obvious counter example to the theory as so far construed: God has many essential characteristics that are not morally significant. God is essentially immutable, necessarily existing and ontologically self-sufficient. But it would be absurd to say that something was good in virtue of its being immutable, necessary or having aseity.

The divine nature theorist might try to revise his view, saying instead that a thing is good just in case it bears some resemblance to God's essential behavioral qualities. Necessity, immutability and aseity have nothing to do with behavior, and so they pose no threat to this view. And yet, other counter examples can still be raised. God is essentially rational, prudential, decisive, and is essentially not impulsive or whimsical. But it seems silly to think that being irrational or imprudent would have any affect on the moral value of ones action.

The divine nature theorist might take a further step back, saying instead that a thing is good just in case it bears some resemblance to God's interpersonal behavioral qualities. Being rational doesn't have anything to do with how one treats others, but lovingness and kindness and justice does.

But now there's an even more pressing problem, in that these qualities don't seem like essential characteristics of God. God cannot be necessarily loving or just, since there are no other necessarily existing people to be loving or just towards. Some have suggested positing the trinity to solve this very problem, saying God's persons necessarily love each other. But love is just one of many morally significant properties the theist is trying to ground in divine essence. What about other morally significant behavioral qualities, like being someone who punishes the wicked, or who protects the defenseless? There are no necessarily existing wicked or defenseless people for God to stand in the appropriate interpersonal relationships with.

The divine nature theorist might try one last revision. He might suggest that goodness is grounded not in any qualities God actually has, but in counterfactuals about what God would do given the opportunity. And so protecting the defenseless is good because, given the opportunity, God would always protect the defenseless.

But now the divine nature theorist has made God entirely dispensable to his theory. Even atheists can believe that, if God were given the opportunity, then he would protect the defenseless. Divine nature theory, then, reduces to something very much like ideal observer theory, and is no longer an inherently theistic account of goodness.

Possibility of the Actually Infinite

Mathematicians define two sets as being the same size whenever there is a bijective mapping from one to the other. That is to say, whenever each member of the one set can be paired up with exactly one member of the other. At face value this makes perfect sense. If you're at a dinner party and every guest has brought a significant other of the opposite sex, then clearly there are just as many men at this party as there are women.

Why does this matter? Because, if same size is understood as mathematicians define it, then it easily follows that a set can be the same size as its proper subset. In other words, the common objection to the possibility of actual infinities simply fails by definition.

It's all the more clearer with an example. The natural numbers {1, 2, 3, ... } and the even numbers {2, 4, 6, ... } are the same size because a bijective mapping exists between them: f(x) = 2x, which pairs 1 with 2, and pairs 2 with 4, and 3 with 6 and so on. And, of course, the even numbers are a proper subset of the natural numbers.

Now you might think, this sounds super abstract. Maybe this is some weird technical idea that mathematicians throw around, but surely all this mathematical mumbo jumbo isn't anything like my understanding of what it means for two sets to be the same size. But actually, it is.

When someone wants to know how big a set is, they count the members. They point to an object and say "one", they point at another and say "two", and so on. They don't know it, but they're proving the existence of a bijective mapping between the set in question and the subset of the naturals they're vocally describing. And because the size of the set they're counting out is identical to the value of the last member, they (all by intuition and not understanding the math) infer the size of the set in question. It's odd to think that something so simple as counting has a rigorous mathematical basis that utilizes the mathematicians very technical definition of same size, but it in fact does.

Of course we needn't proceed one member at a time when counting. We could just as well point to a pair and say "two", and another pair and say "four", and so on. Or we could go by groups of ten, or hundreds or, even, the collection as a whole. It follows from this that, contrary to popular opinion, we can count an infinite set after all, we just do it all at once instead of step by step.

Here's the catch, then. People who think that actual infinities are impossible (almost always because they think it's impossible for a set to be the same size as its proper subset) owe the rest of us an explanation of what 'same size' means, if not what mathematicians mean. Since they are unable to give any answer, their objections to actual infinities (usually to the possibility of an eternal past) fall flat.

Evil as an Absence of Good?

It's very common for people to think that evil doesn't actually exist in of itself. Rather, like a hole in your shirt is just an absence of fabric, some would say evil is just an absence of good. Of course that doesn't mean there is no truth about evil, or that it wont affect your life. A large hole in your jacket, despite not being a thing in of itself, will still make you miserable on a cold, wet winters day.

But there is an obvious problem with this view, in that it would require all things (at least within the relevant domain) to be either good or evil. You can't have shirts that are neither whole nor have a hole, but you can have actions that are neither good nor evil—in fact most actions seem morally insignificant. What could be the moral value in cutting your grass, or eating a cheese burger? If you find some morally significant feature, one can always easily stipulate a scenario in which that feature isn't present.

And so defenders of this view must embrace the implausible and maintain that every action, no matter how seemingly insignificant, has moral value to some degree. This is why I favour theories of moral ontology on which good and evil are both real in a robust sense, neither being an absence of the other.

Skeptical Theism and Divine Deception 2

I have become convinced that my argument outlined in Skeptical Theism and Divine Deception is not successful.

The problem is that I failed to distinguish between having justified belief, and being able to justify ones belief. The difference is that we can have justified belief without being aware of it. On externalism, the justification for a belief can be something that the subject might not even have access to, like the causal history that produced his belief. But justifying ones belief is an action rational people perform, a sort of giving of an explanation or an account of how one is justified in holding that belief.

It follows from externalism that, for all I know, the skeptical theist might be justified in believing God always tells the truth—my premise (3) is indefensible. And yet the spirit of my argument persists. If you think about the role divine revelation plays, it's always intended to account for the justification of religious belief. Religious folk would say "God has told us these things and therefore they are true," implicitly assuming that God is not lying.

But now it's clear that the implicit assumption is not plausible so long as we're committed to skeptical theism. After all, for all the skeptical theist knows, God could have a morally sufficient reason to lie about religious matters. (See the previous post for a more in depth analysis of this).

So while the skeptical theists' religious commitments might be justified, they are not something he can justify. And, since we should exercise a healthy skepticism about beliefs we cannot ourselves justify, there is still tension between skeptical theism and religious belief. It seems, at least, the skeptical theist should be no more confident about his religious beliefs than he thinks is appropriate for gratuitous evil.

We can re-formalize this argument as follows, where P is the sort of belief we can only justify by appealing to divine revelation (namely, religious belief):

1. Without appealing to divine revelation, there is no way to justify the belief that P
2. Skeptical theists cannot appeal to divine revelation to justify their beliefs
3. Therefore, skeptical theists cannot justify their belief that P
4. Doubt should be reserved for beliefs we cannot justify
5. Therefore, skeptical theists should reserve doubt for their belief that P