Book is here. Prepublication draft below.
The problem of evil is widely considered to be one of, if not the, most significant threat to traditional theism, by which I mean the kind of theism that posits a being that is omnipotent, omniscient, and omnibenevolent (roughly: all-powerful, all-knowing, and all-good). This volume focuses on what's often called the Evidential Problem of Evil, one version of which runs as follows:
If God exists, then there are no gratuitous evils
Gratuitous evil exists.
Therefore, God does not exist.
A 'gratuitous' evil, in this context, is an evil for which there is no adequate God-justifying reason. Yes, God might allow some evils if that's the price he must unavoidably pay to allow for still greater goods. But God won't allow gratuitous evils - evils that would be pointless from God's perspective.
Notice that this simple argument is deductively valid: necessarily, if the premises are true, then so is the conclusion. Theists cannot consistently accept both premises while denying the conclusion. Typically they reject the second premise.
In this foreword, I thought I'd sketch out a map of my own thinking on the Problem. I hope it will provide a useful overview of the issues and a helpful point of comparison when reading the book.
It seems to me that, in response to the Evidential problem, theists typically pursue at least one (and sometimes all) of the following four strategies:
1. Theodicy strategy
The first strategy is to identify reasons that would justify God in allowing observed evils, thereby showing that they are not gratuitous evils after all. Such explanations for why God permits such evils are called theodicies. Two classic examples of theodicies are:
Free will. God gave us free will, which is itself a great good, and which also allows for the further great good of our being able to freely choose to be good. God could have made us puppet beings always compelled to behave well, but then we would not be morally responsible for our good behaviour. God cut our strings and set us free so that we might freely choose the good. The downside to this is that we sometimes choose to do evil, not good. That evil is the price God must inevitably pay to allow for these greater goods. And so the evil is not gratuitous.
Character building. Sometimes the horrendous pain and suffering humans experience can make us grow stronger and become better people. Those who have survived an awful experience sometimes say they don't regret having endured it because of the valuable insights it allowed them. Further, the pain, suffering and poverty of others can provide us with opportunities to be compassionate and charitable. Compassion and charity are so-called 'second order goods' that require 'first order evils' like suffering and poverty. Such evils can be explained as a consequence of God giving us these important opportunities to develop morally and spiritually.
While these and many other theodicies have been offered to explain why God allows the evils he does, such explanations appear - even to many theists - both individually and collectively deeply inadequate.
For example, the appeal to free will explains, at best, only those evils for which we're responsible as free moral agents. What of the hundreds of millions of years of horrendous animal suffering before we humans showed up? What of natural diseases and disasters and the suffering they cause? What of the horrendous suffering of parents and children caused by childhood mortality rates of around 50% for almost the entire 200,000 year history of homo sapiens? None of this is a product of free will.
The character building explanation also fails to account for most animal suffering. In addition it fails to explain the suffering of the vast numbers of humans who have bowed out of this life both physically and psychologically crippled by the experiences through which they have been put. Their characters have not been built. They've been destroyed.
If there exists even a teaspoonful of gratuitous evil, then there is no God. The various theodicies that theists have constructed appear, both individually and collectively, to fall spectacularly short of explaining the horrendous pain suffering of humans and other sentient creatures. There remains rather more than a teaspoonful of gratuitous evil to explain away.
2. Skeptical Theism strategy
A second strategy theists employ to try to deal with the Evidential Problem is to insist that we can't reasonably assign a low probability to there being God-justifying reasons for all the evils we observe. We can't reasonably insist that it's unlikely there are such reasons just because we cannot think of such reasons. Theists often offer analogies at this point. If I look into a garage from outside and cannot see any elephants in there, then it's reasonable for me to believe there are no elephants in there. But if I look in and cannot see any insects, it is not reasonable for me to conclude that there are no insects present. My ability to spot insects at that distance is very limited. There could easily be insects present that I can't see. Similarly, given our limited human capacities, we can't reasonably conclude that, given we cannot think of any God-justifying reasons for the evils we observe, there are unlikely to be such reasons. And if there are such reasons, then no evil is gratuitous.
The above response is that of a Skeptical Theist. Skeptical Theism does seem to many to successfully defuse the Evidential problem of Evil. Rather than identifying the reasons that would justify God in allowing observed evils, Skeptical Theism insists that, whether or not we can successfully identify such reasons, we're in no position to conclude that the existence of such reasons is unlikely.
However, Skeptical Theism runs into problems of its own. It entails that we similarly have no clue whether it would be worse, all things considered, if I tortured my cat to death with a red hot poker than if I didn't, because, for all I know, there's an all-things-considered good reason for me to torture my cat to death. Just because we cannot think of such a reason doesn't give us good grounds for thinking no such reason exists. Perhaps my torturing my cat to death is necessary for some very great good further down the line. We cannot say this is unlikely, says the Skeptical Theist. But surely we are in a position to say this is unlikely.
Worse, Skeptical Theism entails that we have no clue whether God has an all-things-considered good reason to deceive us about whether Jesus offers us salvation. If there is such a reason, and if God always does what is best, all things considered, then for all we know God is deceiving us about 'Jesus saves' and truth of Christianity. And also about the existence of the external world, given that for all we know God has a good reason to deceive us about that too.
In short, Skeptical Theism appears to open a skeptical Pandora's box - generating skeptical consequences that are not only implausible, but undercut any grounds for supposing Christianity is true, be those grounds empirical evidence, testimony, scripture, religious experience, or divine revelation. For a Christian, playing the Skeptical Theism card to deal with the Evidential Problem of Evil salvages the reasonableness of their Christian faith only by then undermining the reasonableness of their Christian faith in another way.
Some Skeptical Theists, having spotted that Skeptical Theism entails that for all we know God has good reason to deceive us about the truth of Christianity, insist that we can nevertheless still trust God because God, being perfectly good, would not deceive us, not even if there were a compelling reason for him to do so. One difficulty with this response is that the Bible says God sometimes deceives (2nd Thessalionians says God sends a 'powerful delusion, leading them to believe what is false.' (2:11). The God-would-never-lie response also raises the question: why, if a morally perfect God won't lie to achieve a greater good, will he inflict horrendous suffering on children to achieve a greater good?
3. Counter-balancing arguments strategy
A third strategy, in response to the Evidential Problem of Evil, is to suggest that while there appears to be a great deal of not just evil, but gratuitous evil, we can know that appearance is deceptive if we have far stronger grounds for supposing there is a good God that would never allow such evils. And of course, many theists believe they do possess such grounds. But do they? Most of the most popular arguments for the existence of God, certainly in their simplest forms, appear to provide no clue to as the moral character of our creator, first cause, intelligent designer, or Prime Mover. It's a huge, and, as it stands, unwarranted leap to go from 'there is some sort of intelligence behind the universe....' to '.... and it's perfectly good.' While there are a few arguments specifically for a good god, they are among the most contentious arguments. Take moral arguments, of which this is a typical example:
If there is no good God, then there are no objective moral facts.
Objective moral facts exist.
Therefore, God exists.
This argument assumes there cannot be objective moral facts without a good God, which is, to say the least, contentious (many leading Christian philosophers reject this claim). The argument also assumes that there are objective moral facts. While there appear to objective moral facts, appearances can be deceptive (it really seems to us that the Earth is stationary, but it isn't). If the evidence against a good God were sufficiently powerful (and it is), then - even if it could be shown that if there's no good God, then there are no objective moral facts - the right conclusion to draw would then be that there are no objective moral facts (appearances notwithstanding), not that there is a good God.
In order to more than just counter-balance the otherwise compelling argument against a good God provided by the Evidential Problem of Evil, an argument for a good God would need to be still more compelling. Yet even many theists admit that the arguments specifically for a good God fall far short of being decisive.
4. The 'I just know' strategy
Finally, theists presented with the Evidential Problem of Evil may insist that while it may indeed provide strong evidence against a good God, and while the arguments for a good God looks at best flimsy, it's nevertheless reasonable for them to believe in a good God given their own direct personal experience of such a God. Indeed, they can just know a good God exists by means of direct personal experience. Their direct personal experience trumps the evidence to the contrary.
Claims to such evidence-trumping personal experiences are common when it comes to beliefs in extraordinary hidden agents - ghosts, fairies, angels, dead ancestors, nature spirits, demons, gods, and so on. Suggest to Mary that there's overwhelming evidence that the deceased don't hang around and communicate with the living, and Mary may insist that, nevertheless, she just knows that her dead Auntie is currently in the room with her.
How might we 'just know' that God exists? Many theists insist that God has furnished us with a reliably functioning God sense or sensus divinitatis that allows at least some of us to know directly that God is real. They may also insist that it could be reasonable for someone to trust such their personal experience even if they have been presented with strong evidence there's no such being. Compare a situation in which, say, I'm presented with compelling evidence that there are no oranges currently available anywhere in the UK (evidence that a devastating blight has destroyed every orange in the country, say). If I now appear to be directly aware that I'm eating an orange, then it can be reasonable for me to believe there's an orange present, notwithstanding that mountain of evidence to the contrary. My personal direct experience trumps the evidence.
While I have some sympathy with the principle that we can reasonably believe and indeed directly know that something exists even when the available evidence strongly supports the conclusion that it doesn't, it seems to me that the orange analogy is misleading. I have no reason to distrust my orange experiences more generally. However, I possess a great deal of evidence that we humans are highly prone to false positive experiences of extraordinary hidden agents. And a good God is just another example of such extraordinary hidden agency. Given this additional knowledge about our proneness to false positive beliefs in such agency based on subjective experience, plus the strong evidence provided by the Evidential problem of Evil that there's no good God, relying on my subjective impression that there is a good God looks pretty unreasonable. That would be akin to my believing there's an orange present when I possess not only strong evidence that no oranges are present, but also strong evidence that I've been drugged with a substance that can induce compelling hallucinations of citrus fruit. Even if there is an orange present that's causing my current experience, I can't now reasonably believe that there is.
Many atheists consider the Evidential Problem of Evil fatal to traditional theism. They are right to do so. It is patently absurd to believe this world is the creation of a supremely powerful and malevolent deity, given the depth of good we observe - love, laughter, rainbows, ice-cream, and so (though, as I explain in my paper 'The Evil God Challenge' - we can similarly construct theodicies and appeal to Skeptical Theism to defend belief in such an Evil God). It's scarcely less absurd to believe this is the creation of a supremely powerful and benevolent deity, given the depth of observed evils. The real mystery, in my view, is why so many of us fail to recognise this obvious fact.
 For more on this objection see Stephen Law, 'The X-Claim Argument Against Religious Belief',
 Stephen Law, 'The Evil God Challenge'.