In Breaking the Spell: Religion as a Natural Phenomenon, Daniel Dennett tries to answer the question, ‘Why are we religious?’ Looking at the issue from the standpoint of evolutionary biology, Dennett hopes to explain the origin and endurance of religion naturalistically.
As I’ve explained in my article on the supposed evolution of belief, I’m not completely opposed to this kind of project, nor am I particularly concerned about the results. There is no reason for Christians to worry about theories which naturalistically explain the tendency for people to be religious. There are several reasons for this, not least of which is the fact that providing explanations of the origin of beliefs are not grounds for dismissing the belief. To make such a dismissal would be to commit the genetic fallacy. The rationality of any given belief must be analyzed on its own merits.
And this is what Dennett fails miserably to do, which would be fine except for the fact that he promises to do it. If Dennett were merely describing the ways religion could naturalistically arrive, then there would be no problem. But the book is not quite neutral on the subject, and it is clear that Dennett thinks that there are actually good reasons to reject religious beliefs. In one short section, Dennett takes on the task of refuting natural theology. Yet, even by his own admission, the section contains little or no reasoning- it only gives broad pronouncements. To introduce the section, Dennett writes,
“At long last I turn to the promised consideration of arguments for the existence of God. And, having reviewed the obstacles- diplomatic, logical, psychological, and tactical- facing anybody who wants to do this constructively, I will give just a brief bird’s-eye view of the domain of inquiry, expressing my own verdicts but not the reasoning that has gone into them, and providing references to a few pieces that may not be familiar to many.” 
Did you catch that? Dennett admits that he is not going to provide any reasoning, but merely pronounce his own verdicts on the matter! Why on earth should he do that? It’s already clear at this point in the book that Dennett is an atheist, so we are already aware of his ‘verdicts.’ What we need is some solid reasoning.
And of course, what he does offer us on this subject is extremely weak. He brushes off historical arguments, claiming that they are manifestly question-begging. Well, some of the most unsophisticated ones are, but this does not show that all historical argumentation is useless. Dennett apparently thinks that the only historical arguments are the ones of the fashion:
1. The Bible is God’s word.
2. God makes no mistakes.
3. Therefore, the Bible makes no mistakes.
And, therefore (of course), we know that God exists, because the Bible tells us so! Of course these sorts of arguments are absolute trash, but what about arguments for the Resurrection of Jesus Christ, which do not assume that the Bible is perfect but merely considers the books of the New Testament (and others) as historical sources. Dennett probably doesn’t think that a good case for the resurrection can be made, but the point is simply that he cannot brush off all forms of historical argumentation as if there is nothing substantial to discuss.
Philosophy is actually Dennett’s primary expertise, so we should expect him to be particularly qualified to discuss philosophical arguments for God’s existence. Here, again, he falls far short. He brushes off the Cosmological Argument with simplistic objections like ‘What caused God?’ Then he basically dismisses the whole argument as too ‘arcane’ to bother with. He doesn’t even do this much when discussing the Design Argument, which he claims to have covered ‘quite extensively’ in his book Darwin’s Dangerous Idea, and he decides to merely provide a summary from that book rather than offer any reason for us to think that the Design Argument is flawed.
And that’s all there is to say about that. Whatever the use of this book, it certainly provides no compelling reason to think that God doesn’t exist.
Besides developing an account of the naturalistic emergence and evolution of religious belief, Breaking the Spell also attempts to address the question, ‘Is religion good for us?’ Dennett admits that much more research needs to be done before conclusions can be reached, an admission that he makes concerning almost every topic and theory discussed throughout the entire book. Yet, despite the fact that the data is too limited and the variables too many to come to a conclusion on the matter, Dennett still seems, at least implicitly, to conclude that religion is bad for us, or at least is seriously problematic.
Dennett does get quite a bit right. He is right that we should not try to shield religious beliefs and protect them from skeptical or reflective inquiry. He is right that we should pursue a great deal of research into these topics. I am personally not afraid of such scrutiny of my religious beliefs. But Dennett has a long way to go if he hopes to convince me that religion is only a natural phenomenon.