Is belief in god rational?

The question of whether belief in god is rational or not seems presume an answer to a prior, and perhaps more important question – namely: do we want belief in god to be rational, as opposed to being fruitful, joyous, beautiful, etc.? To put it another way, it’s long been of interest to me why this contest is often fought in the domain of rationality, where everyone who is not a supernaturalist of some sort agrees that there is no possibility of providing any sort of knock-down argument for belief in god, at least where arguments are understood to follow standard rules of logic, involving non-contradiction, the possibility of refutation, and where conclusions are adopted once they are shown to be the best justified of available alternatives.

Rather, the more compelling arguments in favour of belief in god point to various benefits of believing in god, whether these benefits are social, psychological or moral. While it’s far from clear that any of these other purported benefits hold up to scrutiny, or can’t be purchased at lower cost from other sources, it seems to me that we’d need to adopt a definition of “rational” that is essentially teleological (goal-based), rather than one that aims at truth, for it to be possible for belief in god to be described as rational.

To do so would however make this contribution to the debate – and the debate itself – worthless. If “rational” can mean all things to all people (seeing as what counts as a worthwhile goal is an essentially subjective matter), then the debate takes place on relativist terms, where whatever you say has as much validity as whatever I say, and where no independent arbiter of merit can gain a foothold.

Let us then start by adopting a minimal definition of rationality, so as to not stack the deck in favour of any particular conclusion. For a belief to be rational, I suggest that it has to be responsive to evidence, that it entails holding beliefs that are more likely to be true than false, and that beliefs held are then more likely to generate further beliefs that are likely to be true (rather than leading one further from the truth). To put this more simply, it is irrational to believe in a proposition if there is no evidence for it, or if there is a better explanation for whatever it is we are seeking to understand.
With that as my starting point, it should also be made clear that I will not be addressing some standard arguments regarding the existence of god, namely the ontological and cosmological arguments, and the argument from design. The fact that very few Christian apologists even bother attempting to ask us to take these arguments seriously should, I hope, excuse my omission of these demonstrably poor attempts at justifying the existence of god on logical grounds.

The same applies to the infamous wager offered by Pascal, where he asks us to perform a cost-benefit analysis for belief in god, and argues that seeing as little cost is incurred through belief, but that we stand to miss out on significant benefits by not believing, we are better off believing than not doing so. As countless critics have responded, this wager does not help – whether or not you accept Pascal’s matrix of payoffs – seeing as it doesn’t tell us which of the actual (or possible) versions of god to believe in.

More generally, then, and following the definition of “rational” offered earlier, we should perhaps begin with allowing ourselves to recognise that we sometimes hold beliefs that we hope are true, rather than believe to be true. On this model, “belief in belief” is what provides us with comfort or meaning, rather than the fact that our beliefs are, in fact, true. For example, the lover who suspects that her partner is being unfaithful will persist in believing in his fidelity, often long after the point at which outsiders reach a different conclusion. This sort of behaviour is well known to psychologists as evidencing “confirmation bias” – where we weigh confirming evidence for our beliefs as being far more significant than disconfirming evidence.
How do we usually relate to evidence – or rather, how do we usually intend, or hope, to relate to evidence? To answer that question, let us consider our relationship to propositions in general – in other words, to statements that can be true or false – to better understand see whether beliefs such as belief in god conform to our typical standards of justification.

We are constantly forced to make choices. Fortunately, many of those choices have no serious consequences – most of our activities affect ourselves far more than they affect others, and rarely affect us in unacceptably negative ways. However, many of our decisions do affect other people, sometimes quite seriously. And when decisions affect other people, they start to impact on us too: people’s perceptions and responses to us will constantly be shaped by the impact – negative or positive – that we have on their lives. It is thus crucial to be informed when making decisions – to have as many of the facts at your disposal, and then to treat those facts responsibly, in other words to not conveniently ignore those facts that you are uncomfortable with, or would prefer not to believe.

But what is it to be informed? One way of answering this question is to say that the state of being informed in a particular context is to achieve a sort of synchronisation between the contents of your head and the world outside, in other words to believe those things that are true and not to believe those things that are false. Unfortunately, it appears that many of us get this order wrong, in that we tend to treat the world as if it should correspond to our beliefs, rather than treat our beliefs as if they need to correspond to the world.
What this means is that we are reluctant to change our minds in light of evidence that our beliefs are false – we often first prefer to think that the information we are getting is false, rather than thinking that something we believe may be false, and that is why the world doesn’t act the way we think it should.

Why would we be so complacent in choosing what to believe? Often the answer seems to relate to the fact that it doesn’t much matter what you believe in practical terms – other people are usually willing to still talk to you or do business with you, whether or not you believe something different from them. Another part of the answer may be that many of our beliefs are never exposed or tested. We are not given an opportunity to engage with or question them because they simply never come up, or if they do, they come to the surface in contexts where it’s easy to not bother challenging them. So perhaps it’s easy to understand why we often simply carry on believing the things that we currently believe. But it is an entirely different question to ask whether we should be so lazy or complacent in choosing what to believe – and in allowing others to believe what they wish.

One way of approaching this more difficult question is to consider how we feel about different sorts of beliefs. Take, for example, an engineer. It is surely the case that we care about many of her beliefs. If she believes that her design for a particular bridge is sound – in other words that it can support your car as you travel 20 metres above another highway – we have some quite serious reasons for being concerned that her beliefs do in fact map on to the world and its physical laws accurately. For some reason, though, we don’t seem to care so much about other beliefs, particularly beliefs relating to morality, politics, aesthetics and metaphysics – in particular, religious beliefs. We do care in one sort of way, in that we may choose to interact with people who happen to share the same sorts of beliefs that we do (for example, religious people may choose to associate with others who believe in god), but for the most part, these disputes don’t seem to affect many of the day-to-day events of our lives.

So what is the difference between these different sorts of beliefs? Some may think that the answer was contained in the observation that metaphysical beliefs (for example) simply don’t affect our lives in the way that the construction of a bridge does. But is this true? Some people want to claim that the difference between these two sorts of beliefs is more that we tend to treat the engineer’s beliefs as facts and your beliefs about metaphysics as opinions. But then we can ask, surely some opinions are actually true in the same sort of way as the engineer’s beliefs are, and other opinions are simply false – and if this is so, should we not care about maximising the chances that our beliefs are true?

The problem here is that not only are beliefs interconnected, but also that a mistake in the foundation of your belief-network can corrupt any beliefs that depend on that initial belief. If this is true, it means that we should care about whether all of our beliefs are true or not, and not only care about those beliefs that relate to empirical reality. And by “true”, we can’t mean something like “true for me, but maybe not true for you”, because that’s not what the word “true” means. When we say something is true for me but maybe not for you, we are not actually talking about “truth” as commonly understood, but rather saying that I believe it, and perhaps you don’t.

On the subject of how beliefs can affect our lives, it’s true that none of this would matter if our beliefs made no difference to the ways in which we interact with the world. But history offers us countless examples of ways in which beliefs – particularly about things with little or no evidence – can have serious and sometimes catastrophic effects on the world. For example, consider the belief, held by many even today, that one particular race or gender is superior to another. Or consider the belief that there is an afterlife where you can get rewarded or punished for your actions in this life – didn’t that belief have at least something to do with the attack on the World Trade Centre on September 11, 2001? In other words, without the motivation offered by eternal reward for a mortal sacrifice, would people still willing to sacrifice thousands of lives, including their own, to make a political statement?

What this adds up to is that we offer exemptions to some beliefs from the usual standards – you can believe whatever you like, so long as you don’t perform actions that harm other people on the basis of those beliefs. But the broader point being made here is that if we are willing to offer some sorts of beliefs exemptions from corresponding to evidence, we should extend the same charity to all beliefs that can’t be proven to correspond to the world. The uncomfortable consequence of this is that if we are going to hold this attitude with regard to belief, your demand to be entitled to believe in god, or a soul, means that the racist is equally entitled to believe he is superior to you, or for the sexist to believe that she is your superior. A very simple principle – consistency – demands that we treat things that are similar in the same sorts of ways, and you need to either show that these examples are not similar, or if that fails, to accept that they need to be treated in similar sorts of ways.

The similarities between these metaphysical beliefs and those of the racist or sexist are firstly that they are all incapable of justification by the standards described earlier. In fact, in both cases the idea of objective justification is often actively rejected, and replaced by appeals to a privileged kind of knowing, often called faith. They are also similar in that religious beliefs can – at the thick end of the wedge – affect other people as negatively as racist or sexist beliefs. And if we want to avoid respecting all beliefs that don’t correspond to evidence, we are then required to instead be sceptical about all of these sorts of beliefs.

Being sceptical of such beliefs does not necessarily mean believing that they are false. It simply means not being dogmatic about them being true, and being open to the possibility that they are false. Of all such beliefs, it may well be that some or many of them are in fact true, but the problem remains that we cannot demonstrate them to be so. This means that if you are going to believe them, you commit yourself to a certain amount of risk, especially if you are going to perform actions motivated by these beliefs. Again, though, it needs to be made clear that if one is prepared to believe things on the basis of little or no evidence, one increases one’s chances of believing things that are false – and that this is an irrational way of relating to the world.
Back to our engineer: what makes her belief knowledge is that it is demonstrably so likely to be true that we may as well call it true. What stops my belief that the death penalty is unjust from being knowledge is that I cannot demonstrate it to be true. I could still however make the case that it’s more likely to be true than the alternatives are. So while it remains in the realm of opinion, rather than fact, it’s nevertheless the sort of opinion that is worth taking more seriously than the opinion that I have a disembodied and immortal mini-me, that some call a soul.

The reason for this is that my opinion on capital punishment is at least somewhat related to evidence – I can argue for it, and offer you reasons why it’s a good belief to hold. By contrast, my primary reason for wanting to believe in a soul is that it’s part of an often repeated, and quite ancient, narrative. So a large part of our answer as to which opinions to take seriously, or which opinions to treat as knowledge, has to do with how responsive the opinions in question are to evidence, in other words how well justified they are. Crucially, we must remember that we often act in ways motivated by our beliefs, and that we should therefore care about holding true beliefs rather than false ones.

Of course nobody can claim that they are sure that these metaphysical beliefs are false. Nobody can claim that they are sure it is false that it actually is unlucky to walk under a ladder. Perhaps there is some law of nature which dictates this unlikely principle, and we should promptly begin investigating this in the laboratory. But the important thing to note is that if there is no evidence for a belief (and, frequently, existing evidence contrary to that belief), it is irrational to believe it – even though it might actually be true. Likewise, we can make mistakes of the other sort – we can believe something because it seems to be justified by the evidence, but then we eventually learn that the belief in question was false. As example, consider the belief that the earth is flat – before we had boats that could travel long distances, telescopes or photographs from space, all the evidence we had access to meant that it was fully justified to believe that the earth was flat, even though that belief turns out to be false.

Our beliefs can only be justified to the extent of the information available to us. But if we happily succumb to confirmation bias, and refuse to consider the evidence we’re not bothering to look at, then justification is clearly not present, and our belief is therefore not rational. If we are to claim that something is true, it is our responsibility to access enough information related to that belief to entitle us to claim that the proposition in question is true. Once we have this evidence, we would also be entitled to being surprised – even disappointed – if others do not share that belief. But we are not entitled to ignore that homework, and just insist on respect for our beliefs, because then others are equally entitled to hold beliefs that you regard as false. The same standard needs to apply to us all.

Of course, it may well be that for some issues (such as the death penalty, or the existence of God) the evidence is too difficult to find or understand. But that does not necessarily mean that knowledge is impossible for those issues – it just means that we have to work far harder to justify it, not simply rely on what people considered justified in historical times, when education was somewhat of a novelty, and where the scope of education was severely limited. We should also not forget that just because we as individuals cannot justify something as knowledge right now does not mean that it will not eventually be regarded as knowledge. In the past we did not have the technology to know the earth was round, and now we do. We also could never prove that the races and genders were identical in all features relevant to being treated equally, and now we do. Examples such as this should remind us that we are learning things all the time, and that we should remember that there is still much to learn.

We do, however, currently know far more than our ancestors did. A belief in a supernatural entity watching over us would certainly have been rational at a time when we had no knowledge of meteorology, and where your neighbouring village’s crops were destroyed by lighting and yours not, because your village had said the correct prayers, or left the appropriate offerings. The situation is very different today – as example, consider neuroscience, which is slowly filling in all the gaps that used to need the explanatory power of a soul. It is no longer rational to believe that humans are anything but exclusively material entities, because all the evidence we have points to that conclusion, while there is no evidence (and in fact, little possibility of evidence), for the belief that something immaterial and immortal hovers over or inhabits our beings.

Of course it is possible that god exists. But if we assume that god does exist, what are we to do about it? Note that we have already taken serious liberties here, in that we have no evidence for the existence of god, but rather only some large mysteries that could be explained by god, but that also could be explained by science. If we were to take the gamble and select the god hypothesis, where does that leave us? All religious systems – including the Flying Spaghetti Monster – offer equally defensible accounts of god’s properties, at least according to the standards of rational enquiry. The fact that some of us are Christians, and others Hindu, is solely due to an accident of geography – where you happen to have been born – rather than any evidence that there is one god as opposed to many. This should remind us of the point made by Wittgenstein in saying: “A nothing will serve as well as a something about which nothing could be said”. We can, if we choose, believe in god, but that would still offer us no justification for any beliefs about what god wants, and how we should live our lives. Inserting god into the equation, in other words, explains everything while simultaneously explaining nothing.
This sort of exemption from ordinary standards of rationality is particularly interesting in that it seems to rest on an assumption that religious beliefs can be walled off from other beliefs – in other words that we can easily switch between critical and more lax standards of judgement, and that the tolerance with regard to metaphysics won’t make us generally lazy, gullible, or foolish. That risk doesn’t seem worth taking.

What we think – the contents of our heads – is the primary factor that causes us to do what we do. In other words, your beliefs determine your actions. Our beliefs with regard to political or religious questions can have serious, and sometimes catastrophic, implications in terms of what we physically do. We certainly care about how people act, and if you care about how people act (and would sometimes like them to act differently), you need to first be concerned about how they think, and to get them to think differently. This incurs on us a moral responsibility to guard against the intrusion of errors in own thinking, and also a responsibility to be aware of how allowing ourselves exemptions from justification in terms of our beliefs licences and legitimises any beliefs that do not hold themselves responsible to evidence.

To conclude: It is not obvious that religion is a matter of truth anyway, or that religious states of mind should be addressed in terms of truth and falsity – maybe religion is not a matter of particular beliefs at all, but rather consists of certain dispositions or attitudes. On this model, religious practice could be an aesthetic/emotional response to the world, like that offered by poetry; or it may be a question of immersion in a set of practises, for social purposes that provide some meaning to life, such as the unifying effect generated by being a supporter of a football team. In these cases, one would just assess these states of mind to see whether they are admirable/useful, etc., or whether the benefits could be attained, as mentioned earlier, in ways that come at less potential cost.

But once we ask the question “is belief in god rational?”, we are then not understanding religion in this way, but rather dealing with religious beliefs as claims to truth, which may or may not come with emotional and social benefits. As I have suggested, most of the explanatory role previously served by god has been supplanted by better justified scientific explanations, and we have every reason to expect that humans will find fewer and fewer explanatory gaps that have room for god – and that where we do, plugging god into that gap will serve as much of an explanatory role as an infinite number of alternatives would, seeing as we can know nothing about the nature of such a god – unless we are willing to embrace the circular argument that “god is good because she says she is”, where “good” could be taken to mean whatever characteristic was under discussion at the time.

By Jacques Rousseau

Jacques Rousseau teaches critical thinking and ethics at the University of Cape Town, South Africa, and is the founder and director of the Free Society Institute, a non-profit organisation promoting secular humanism and scientific reasoning.