Saturday 26 April 2014

Is Belief in God Irrational? A response

Joshua Cockayne
Firstly, I’d like thank Anna, Jon and Gary for their insightful comments and for raising some interesting areas to follow up. I will try to respond to these as best as possible.

First, I will respond to the question concerning religious diversity: the worry raised was that my defence of theistic belief would allow us to defend the rationality of incompatible beliefs such as beliefs in Hinduism and Christianity. 

I should clarify that the kinds of experience I had in mind, which provide immediate justification for theistic belief would be experiences of God as a person but not necessarily of the God of Christianity or Islam, for example. The kinds of experience I am interested in are much more basic in nature, such as ‘I am having an experience of a loving God’. This issue should be kept separate from the issue of which religion is closer to the truth, but it is interesting to note that my account will only apply to religions that rely on a personal relationship to God as a person. Admittedly each religion will interpret experience of God in light of their own tradition, but my focus is specifically on theistic belief rather than religious belief in general. In this sense, the kinds of beliefs that are interesting are very basic in nature and my account does not amount to a defence of fully fledged religious belief, only basic theistic belief. My example of Christian religious experience was perhaps misleading, but I merely used this example as it is the tradition with which I am most familiar.

To reply to Anna’s questions concerning whether theistic believers have any propositional beliefs about God; I do not claim that believers only have non-propositional beliefs but rather that belief in God requires experiencing God as a person. I can believe lots of things about God, which are entirely propositional, and these are theistic beliefs. A person can have only propositional theistic beliefs; arguably, an atheist has only propositional theistic beliefs, just negative beliefs. The contrast I attempted to draw was between beliefs about God (which can be entirely propositional) and beliefs in God (which require a non-propositional element), but I do not claim that theistic believers only have non propositional beliefs. In fact, much like when Mary (in Stump’s version of the story) leaves the room for the first time, what she learns is something essentially non-propositional about her mother, but it does not then follow that Mary has no propositional knowledge of her mother. Mary would come to believe that all of the things that she had learnt about her mother whilst in the room are now true of this person that she experiences for the first time. Mary could have beliefs about her mother prior to leaving the room but she could not truly be said to know her mother. Similarly, I claim that belief in God requires experiencing God as a person and not just having certain beliefs about God. Therefore, I think believers are justified in believing in God even without publically demonstrable evidence. However, it does not follow that religious believers have no propositional beliefs.

Secondly, I turn to the worry raised by Anna and Gary as to whether ‘theistic seemings’ can provide justification for belief and whether these are defeated by the possibility of illusion or drug use. In my original post, with reference to James Pryor, I claimed that theistic believers are justified in believing P, iff P seems to be the case and there is no defeating evidence for P. Therefore, to respond to a question that Jon raised, if there are strong arguments against theistic belief which cannot be responded to then these would act as rebutting defeaters to theistic beliefs. The kind of worry raised by Anna or Gary, as I understand it, is that the possibility of illusion or that my experience might be caused by hallucinogenic drugs would function as an undercutting defeater to my belief in God. 

The most promising reply to undercutting defeaters of this kind is to consider how we might respond to the external world sceptic who asks whether our experiences of the world might all be the result of some evil demon. Michael Bergmann suggests that certain higher order seemings about the felt veridicality of perceptual experiences deflect the sceptical hypothesis that we are merely brains in vats (for more detail on this reply, see my post on the Workshop on Defeat and Religious Epistemology). The same can be extended to theistic beliefs; the possibility of illusion does not provide an undercutting defeater for belief so long as the higher order seeming about the veridicality of belief is stronger than my seeming about the possibility of illusion.

I would be interested to hear your thoughts on these issues and I apologise for anything I have overlooked.


  1. Thank you very much, Joshua. I think I now see more clearly your view on the issues I raised. If you have time, I’d like to ask you a couple of further clarifications.

    1) One of my doubts was about how a non-propositional representation of the Christin God would differ from that of the Muslim God, or of the God of other religions… If I correctly understand, you have replied that indeed the kind of non-propositional representation you are talking about does not ‘incorporate’ such differences. You are talking about a basic non-propositional representation of ‘the person of a loving God’, which constitutes the content of (various instances of) religious experiences and can justify various kinds of beliefs IN God (provided that such God is conceived as a person, and not just as a kind of metaphysical principle, or anything like that).
    I take this point; it seems right to me. However, even if the non-propositional representation you are talking about is so basic and (in a sense) ‘non-specific’, I guess that in order for it to be a representation *of the person of God*, rather than of any other kind of finite, contingent person, it should at least ‘incorporate’ some fundamental attributes such as eternity, omnipotence, the fact of being responsible for the creation everything, etc… If this is right, how can such attributes be represented non-conceptually? And if they cannot, how can the non-propositional representation you are talking about justify a belief in GOD?

    2) The second clarification actually is a clarification of one of my questions (n.3), which I'm afraid I didn’t formulate very well.
    In your original post it was very clear that you were not denying that religious people have ALSO propositional attitudes concerning God and religion, but you were just arguing that part of their attitudes – i.e. their BELIEFS IN God – are not propositional.
    So my question wasn’t whether you agree or not that religious people have also some kind of attitude towards *propositions* such as GOD EXISTS, GOD CREATED HUMANKIND IN HIS IMAGE, etc…
    My question was rather whether you think that such propositional attitudes are described at best in terms of belief, or whether you might agree with me that – at least in some cases – they are better described in terms of propositional imaginings (I explained some reasons why I think that often they are not beliefs in a previous post). Actually from what you say I think you don't agree with me - you seem to assume that the belief explanation is the right one. But if you have any thoughts about the reasons I gave to think otherwise, of course I'd be very happy to hear them.

    I hope I've explained myself clearly… Sorry if not! I’ve many other thoughts on your arguments, which hopefully there will be occasions to further discuss in the future. Thanks again and good luck with you very interesting PhD!

  2. Thanks Anna, your comments have been very helpful. I think that you are right that the non-propositional representation seem rather ‘non-specific’ and I'm not sure that I have entirely figured out how this fully supports religious belief. However, I think that we can have a non-propositional representation of a loving God in the same way that Mary (in Stump’s version) can have a representation of her mother’s love when she meets her. Mary still needs to know that the woman she meets is her mother, what a mother is and various other details which contextualise her non-propositional beliefs. I think the same can be said of our experiences of God and that our non-propositional representations of a loving God, understood contextually through religious tradition, for example, can be justified if there are no defeaters to belief.

    To respond to your original post, and your claim that religious attitudes are best regarded as a kind of imagining because they are not sensitive to evidence, I don’t think this is correct. In fact (as I claimed in my original post) I think that the justification for religious belief is defeasible as it is for perceptual beliefs and that both are receptive to evidence. I think the reason religious beliefs seem insensitive to evidence is for the same reason that we don’t take perceptual sceptics seriously; because scepticism seems more obviously false to us and we cannot seriously entertain the sceptical hypothesis. This (as I claimed in my response) can act as a higher order seeming to deflect the undercutting defeater from scepticism. However, I take it that if we could somehow prove our perceptual beliefs to be false, then we ought to give up on them, and the same applies for our religious beliefs, I think. In this regard, religious attitudes don’t look very much like propositional imaginings at all. If God does not exist, then these are just unjustified beliefs.

    Perhaps I have misunderstood your argument, but I would be interested to hear why you think this isn't the case


Comments are moderated.