Jaynes notes that many people perceive their own thought processes as a kind of dialogue between the 'self and another internal protagonist inside the head. Nowadays we understand that both 'voices' are our own — or if we don't we are treated as mentally ill. This happened, briefly, to Evelyn Waugh. Never one to mince words, Waugh remarked to a friend: 'I haven't seen you for a long time, but then I've seen so few people because — did you know? — I went mad.' After his recovery, Waugh wrote a novel, The Ordeal of Gilbert Pinfold, which described his hallucinatory period, and the voices that he heard.

Jaynes's suggestion is that some time before 1000 BC people in general were unaware that the second voice — the Gilbert Pinfold voice — came from within themselves. They thought the Pinfold voice was a god: Apollo, say, or Astarte or Yahweh or, more probably, a minor household god, offering them advice or orders. Jaynes even located the voices of the gods in the opposite hemisphere of the brain from the one that controls audible speech. The 'breakdown of the bicameral' mind was, for Jaynes, a historical transition. It was the moment in history when it dawned on people that the external voices that they seemed to be hearing were really internal. Jaynes even goes so far as to define this historical transition as the dawning of human consciousness.

There is an ancient Egyptian inscription about the creator god Ptah, which describes the various other gods as variations of Ptah's 'voice' or 'tongue'. Modern translations reject the literal 'voice' and interpret the other gods as 'objectified conceptions of [Ptah's] mind'. Jaynes dismisses such educated readings, preferring to take the literal meaning seriously. The gods were hallucinated voices, speaking inside people's heads. Jaynes further suggests that such gods evolved from memories of dead kings, who still, in a manner of speaking, retained control over their subjects via imagined voices in their heads. Whether or not you find his thesis plausible, Jaynes's book is intriguing enough to earn its mention in a book on religion.

Now, to the possibility I raised of borrowing from Jaynes to construct a theory that gods and binkers are developmentally related, but the opposite way around from the paedomorphosis theory. It amounts to the suggestion that the breakdown of the bicameral mind didn't happen suddenly in history, but was a progressive pulling back into childhood of the moment when hallucinated voices and apparitions were rumbled as not real. In a kind of reversal of the paedomorphosis hypothesis, the hallucinated gods disappeared from adult minds first, then were pulled back earlier and earlier into childhood, until today they survive only in the Binker or little purple man phenomenon. The problem with this version of the theory is that it doesn't explain the persistence of gods into adulthood today.

It might be better not to treat gods as ancestral to binkers, or vice versa, but rather to see both as by-products of the same psychological predisposition. Gods and binkers have in common the power to comfort, and provide a vivid sounding board for trying out ideas. We have not moved far from Chapter 5's psychological by-product theory of the evolution of religion.

It is time to face up to the important role that God plays in consoling us; and the humanitarian challenge, if he does not exist, to put something in his place. Many people who concede that God probably doesn't exist, and that he is not necessary for morality, still come back with what they often regard as a trump card: the alleged psychological or emotional need for a god. If you take religion away, people truculently ask, what are you going to put in its place? What have you to offer the dying patients, the weeping bereaved, the lonely Eleanor Rigbys for whom God is their only friend?

The first thing to say in response to this is something that should need no saying. Religion's power to console doesn't make it true. Even if we make a huge concession; even if it were conclusively demonstrated that belief in God's existence is completely essential to human psychological and emotional well-being; even if all atheists were despairing neurotics driven to suicide by relentless cosmic angst — none of this would contribute the tiniest jot or tittle of evidence that religious belief is true. It might be evidence in favour of the desirability of convincing yourself that God exists, even if he doesn't. As I've already mentioned, Dennett, in Breaking the Spell, makes the distinction between belief in God and belief in belief: the belief that it is desirable to believe, even if the belief itself is false: 'Lord, I believe; help thou mine unbelief (Mark 9: 24). The faithful are encouraged to profess belief, whether they are convinced by it or not. Maybe if you repeat something often enough, you will succeed in convincing yourself of its truth. I think we all know people who enjoy the idea of religious faith, and resent attacks on it, while reluctantly admitting that they don't have it themselves.

Since reading of Dennett's distinction, I have found occasion to use it again and again. It is scarcely an exaggeration to say that the majority of atheists I know disguise their atheism behind a pious facade. They do not believe in anything supernatural themselves, but retain a vague soft spot for irrational belief. They believe in belief. It is amazing how many people seemingly cannot tell the difference between 'X is true' and 'It is desirable that people should believe that X is true'. Or maybe they don't really fall for this logical error, but simply rate truth as unimportant compared with human feelings. I don't want to decry human feelings. But let's be clear, in any particular conversation, what we are talking about: feelings, or truth. Both may be important, but they are not the same thing.

In any case, my hypothetical concession was extravagant and wrong. I know of no evidence that atheists have any general tendency towards unhappy, angst-ridden despond. Some atheists are happy. Others are miserable. Similarly, some Christians, Jews, Muslims, Hindus and Buddhists are miserable, while others are happy. There may be statistical evidence bearing on the relationship between happiness and belief (or unbelief), but I doubt if it is a strong effect, one way or the other. I find it more interesting to ask whether there is any good reason to feel depressed if we live without God. I shall end this book by arguing, on the contrary, that it is an understatement to say that one can lead a happy and fulfilled life without supernatural religion. First, though, I must examine the claims of religion to offer consolation.

Consolation, according to the Shorter Oxford Dictionary, is the alleviation of sorrow or mental distress. I shall divide consolation into two types.