This claim comes from the school of thought that grew out of Wittgenstein’s work, especially his theory of language games, which works on the principle that the meaning of words is in their usage, and different areas of knowledge use language in different ways – they are like different ‘games’ – so that just as you wouldn’t take the rules of the game of football and apply them to chess, you also can’t take the rules of the scientific language game and apply them to the religious language game.
This was a challenge to Logical Positivism, whose strict interpretation of meaning along empirical lines represented by the verification and falsification principles implied that there was an objective viewpoint from which you could judge the meaningfulness of sentences. If science and the empirical method represented this standpoint that meant that the rules of their game should be applied to all the others.
It is clear that the theory of language games relies on an anti-realist theory of meaning. The anti-realist holds that meanings are to be understood by reference to what circumstances you would be justified in asserting them, as Michael Dummett calls them ‘assertability-conditions’. Clearly the phrase ‘I baptise thee in the name of the Father and the Son and the Holy Spirit’ would gain its meaning from the context in which it was said – it is understandable among the community of believers participating in the ritual of baptism as a performative sentence uttered by a priest which announces the changed status of the participant to a full member of the community. This is a fairly specialised area, and the language is performing a specialised job. Language games theory’s greatest strength is that it accounts for the way in which much language is used; there is not one correct way of applying rules of meaning outside of the circumstances in which the sentences are being said – so jokes, cursing, blessings, cries of pain, analytical sentences, scientific sentences, all have their own internal coherence and meaning.
So language games theory lets each area of knowledge have meaning based on its own rules, so that someone else cannot stand outside of the religious language game and judge it by different rules – one criticism of this is that then believers can say anything that they want – any old nonsense can be passed off as religious truth. Related to this criticism is the fact that many religious statements do seem to be making assertions that exist at least partly in other areas, historical, even statements about the world like science, so surely it is an oversimplification to think that each area is self-contained with its own rules, and therefore that in some sense religious language must be at least partially open to judgement by say, historical or scientific rules .
This criticism reflects a real misgiving many have with language theory that makes it become a kind of fideism – the belief that faith is independent of reason, and therefore not open to criticism from it. D Z Phillips also notes the common criticism that if religious beliefs are isolated, self-sufficient language games, it becomes difficult to explain why people should cherish those beliefs so much: “religious beliefs begin to look like hobbies, something with which men occupy themselves at weekends”.
D Z Phillips maintained that to some extent these misgivings were justified, but he also believed they were partly based on misunderstandings about the nature of the theory. For instance, he argues that there are definite distinctions that should be made in the use of words in religious contexts from other contexts. The use of the word belief is one instance. He uses Wittgenstein’s example of a person saying “I believe in the Last Judgement”, and his friend says “I’m not so sure”, and in another case where the person says “I believe there’s a German plane overhead”, and his friend says “I’m not so sure”. Clearly the gap between the two people in the first instance is fundamental – whereas in the second instance there is really not much between them. D Z Phillips says that this highlights the really foundational way some religious language is used, such that the calling the argument between the believer and the non-believer a ‘disagreement’ or contradiction is really insufficient – the two arguing about the German plane were having a disagreement – the first two not. You can’t disagree with or contradict someone over something unless you share a common understanding of the thing. Phillips uses the examples of the man who says the sun is ninety million miles from the earth contradicting the man who says it is twenty million miles from the earth – they contradict each other because they share a common understanding. But the person who says ‘God does not exist’ does not contradict the person who says ‘God exists’ because for the believer – the question of God’s non-existence is literally meaningless – God’s definition includes his necessary existence – he is not a being among beings who might not exist.
Language games theory’s value to the debate on religious language was that it helped to bring out the ‘grammar of belief’, that it doesn’t involve the weighing of evidence, or reasoning to a conclusion, but seeing how it regulates a believer’s life. Phillips likens religious belief to a picture, which to some people is constantly in the foreground, shaping how they act, and to others (non-believers) it just doesn’t figure in their life at all, it plays no part in their thinking. Therefore, there is a real and important sense in which religious language can only be understood properly in the context of religious belief. Wittgenstein and his subsequent followers showed how important mistakes could be made if people outside of the religious context judged the beliefs on the basis of assumptions that the language the believer was using was being used in the same way as the language from another context.