This article develops a new argument – The Argument from Atonement – that commonplace Christian assumptions about heaven imply the existence of purgatory. According to this argument, many Christians will die with residual moral guilt due to possessing unfulfilled moral debts towards fellow non-divine humans. Some of these debts cannot be fulfilled immediately after death and are not completely fulfilled or cancelled by Jesus's atoning work. Christians with such debts are fit neither for heaven nor for hell and thus must occupy purgatory until their moral debts are fulfilled.

Responding to key questions raised by the other three, this article discusses the factors which led to the development of Christian fideism and why Christians were seen as a threat to wider society. It considers whether early Christian discourses always represent (of characters in narratives), or demand, belief alongside trust and other relational aspects of pistis, and argues that it is sometimes possible to have effective pistis without having right beliefs. It discusses the variable relationship between belief and doubt in New Testament texts, and suggests how the faith of St Teresa of Calcutta might have been viewed by early Christians.

Many philosophers ardently subscribe to what can be called the doctrine of public-world fallibilism (DPWF), i.e. the doctrine that human persons can never have infallible awareness of the truth of propositions such as that expressed by the sentence ‘There is an olive on the kitchen floor’. It has, of course, been standard to contrast such claims with epistemically tentative first-person phenomenological reports, e.g. ‘It seems to me that there is olive on the kitchen floor’. According to the DPWF, for any public-world proposition p and person N who accepts that p, there is always some epistemic possibility (it may, of course, be quite small) that N is mistaken about p. I hope to establish – in something of a Cartesian spirit – that there is eminently respectable warrant for maintaining that theism requires the falsity of the DPWF. Further, I believe that this thesis can be seen to have far more epistemological significance than might initially be supposed.

In recent essays John Bishop proposes a ‘doxastic venture’ model of religious faith. This author notices that a so-called doxastic venture model of theistic faith is self-defeating for the following reason: a venture suggests a process with an outcome; by definition a venture into Christian faith denies itself an outcome in virtue of the transcendent character of its claims – for what is claimed cannot be settled. Taking instruction from logical positivism, I stress the nonsensical character of religious claims while attacking Bishop's model. However, I wish to avail myself of this same model to describe a state of belief among certain parties which does not refer to transcendent matters, in order to show that a doxastic venture is indeed a valid description of a state of belief, and that pursuing this model shows in relief the transformative nature of belief, along with its essentially scientific status. It is my ambition to show, turning Bishop's model against itself, that a state of religious belief suffers from a precise logical equivalence to a condition of agnosticism. I ask whether we are justified in believing in belief.

Paul Helm's discussion of my book is a clear illustration of some of my central claims about Anglo-American philosophy of religion: he instantiates its undue preoccupation with beliefs, and its erasure of gender. In my reply I show how Helm conflates my objection to such preoccupation with the absurd claim that beliefs are unnecessary, and how he conflates philosophy of religion – even rationality itself – with its Anglo-American variants. He refuses to engage with the masculinism implicit both in his selectivity and in the boundaries he seeks to maintain around the discipline; boundaries which, I argue, need to be opened up for philosophy of religion to flourish.

Before beginning my response, let me express the honour I feel in having these three friends and distinguished philosophical colleagues comment so thoughtfully on my ideas in Divine Discourse. I warmly thank them for their ‘labours’.

I propose mirroring the general structure of the book itself in my response. First, I'll consider what Helm says about my delineation of the topic, second, what Quinn says about my discussion of God speaking; third, what Westphal says about my discussion of interpreting for God's speech; and last, what Quinn says about my discussion of the epistemology of believing that God speaks.

‘Faith’ (pistis) is a key term in early Christianity. It not only describes the self-understanding of an individual Christ-follower, but also operates as the social marker of the Christ-groups. Rather than adding to the renaissance of studies on faith in recent years, this article seeks to illuminate what faith is by focusing on the phenomenon of doubt, broadly understood. After some linguistic reflections, the article identifies six basic types of doubt in early Christian writings and then compiles eighteen coping strategies and patterns of resilience reflected in the most prominent texts. The article is not an attempt at systematizing early Christian reflections on doubt but is rather a cornucopia of insights into how the first Christian theologians talked about doubt, dealt with it, and tried to overcome it.

I discuss the role of religious experience in Richard Swinburne's probabilistic case for theism. Swinburne draws on his principle of credulity to argue that, if in addition to other evidence we consider that many people have theistic religious experiences, theism comes out as more probable than not. However, on many plausible probability assignments for the relevant non-experiential evidence, the conditional probability of theism already converges towards 1. Moreover, an argument analogous to a general Bayesian argument against phenomenal conservatism suggests that, after we take account of evidence from religious experience, the probability of theism cannot be greater than the prior probability that the best rival hypothesis is false. I conclude that these observations are compatible with what Swinburne would call ‘weak rational belief’ in theism and that such weak belief can be strong enough for rational faith.

Eleonore Stump develops and defends a theodicy of redemptive suffering. In particular, God's permission of suffering (at least some classes, if not instances, of serious undeserved, involuntary suffering due to natural or free causes) is justified just in case it benefits those who suffer, it is the best possible means in the circumstances for their benefit, and God knows this is the case. The main aim of this paper is to show that for Stump's theodicy to have a good chance of working, it is reasonable to think that it requires the Molinist claim that God has middle knowledge.

In conversation with Morgan (2015), I point out that the view of relational faith I have elsewhere defended (McKaughan 2013, 2016, 2017) fits rather well with the understanding of pistis that emerges from Morgan's careful reading of New Testament texts. Moreover, the fact that New Testament authors display little interest in examining interior aspects of faith makes it difficult to justify the claim that their understanding of the pistis lexicon requires believing in the modern sense as the attitude Christians must take towards relevant content, in contrast to various other positive but non-doxastic attitudes that philosophers recognize today. Such faith is of contemporary interest, given its congruity with early Christian tradition, the role it can play in helping relationships to persevere through various kinds of challenges (including doubts significant enough to preclude believing), and for the wide range of evidential circumstances in which it can be enacted with intellectual integrity.

I argue that Social Trinitarians can and should conceive of God as a group person. They can by drawing on recent theories of group agency realism that show how groups can be not just agents but persons distinct from their members – albeit, I argue, persons of a different kind. They should because the resultant novel view of the Trinity – that God is three ‘intrinsicist’ persons in one ‘functional’ person – is theologically sound, effectively counters the most trenchant criticisms of Social Trinitarianism, and enjoys independent theological support from the biblical notion of ‘corporate personality’.

In this article I discuss two interrelated problems found in Richard Swinburne's cumulative case argument for the existence of God. First, I argue that the probabilistic proof apparatus introduced by Swinburne in The Existence of God (1979; 2004) would require him to provide a normalized preference order of God's intentions based on a measure of the ‘relative moral goodness’ of possible-world states. However, the approach offered in the second edition of The Existence of God (2004) fails to do so for various reasons. Second, a slightly different version of the argument briefly indicated by Swinburne might avoid the problems of normalizability but falls apart when meeting the criteria of relevant confirmation.

I present an argument for God's existence based on the idea that the possibility of God requires the existence of God as a ground. After setting this argument out, I compare it to other arguments for God, concentrating on an argument from Descartes's Third Meditation. I then address various objections and conclude by setting out a non-theistic version of the argument.

In this response to D. Z. Phillips's critique of my interpretation of Wittgenstein's view of magic and ritual, I counter Phillips's claim that I have misrepresented the Wittgensteinian view of ritual, consider the instrumentalist dimension of the Remarks on Frazer's Golden Bough, offer some objections to Phillips's expressivist view that a ritual ‘says itself’, and detect obscurantism in his approach to the study of religion.

Both Thomas Talbott and Friedrich Schleiermacher have argued, in somewhat different ways, that in the context of Christian theism the damnation of anyone would render it impossible to extend genuine blessedness to anyone else. I examine both Schleiermacher's and Talbott's version of this argument, which I call the ‘incompatibility argument', and respond to criticisms levelled by Jerry Walls and William Lane Craig. I argue that the argument is more powerful than its critics admit, and that it poses a potentially devastating challenge to what Thomas Talbott calls ‘moderately conservative theism', according to which the damned autonomously choose their own damnation by forever rejecting God's offer of salvation.

In this article three types of objection to a realist account of religious experience are explored: (1) the unusual character of its object; (2) its unusual accompanying conditions; and (3) the conflicting content. In response to (1) it is noted that despite divine freedom not all types of encounter preclude predictability, while parallels are drawn with perception of other complex objects such as persons. At the same time the whole notion of simple perceptions is challenged. In response to (2) parallels to the affective element are found not only in moral and aesthetic experience but more widely. Finally, in response to (3) apparent irreconcilable conflicts are lessened by observing how all such experiences take place within the context of traditions whose surface incompatibility does not necessarily indicate deep divisions.