The claim of a language game theorist is that the meaning of a phrase is found in its use or context. This was first proposed by Ludwig Wittgenstein in the early 20th century, whose thought grew out of the ideas of the Vienna Circle. At first Wittgenstein’s view resembled a positivist stance on language – that words and sentences were able to convey meaning because they referred to pictures of actual things and occurrences, but later he came to develop the view that meaning was not found directly in this one to one relationship, but rather that it was within the context of the words themselves and the way in which they are used. The consequences of this for religious language are that one has to be careful not to presume that a word will have the same meaning in different contexts and thus a sensitivity to the ways in which religious believers use language and what they are doing when they use it.
Language games theory has to show that religious language can be meaningful without being reductionist and without ignoring the ways in which believers actually believe. It also needs to be able to stand up to the accusations of fideism and relativism that its opponents have charged it with. I will consider in this essay the extent to which it actually does this.
A J Ayer criticised Wittgenstein’s views, saying that if his language games theory was correct, and we are required by it to withhold our judgement of religious statements because their validity can only be decided by the coherence of rules internal to the religious language game, then we are also required to do that for those people who claim to be witches and talk to fairies. The general problem here is that we do tend to believe that we can subject most talk to rational scrutiny and decide if it is valid or not. The belief that there are certain areas of life that are not amenable to this is called fideism – the belief that faith is beyond reason. It is important then to understand why Wittgenstein believed that it was a mistake to use the rules of one language game to judge another (if indeed he did actually believe this!).
Wittgenstein thought that many blunders were made in language by conflating two different uses of a word and believing that they meant the same in both contexts. He gives the example (developed later by D Z Phillips) of one person who says that they believe there is a German bomber overhead and another person who says he believes in the Last Judgement. The two statements are using the word belief but there is a chasm between the different meanings.
In other words, the factual significance of religious discourse is not the crucial thing – it has other important kinds of significance. Belief that a bomber is overhead comes down to a matter of observation, but belief in a last judgement does not solely depend on this kind of input. Religion, then does not come down to a set of hypotheses; Wittgenstein is clear – if you treat it as such you are bound to think it is mistaken. Rather, he says that belief in the Last Judgement is much more like a picture that I always think of when I act rather than anything that might play a predictive role, and that you either have that picture or you don’t, there is no sense of which person is right or wrong about it.
Many criticisms of Wittgenstein’s theories on religious language are actually criticisms of parodies of his position, often because he spoke infrequently on religion and many of his comments have only been made available posthumously as notes from his lectures. For this reason we should be wary of stating that he thought religion was a language game. He himself never states this, but rather uses examples of religious statements to make us pay attention to the subtle differences that arise in meaning in different contexts. This is perhaps best summarised in his admonition to “look and see”. Some later philosophers have taken him at his word and done this, in the process developing and extending his theory.
One key development of this is ‘functional analysis’ which has identified functions of language such as performative (eg. “I baptise thee” or “I now pronounce you husband and wife”), interrogative and imperative. J L Austin develops this in his theory of locutionary language and shows how far from a straightforward positivist interpretation of meaning much religious language is. Wittgenstein is then very important in breaking the dominance of positivism on questions of meaning. But some have claimed that his work capitulates too much ground to the positivists. In other words it retreats into the realm of non-cognitivism much like Flew argued.
For instance, some developments of his thought might sound like the sort of non-cognitivism proposed by R M Hare in his theory of a ‘blik’. If a belief functions as a ‘picture’ which is at the bottom of all someone’s actions and it is not open to questions of truth or falsity, then it seems to be no more than a fancy, as alluded to by Ayer.
R B Braithwaite has said that religious language has a moral function. He claims for instance that the narratives of Jesus’s teaching and healing should not be seen as merely assertions of fact but as expressions of intentions to live morally by the believer. He believes most people find it easier to act in a certain way if they associate those actions with certain stories. There seems to be some psychological basis for this – clearly humans are fascinated with stories – they are universal throughout cultures – and there appears to be an old tradition of using stories for moral formation. But some have claimed that Braithwaite changes religious language into a set of moral principles decorated with stories – ie. that it is reductionist. Donald Hudson says that this error is a violation of the “depth grammar” of religion.
So some of the developments of Wittgenstein’s ideas have fallen into the reductionist trap. D Z Phillips, another of Wittgenstein’s followers, was aware of these problems, especially the particular challenge that language games does not adequately describe how believers themselves take their religion to be, or at least a lot of it. For instance as Dawkins points out to Rowan Williams there are certain truths within religion which have to be taken as statements of fact, no matter how poetically they are described (like the incarnation – either it is true that God became man in Jesus or it isn’t – and believers have to taken certain of these as propositionally true before they can believe the rest). This is why Peterson et al say that language games theory fails to take seriously what ordinary religious believers take themselves to be saying.
Don Cupitt would argue that no experience comes to us pure and unmediated by language – indeed that all experience and thought is essentially linguistic – we can’t experience it if we don’t have words for it! This is an important recognition of the centrality of language to all human concerns, religion being one of them.
In summary then Wittgensteinian thinkers have done the important job of showing the ways in which religious language is significant as part of the complex web of actions, rituals, moral behaviour, thought and experience that make up religion. But in doing this I think they have too much emphasised this functional aspect of religious language and neglected its informative element. Ultimately many Wittgensteinian thinkers would have to concede questions of factual and cognitive significance to the field of science, and leave unsaid whether language can say anything about God as the ultimate reality. Even Aquinas, a distinct agnostic when it came to questions of what we can say about God was prepared to argue that with analogy we could say something positive about Him. Therefore I believe language games theory to be only moderately successful at explaining how religious language can be meaningful.