The analysis of religious assertions in terms of a language game or in terms of eschotological verification are the two most notable defences today of the factual significance of religious language. But both of these approaches, I believe, are to be found wanting, not only on philosophical grounds, but especially on the grounds of faith. Neither of these approaches reflects ordinary faith, the faith of ordinary believers. And it is in terms of such ordinary faith that we can find the key to a more adequate answer to the challenge of verifiability.