Is it ever reasonable to accept the claim that a miracle has occurred? At least for fairly traditional Christians, this question has a special point, since their faith is centred upon acceptance of a miracle, the resurrection of Jesus from death. If it is unreasonable to accept any claim to the occurrence of that miracle, then their faith is ipso facto unreasonable. J. L. Mackie has recently propounded an argument, based closely on David Hume's essay on miracles, in part 10 of the Enquiry concerning Human Understanding, which provides, he says, ‘a very powerful reply to any claim that a miracle has been performed’ (The Miracle of Theism, 26). My aim is to examine his argument, and to make some comments on his understanding of the notion of miracle and its place in religious belief.

In recent years a number of arguments have been advanced to show that there are conceptual difficulties with a variety of divine attributes. Some have claimed that there is an inherent inconsistency in the notion of omnipotence, others that omnipotence was logically incompatible with omniscience or omnibenevolence, and yet others that omniscience is irreconcilable with immutuability.

When are sentences A and B the same belief? Following Quine, observation sentences A and B are the same belief when they share the same stimulus–meaning, similar patterns of assent and dissent by subjects when the sentences are queried in the presence of the same non–linguistic stimuli. As for non–observation sentences we note a suggestion of Karl Schick: apply linguistic stimuli in the form of utterances of the language, and map the connections between sentences in the language in terms of linguistic conditioned–responses to utterances. The mapping will yield a network of relations between non–observation sentences themselves, and between the latter and observation sentences at the ‘periphery’. Thus, each sentence receives its place in the overall criss–crossing of relations in the network of the language. Out of a commitment to the ‘autonomy of meaning’, we can say that when A and B are non–observational, they are the same belief when they occupy similar places in the network of sentences in a given language, or corresponding places in corresponding networks of two languages. (Since we can identify the place of sentences in the language network, and since the present suggestion identifies the sameness of belief with location identity, it turns out that there needn't be indeterminacy of translocation.)

My text for this paper is taken from Lewis Carroll's Through the Looking-Glass. It is that well–known passage where Alice is having a frustrating conversation with the White Queen. Alice responds to some of the Queen's absurd statements by saying, ‘One can't believe impossible things.’ At which point the Queen snaps back: ‘I daresay you haven't had much practice. When I was your age, I always did it for half–an–hour a day. Why, sometimes I've believed as many as six impossible things before breakfast.’

Courage is a basic virtue to any heroic society. It is the defining virtue of the aristocratic warrior in the Iliad. It came with a set of other related virtues, all functioning in a social setting unique to that heroic era. However, as society evolved beyond the heroics of war to the civility of settled city–states, courage would be reviewed and redefined. In fact the whole virtue complex would undergo fundamental changes. Still later, when from out of the cities philosophers rose, they would, in their commitment to a higher justice or righteousness than what the city had to offer to date, submit courage to a third critique and transformation. We see this in Greece; we may also see it in China.

Much work on mystical experience has taken for granted a certain view about the relation between experience and its interpretation. This ‘traditional view’ has received perhaps its most explicit statement in Stace's Mysticism and Philosophy. It is a view which is attractive to proponents of the doctrine of unanimity, the doctrine (roughly put) that at the phenomenological level all mystical experiences are basically similar. Recently, however, in a growing body of literature, the traditional view has come under heavy fire. Its critics adopt a Kantian, indeed a hyper–Kantian, picture of experience. And they see the traditional view, accordingly, as ‘naïve’ and ‘simplistic’. In addition, hyper–Kantians typically reject the doctrine of unanimity.

‘Revelation’ has not appeared at all frequently in the titles of contributions to this journal (nor, from memory, has it at all often been their unannounced subject matter). On the other hand, neither does it seem to have been formally banished. The term is occasionally used, still, without any obvious sign of unease. Perhaps the majority of contributors have tacitly abandoned it, as incompatible with a broadly phenomenological approach to religions. It is possible to describe expressions of religion (including claims to ‘revelation’) and analyse their doctrinal statements (including assertions that they contain a given knowledge of deity); but to use the term in description or analysis might seem to concede too much. If there is a revealed religion, the others might hardly seem worth describing or dissecting.

As the title suggests, the thesis of this essay is that any view which would affirm the orthodox concept of a theistic God and yet deny the existence of a devil is inherently untenable; or, alternatively put, that ‘fundamentalism’ (if it may be so called), which affirms the existence of both God and a devil, and atheism, which denies the existence of both, are the only viable options with respect to the question of the theistic God. The support for this conclusion comes from the problem of evil, together with considerations drawn from common sense.