When compared with the classical arguments for God’s existence – Cosmological, Teleological and Ontological – Religious Experience might seem like the best basis for believing in God, because the God revealed through well-known experiences is more obviously what Pascal called “the God of Abraham, Isaac and Jacob” than the abstract “God of the Philosophers”.  In this way, Religious Experience would better support Religion or Classical Theism than the other arguments, which seem to support deism at best.  Nevertheless, on closer analysis Religious Experience does little more to support belief in the God of Religion, a personal deity, than the other arguments.  Because of this, coupled with the unique difficulties which beset Religious Experience as an argument for God’s existence, Religious Experience is not the best – or even a good – basis for belief in God.

Firstly, as William James argued in his “Varieties of Religious Experience” (1902) genuine religious experiences are ineffable and resist description in what Ramsey called “ordinary language”.  When somebody reports having “seen” the Virgin Mary or having “heard” the voice of God, the experience is not really like other sense-experiences through the eyes or ears.  Further, the object that people experience is not really personal.  As Otto argued, genuine religious experiences are of “the numinous”, “the Absolute” rather than any anthropomorphic being.  Stace concurred, arguing that genuine mystical experiences are of “an ultimate nonsensuous unity in all things, a oneness or a One to which neither the senses nor the reason can penetrate.” Happold agreed, arguing that genuine mysticism is characterized by a sense of love and union with all other beings which overcomes the anxiety we all feel at being separate and alone.  For James, Otto, Stace and Happold Religious Experiences point not towards the existence of “the God of Abraham, Isaac and Jacob” but to a “higher power” which, as James pointed out, “need not be infinite, it need not be solitary. It might conceivably even be only a larger and more godlike self…” (Varieties of Religious Experience, Postscript) In this way, Religious Experience does not really serve to support belief in the God of Religion or even Classical Theism any more than the classical arguments do and so it is not the best basis for belief in such a God.

Secondly, Religious Experience is a weaker basis for belief in a “higher power” God than the classical arguments are.  This is because:

1. Religious Experience is difficult to define.  While James, Otto and Stace recognize only solitary experiences, Richard Swinburne and Caroline Franks-Davis allow for corporate experiences as well.  While James, Otto and Stace suggest that all genuine experiences are beyond literal description, Swinburne and Franks-Davis allow for experiences which can be described using everyday language.  Because scholars differ about which experiences are possibly “authentic” and which are not, Religious Experience as a phenomenon is a less convincing basis of observation on which to build an inductive argument for God’s existence.  When compared with the Cosmological Argument, nobody questions Craig’s first premise “everything that begins to exist has a cause” even if they go on to criticize other premises or the conclusion of his argument. It follows that because of the difficulty in defining Religious Experience, is a weaker basis for an inductive argument for God’s existence than the classical arguments are
2. As Swinburne points out, the argument from Religious Experience depends on assuming the Principles of Credulity and Testimony.  While it is far to say that the other arguments from observation also depend on the Principle of Credulity, Hume’s critique of Miracles in “An Enquiry concerning Human Understanding” Section X (1758) fatally undermines the Principle of Testimony as it applies to reports of religious experiences.  It is always more likely that somebody has made a mistake than that their experience, which goes against the established laws of nature, is genuine.  While it is, as Hick pointed out, bad science to ignore such reports, it is, as James pointed out, impossible to exclude the possibility that the person reporting a religious experience is mistaken because credible physiological and/or psychological explanations exist to account for everything reported.  For examples, as James discusses in his chapter on Conversion Experiences, psychiatry might account for very many of these including those of St Paul (as a response to a moral crisis according to Leuba) and of St Augustine (as a delayed adolescent crisis according to Starbuck).  While St Paul or St Augustine might themselves be convinced that their experience was authentic and be justified in believing in their object, there is no necessity for other people to believe either in the authenticity of their experiences or in what they seem to refer to.  So, because testimony always lacks credibility, an inductive argument based on Religious Experience will be weaker than the classical arguments in establishing a basis for belief in God.
3. Finally, as Swinburne points out, accepting accounts of Religious Experience as possibly genuine depends on prior probability.  If you are what James called a “medical materialist“, reports of religious experience would be their very nature incredible.  In this way, reports of will only be entertained as the basis for belief in anything by those who are already open to the existence of that thing and the argument from Religious Experience is shown to be circular and so not as persuasive as the other classical arguments.

Clearly, Swinburne would disagree and would argue that Religious Experience is a better basis for believing in God than the other inductive arguments.  In his “The Existence of God” (1979) he set out a cumulative argument for God’s existence which employed Bayes’ Theorem to assess the relative probabilities of God and natural causes as explanations of causation, order and purpose, beauty and morality in the universe.  None of these arguments is conclusive in itself, argued Swinburne.  It takes Religious Experience to tip the balance in favour of God’s existence and provide a basis for believing in God.  Nevertheless, Swinburne’s cumulative argument has been widely criticized, not least by Anthony Flew, who compared it with “ten leaky buckets”.  A lot of bad arguments, each of which fails to justify belief in God in itself, are together not significantly better than one bad argument and so fail to justify religious belief.  Further, Swinburne’s contention that Religious Experience is the strongest argument, the argument which tips the balance of probability in favour of God’s existence, seems odd given that it depends on prior probability and so has no force without the other arguments having already established that God is slightly more probable than the natural alternative.  To what extent can an argument which is circular in itself be the deciding factor?  If the other arguments succeed in justifying an openness to the existence of God without depending on Prior Probability or the Principle of Testimony, then surely they are better bases for belief in God than Religious Experience.  For Swinburne, Religious Experience is a better basis for belief in God than the other arguments from observation even though it depends on Prior Probability and the Principle of Testimony because, as he sees it, Religious Experiences support belief in the personal God of Religion.  However, as has already been established here, this is not necessarily the case.  On close analysis, for those who have had no experience themselves, Religious Experiences can support openness to the existence of a “higher power” at most.  By contrast, the cosmological argument supports belief in a necessarily single, all-powerful creator God and the teleological argument supports belief in an omni-benevolent intelligent designer God, at least when “good” is understood in the purely Aristotelian sense.  In this way, Religious Experience is a worse basis for belief in God than either of the cosmological or teleological arguments, even when it comes to believing in “the God of the Philosophers” with the classical attributes.

In conclusion, Religious Experience is very far from being the best basis for belief in God. On close analysis, reports of such experiences fail to justify anything more than an openness to a “higher power” which would not have to have any of the classical attributes of God.  It is as impossible to exclude naturalistic explanations for Religious Experiences as it is to exclude the possibility that they have been caused by God, so the question of what they point towards must remain open.  Further, as an argument for God’s existence, Religious Experience is beset by problems of definition, credibility of testimony and circularity.  It is certainly not the decisive factor in demonstrating the existence of God that Swinburne claims, but is a bucket that is more leaky than most.