David Hume’s critique of religion reveals what seems to be a vacillation in his commitment to an argument-based paradigm of legitimate believing. On the one hand, Hume assumes such a traditional (argumentbased) model of rational justification of beliefs in order to point to the weakness of some classical arguments for religious belief (e.g., the design argument), to chastise the believer for extrapolating to a conclusion which outstrips its evidential warrant. On the other hand, Hume, ‘mitigated’ or naturalist skeptic that he is, at other times rejects an argumentbased paradigm of certainty and truth, and so sees as irrelevant the traditional or ‘regular’ model of rational justification; he places a premium on instinctive belief, as both unavoidable and (usually) more reliable than reasoning. On this view, a forceful critique of religion would have to fault it, not for failing to meet criteria of rational argument (failing to proportion belief to the evidence), but (as Hume sometimes seems to) for failing to be the right sort of instinct.

After identifying contrasting formulations of the practical postulates of reason in Kant's second critique, I analyse the context of each formulation, showing both how the postulate of the ‘possibility’ of God is consistent with Kant's understanding of a significant transition arising from practical needs as well as how the postulate of the existence of God can be seen as a ‘practical belief’ acting out a ‘hope’. My goal is to re-examine Kant's view of the relation between the practical and theoretical employments of reason in order to distinguish clearly between what Kant sees as required of the moral believer as opposed to permitted.

Søren Kierkegaard (in the Climacus writings) and John Henry Newman have starkly opposed formulations of the relation between faith and reason. In this essay I focus on a possible convergence in their respective understandings of the transition to religious belief or faith, as embodied in metaphors they use for a qualitative transition. I explore the ways in which attention to the legitimate dimension of discontinuity highlighted by the Climacan metaphor of the ‘leap’ can illuminate Newman's use of the metaphor of a ‘polygon inscribed in a circle’, as well as the ways in which Newman's metaphor can illuminate the dimension of continuity operative in the Climacan appreciation of qualitative transition.

Much has been made of the Kierkegaardian flavour of Wittgenstein's thought on religion, both with respect to its explicit allusions to Kierkegaard and its implicit appeals. Even when significant disparities between the two are noted, there remains an important core of de facto methodological agreement between them, addressing the limits of theory and the dispelling of illusion. The categories of ‘nonsense’ and ‘paradox’ are central to Wittgenstein's therapeutic enterprise, while the categories of ‘paradox’ and the ‘absurd’ are central to much of Kierkegaard's attempt (pseudonymous and non-pseudonymous) to dispel religious illusion. Writing of how the ‘urge to thrust against the limits of language’ yields ‘nonsense’, Wittgenstein explicitly appealed to Kierkegaard: ‘Kierkegaard, too, recognized this thrust and even described it in much the same way (as a thrust against paradox)’.1 I want to consider whether Kierkegaard's category of paradox of the absurd is assimilable to Wittgenstein's view of nonsense and paradox. I shall argue that a consideration of Wittgenstein's view of paradox can highlight contrasting strands in Kierkegaard's writings on religious faith, strands which take paradox more or less strictly – in particular, it can clarify several different opinions concerning the status of religious claims. My exploration will bring to the fore some implications of the attempt to make room, in the religious employment of language, for a ‘higher understanding’ of truths which we are said to be able to grasp but cannot express.

The last thirty years have seen a number of major contributions to the philosophical discussion of the possibility and character of an ‘ethics of belief’. In so far as the concern was focused on the problem of what constitutes ‘sufficient’ or ‘insufficient evidence’, the question of the ‘ethics of belief’ has turned into the current philosophical preoccupation with the question of the character of ‘rationality’ and the possibility of criteria of rationality which are either universal or at least cross-contextual. On the other hand, in so far as the concern was focused on the debated thesis that ‘believing is subject to moral appraisal (including the determination of “duties” to believe)’ the question of the ‘ethics of belief’ is with us today as the double inquiry into the question of the ‘will to believe’ and the relation of belief to action. Though the two concerns are not entirely separable, I will pay more explicit attention in this paper to the latter one in assessing some recent claims concerning the position held by John Henry Newman on these matters. After addressing some of the main points in the modern philosophical discussion of the ‘ethics of belief’, I will attempt to clarify Newman's place in the ‘ethics of belief’ discussion and argue that recent evaluations of Newman misrepresent his position in crucial ways and obscure his contributions to that discussion.