Obviously irrational beliefs can be useful. Voting, for example, would break down entirely if people didn’t believe that their vote could make a difference, when in fact the chance of that happening is almost zero (especially since if an election really did come down to a single vote, there would probably have to be a re-run election.) Voting is completely irrational on an individual level but it works fine as a way of running a society.

Religion, however, is a very difficult case, and whether religion is “good for you” is an open question. There are a lot of studies finding better health outcomes in more religious people but the effects are usually pretty small and it’s unclear whether it’s belief, per se, which is beneficial, vs. being part of a supportive community. There are also differences between religions e.g. I seem to remember that Catholicism is better than Protestantism on some measures e.g. reducing suicide rates.

Overall I’m sceptical of any attempts to say that religion is beneficial or harmful overall, since the positive and negative aspects are hard to measure and even harder to weigh against each other. The fact that almost everyone who thinks religion is harmful is a hardcore atheist and most people who think it’s great are very religious themselves, makes me think the whole argument is just a political echo of the debate over whether God exists.

I am not sure that the term “instrumental rationality” is necessarily a good one for – a set of beliefs, not necessarily true, that will help you get through your day – but let us suppose we use it.

Surely the interesting thing is not so much that these “rationalities” conflict but that they lie on top of one another. We are capable of holding in our heads any number of overlapping and contradictory “truths” and, as human beings, we are expert at picking out the one that best meets our current requirements.

Nor does this make us silly irrational beings – for example, if you ask someone to describe how objects inter-react with one another, the description you will get is generally Newtonian in character, even if that person knows perfectly well that, under certain circumstances, the inter-reaction is relativistic in character or, indeed quantum. In almost all circumstances, the Newtonian answer is the simplest answer that gives a reasonable approximation of truth – but we carry the others around just in case.

Similarly, we are quite capable of carrying around an instrumental belief system (the business I have just set up will do fantastically well – the book I am writing will win awards) even though we are also capable of doing the complex calculations which tell us about our real chances of success. In the case of the business, this does not matter so much – the actions that are likely to make our business successful in a rational analysis are likely to be the same actions that will make it successful in the instrumental belief system. We can therefore choose to spend most of our time in the more comfortable surroundings of the belief system because it is, frankly, a less stressful place to be. The belief that we will be successful carries few penalties – so why not?

We don’t need to strip away all these belief systems (or delusions as Dawkins insists on calling them) we simply need to be as sophisticated about how we shuttle between truth and belief or belief rationality and instrumental rationality if you prefer.

For example, I do not believe myself to be a racist. However, a rigorous test of this proposition conducted by walking around South London’s Heygate Estate late at night might conceivably reveal that, contrary to what I think I think, I am indeed more wary of black people than white people under stressful circumstances. Should that affect the general contention that I do not consider myself to be a racist? Well, a bit, certainly but perhaps I am simply reacting rationally to the evidence presented to me by newspapers (who may or may not be racists themselves). What it should certainly prompt me to do is to ensure that, if I am in a situation, such as a job interview, where there is a possibility of the general, everyday belief that I am not a racist blinding me to an unconscious act of racism then I should make certain that I am using all my objectivity (my belief rationalism) to check the decisions I am making.

It is only by shuttling between the comforting general belief and the occasionally uncomfortable process of attempting to think objectively that the validity of the general belief can be maintained – even despite the awareness that I may harbour unconscious prejudices.

I’m a great believer in irrationality as part of the human condition. As to the specifics of the article surely it depends on the people in question. For some people to be free is to have the freedom to be irrational; some find more freedom in having their lives structured in some way by state or church.

Freedom itself is such a subjective concept that it is hard to say one way or the other. Control matters but some people simply dont want too much control and thats all well and good. In terms of framing specific policy the way forward surely is to find a way for the two things to co-exist.

“Voting is completely irrational on an individual level but it works fine as a way of running a society.”

That is certainly an open question: http://www.cato.org/pub_display.php?pub_id=8262

Most people would settle on “least worst” rather than fine. I would argue that it would be best to limit so far as possible the scope with which voting mechanisms impact on individual decisions, and that state governments tend to behave better when citizens have an enforceable right of exit of their jurisdiction besides a right to vote.

Your use of “instrumental rationality” here is highly misleading.

The religious believer doesn’t see belief as a matter of “maximizing” anything; rather, the doctrines she believes in and the practices she undertakes are part of what makes her life within the community she belongs to meaningful. Similarly, Dawkins no doubt sees free scientific enquiry as a vital ingredient of the life of a scientist. The loss of the ability to engage in it (e.g. through the imposition of a theocracy) wouldn’t be the loss of an instrumental good, but would be the loss of everything that makes the life of a scientist mean anything at all.

I think Dawkins should dally with religion a little longer and he’ll probably find accomodation with one form or another.

This article is flawed by presenting an explanation for unqualified personal bias when the qualification would have rendered it redundant before it had started.

People have… “faith in their own ability to control their lives” infers an absolute position, which could be more accurately rendered “faith in their own ability to control the direction of their lives”. Then we’d have less cause to be criticised for exaggeration.