In a discussion with Jonathan Sacks, Richard Dawkins reads a letter he wrote to his daughter asking her to think for herself:

Next time somebody tells you something that sounds important, think to yourself: 'Is this the kind of thing that people probably know because of evidence? Or is it the kind of thing that people only believe because of tradition, authority or revelation?'

This has always struck me as bad advice, even to an adult.Research into cognitive biases has surely strengthened the case for agreeing instead with Edmund Burke:

We are afraid to put men to live and trade each on his own private stock of reason; because we suspect that this stock in each man is small, and that the individuals would be better to avail themselves of the general bank and capital of nations, and of ages. (Reflections on the Revolution in France, par 145)

This is what we do every day. From putting the kettle on in the morning to taking the medicine the doctor prescribed to turning the light off at night we unthinkingly rely upon others' knowledge.As Alfred North Whitehead said, "Civilization advances by extending the number of important operations which we can perform without thinking about them."

And many of our non-religious beliefs come not from our independent thinking but from authority. We believe in evolution and climate change not because we've gathered evidence ourselves or tested the evidence provided by others, but because it is the consensus of scientists.

So far, so good. But on the other hand, the Hillsborough report highlights why Dawkins might be right. I mean this in two ways:

- Irvine Patnick (pdf) and Kelvin Mackenzie both trusted lying policemen.

- Jack Straw now regrets that Lord Justice Stuart-Smith failed to expose the extent to which the police perverted the course of justice in covering up their murderous incompetence. He was guilty of trusting Stuart-Smith's authority too much, as Stuart-Smith trusted the police too much.

So, what is the difference between "good" reliance on authority and bad? In many cases, it's because empirical evidence justifies it. We trust that doctors will prescribe us the correct medicine because most of us have survived our encounters with the medical profession. But this consideration doesn't apply in one-shot questions.I'd suggest two criteria here:

1. Wishful thinking. If you trust an "expert" because you want to, you raise your odds of being wrong. Patnick and Mackenzie wanted reasons to hate working class Scousers so were happy to believe the lies. And Straw - who never saw a powerful arse he didn't want to kiss - wanted to believe that judges were diligent and competent.

2. Incentives. Coppers, I suspect, are no more innately corrupt than anyone else.But they had huge incentives to lie to hide their incompetence.And people respond to incentives.

I would amend Dawkins thus:

Think to yourself: 'Is this the kind of thing that I want to believe? Are people telling me this because they have an incentive to do so? If the answers are yes, don't trust the authority, and look for the evidence.