Abbreviations

Prolegomena to a Philosophy of Religion (Ithaca NY: Cornell University Press, 2005): PR

The Wisdom to Doubt: A Justification of Religious Skepticism (Ithaca NY: Cornell University Press, 2007): WD

The Will to Imagine: A Justification of Skeptical Religion (Ithaca NY: Cornell University Press, 2009): WI

I am delighted to express my gratitude to everyone who has contributed an article on my work to this special issue of Religious Studies. In my replies I would naturally like to respond to criticisms and, where needed, to correct the record. I would also like to explain a few ways in which my mind has changed (if only to prove mistaken those who have denied it can happen). But I will try to do all of this while still concentrating on something else: namely, the central issues my colleagues have raised, and how their work in conversation with mine can nudge our understanding of these matters forward. Needless to say (but let me say it anyway), I will not be able to touch on everything in these articles that it would have been interesting to discuss.

Reply to Chignell

In his sensitive reflections on prolegomenous issues, Andrew Chignell considers the possibility that we in the philosophy of religion might come to link our philosophizing more closely to the rich complexity of religious communal practices. He calls the change he envisages ‘the liturgical turn’. And although Chignell says he's not aiming to defend such a turn in his article, two things seem clear: that he finds the idea of a liturgical turn intriguing, and that he finds my own proposals only partially in accord with it. I have several thoughts.

First, philosophy of religion is not like philosophy of science in featuring wide agreement as to the basic soundness of the practices under investigation. It follows that no recommendation making the assumption that religious practices as we see them today are in good shape, intellectually speaking, can be expected to commend itself to just any philosopher of religion. And although there is some unclarity as to precisely what a liturgical turn would entail, it does appear that it would most naturally turn on this assumption. Otherwise, why would Chignell fail to regard the matter of whether ‘bare’ theism or ultimism is true as one of the philosophically significant issues suggested by liturgy?

Second, by shifting our focus from theism, so often in the glare of the spotlight today, to ultimism we will, I think, have a better chance of hitting on an approach to philosophy of religion suitable for general consumption. Theism is just one of many detailed religious views, known and presently unknown, that we might conceivably explore as we work to achieve a firmer grip on the notion of ultimacy at the heart of so much religious practice, and as we consider whether ultimism is true and ultimistic practice appropriate in the first place. (By the way, ‘religion as “ultimism” ’, one of Chignell's subheadings, is accordingly misleading: there can be ultimistic religion, but ultimism itself, like theism, is only a view or proposition or claim.)

Now a number of philosophers, perhaps including Chignell, already accept that ultimism is true. And presumably they are comfortable with some existing elaboration thereof – perhaps it will be theistic in nature. So their way of pursuing philosophy within an ultimistic framework might well be different from mine. But there is nothing to say that those who accept proposals from PR must go along with the way I myself apply them in the other volumes of the trilogy. Indeed – and this is my third thought (a specific application of the second) – the ultimistic framework I have defended can even accommodate, in one quarter of the field or another, a liturgical turn! I regard the ‘bare’ idea of ultimism as an appropriate guiding idea not because, as Andrew several times suggests, it is more intellectually ‘tractable’ than much that we might focus on but because it is more inclusive.

Is it inclusive enough? In particular, what about religious-seeming claims such as the physicalistic pantheism mentioned by Chignell? Notice first that the latter view cannot, contrary to what Chignell suggests, be regarded as already a brand of ultimism. Chignell here conflates ‘axiologically ultimate’ with ‘axiologically fundamental’ (a term I never use). Perhaps there is a sense in which a pantheist might regard the physical universe as axiologically fundamental, but what I have denied is that the universe is axiologically ultimate in my sense of unsurpassably great. The latter is what I mean by ultimacy in value. So the relevant question is this: should a physicalistic pantheism be regarded as religious even though it is not fully ultimistic?

Here one needs to notice that on the view defended in PR, fresh prolegomenous inquiry – including inquiry into the meanings of ‘religion’ and cognate terms – should be appearing continually in philosophy of religion as it grows, and also that what I say about the ultimistic approach is a proposal about how to move forward from our theism-saturated times, one that, even were it to be accepted for now, might well be refined when we apply the results of future prolegomenous work. Having said that, I would defend my proposal for philosophy of religion today by saying that an ultimistic framework allows for a clearer distinction between the religious and the non-religious, is more capable of accommodating philosophical interests, and is truer to the most sophisticated forms of religious thought and practice humans have produced so far and the most powerful religious experiences they have undergone than any alternative approach of the sort Chignell appears to have in mind when speaking of a physicalistic pantheism.

Chignell's points can be seen as questioning the inclusivity of my approach in another way, too, given that my picture of ultimistic religion, as he sees it, places the emphasis on ‘interiority’, on a ‘first-person phenomenology and psychology’. This, he says, ‘gets things precisely backward’, given the communal nature of religion.

Here my use of such terms as ‘internal’ and ‘external’ in PR may have been misleading. It is important to notice that the former is not restricted to the individual person and that the latter does not cover everything outside the individual, including other persons. Rather the former, which for me is synonymous with ‘personal’, covers the thoughts and experiences and doings of people, whether alone or together. And the latter, which for me is synonymous with ‘institutional’, applies to the products of their activities as we see them in traditions spanning the centuries, which can therefore in the relevant sense be spoken about quite impersonally. I gave examples of communal religion when discussing my personal definition in PR, and I think it will be evident on close inspection that it includes such practices as are of interest to advocates of a liturgical turn.

I turn now to Chignell's probing thoughts about the concept of propositional belief. I think he's right that one might regard belief as directed to propositions even where it is not assumed that the latter are, at the time of occurrent belief, consciously being represented as propositions. But I am not persuaded by his arguments on behalf of the idea that belief is intellectual confidence – a disposition I have distinguished from belief. (I once held this view, but the thinking that went into PR dissuaded me.) It could be, as some of Chignell's remarks suggest, that such graded notions as those of confidence and credence are even more important in intellectual contexts concerned with evidence than is belief. Perhaps, indeed, what we see here is another reason for deliberately de-emphasizing belief. But, whatever the case, I think that in speaking of confidence and belief we are speaking of two different things.

To show this I would rely on my general arguments in PR about the nature of belief, which suggest that it is a matter of thinking in a certain way, not of feeling in any way. But the case of the ‘fearful frequent flier’, one of many realistic cases I have used, still seems effective too, in part because of a vital bit of the story that Chignell leaves out when considering it: our fearful frequent flier lacks confidence not just while flying but also while on the ground. She is almost always agitated to a degree incompatible with confidence when thinking about whether flying is safe. And yet she flies! Would she do so if she didn't believe that the planes she boards will get where they're going safely?

Chignell has a more general objection as well to what he calls my ‘stripped-down’ concept of belief. He thinks that when I speak of propositional belief as entirely distinct not just from confidence but from other forms of belief and moreover from propositional faith, I am ‘needlessly multiplying dispositions’. He refers to a ‘dramatic loss in theoretical economy’. To avoid this, he goes so far as to suggest, one might think of what I call voluntary non-believing faith as merely illustrating how belief can be realized both voluntarily and involuntarily.

Three things seem obvious to me here: that sometimes one is entirely in a state of doubt and thus without belief while needing voluntarily to take up a positive attitude towards a proposition nonetheless; that sometimes one does the latter while remaining in doubt; and that sometimes it is appropriate to call what results from doing so faith. The particular interpretations attached to these ideas, though important, don't matter as much as these central claims. Notice that for Chignell's simplifying suggestion about belief to be correct, we would have to say that it is possible to be in doubt about a proposition – not just feeling doubts but in doubt – while yet believing it. And this seems quite impossible. We've all experienced the loss of one belief or another: you know when you have it and you know when you don't; and even if you've only landed in doubt or uncertainty about a proposition p, you know that your situation is of the latter sort, and also that you can't get back to a situation of the former sort by mental fiat.

And what if in these circumstances one nonetheless voluntarily summons a positive non-doxastic attitude towards p, behaving mentally as though it is true? As it turns out, Chignell regards this as possible, but, following Kant, he thinks of what we see here as hope. Now hope may indeed be included in many cases of this sort, but if we want to capture fully an attitude that involves thinking well of the truth of the relevant proposition but also goes beyond this in voluntarily embracing the proposition, mentally speaking, and cultivating a disposition to think accordingly, it's hard to see how talk of hope could do the job. Certainly the most common concept of propositional hope among us today, which features no more than a positive evaluation and desire together with a modest estimate of probability, simply lacks the weight required for so robust a stance: in particular, there is nothing voluntary about it. Here I think that for theoretical reasons we should want an expanded conceptual space including a concept of non-doxastic faith, rather than, because of a one-sided emphasis on economy, choosing to stay in cramped quarters capable of accommodating only belief and hope.

A final comment. Discussions about such matters, especially when religiously applied, would benefit from some experimental philosophy. We could, for example, use some empirical evidence as to the nature of the non-believing but intellectually committed states, if any, already instantiated in religious communities. If we obtained it, perhaps we would even have found a way of getting started in liturgical philosophy!

Reply to Howard-Snyder

Daniel Howard-Snyder, unlike Andrew Chignell, holds that there is a way of exhibiting non-doxastic faith, but in his trenchant article he disagrees with me about its features. Some of these disagreements are relatively minor and I will not suggest here how they might be resolved. But others go deeper.

For example, Howard-Snyder thinks that some reference to desire is needed in the analysis of faith that p. I do not. I think the reference to a positive evaluation of p is enough. Faith seems to me to be tailor-made for (among others) people who haven't got their desires in order. Think, in particular, of someone who, gripped by depression or other mental illness or perhaps brain injury, is affectively incapacitated, just feeling dead inside, but who can see that it would be good to be in a different state: can't she have faith that she will achieve it? I think we should define faith in such a way that she can. Does she desire to achieve that different state? It certainly seems not.

What Howard-Snyder has settled on in this regard, for his own analysis, is the notion that ‘what's required is a desire in virtue of which one cares that p’. Should I go along with this? The first thing one may notice is that ‘caring that p’ seems hardly different from ‘desiring that p’! But this is a misleading appearance. Howard-Snyder actually has a fairly weak notion of caring that p in mind, according to which it means only not being indifferent to p, and he allows that one might avoid being indifferent to p without desiring that p.

Now if not being indifferent to p (even without desiring that p) is caring that p, then I would quickly agree that faith that p entails caring that p. Here the point captures an intuition I share. But then I would also say that I have already accommodated this insight with the positive evaluation condition. Even the affectively incapacitated individual described above evaluates positively the idea of her recovery. In so doing, she shows that she recognizes the value of recovery. Why then call her indifferent to it? She is certainly not cognitively indifferent. I would suggest that she clearly cares that p, and, given her psychological condition, that she does so without any desire at all – and so without the desire Howard-Snyder's formulation still refers to. If she goes on to do some of the other things that are included in my analysis of faith – for example, imagining her recovery and forming the intention to be mentally guided by this picture – then it will be even harder to make a charge of indifference stick. The upshot is that there is a dilemma here for Howard-Snyder: if one generates a strong ‘caring’ condition that is genuinely new, not already implicitly included in other conditions of propositional faith, then one has an analysis that can plausibly be counterexampled. And if one settles for a weaker condition, then it won't be genuinely new.

What about faith's relation to belief? In PR I argue that faith that p is incompatible with belief that p. Howard-Snyder finds this ‘flat out shocking’. We probably won't find it so if we remember that what we're talking about here is propositional faith only. Much of the time people who speak of faith have in mind a good deal more than a propositional attitude, and this more – faith-in or operational faith – is even in PR seen as being quite compatible with belief (I there argue that faith-in can have as its cognitive core either a voluntary faith-that or an involuntary belief).

However that may be, I have come to realize that the PR view of faith-that is probably too narrow to do justice to actual usage. While preparing for an APA session with Robert Audi and Robert Adams in 2009 – it was on the virtue of faith – I found on the internet examples of not-clearly-inappropriate propositional faith talk that entailed reference to belief. I also found non-doxastic propositional faith talk, mind you! What I would emphasize now is that the non-doxastic conception of propositional faith developed and applied in the trilogy represents a way of having such faith. But I think it is a way of having faith that has been overlooked because of our tendency to think of faith and belief together, and that it may be much more commonly exemplified than we suppose. I also think that it will turn out to be a particularly important way of having religious faith as we head into the future. Moreover, such non-doxastic religious faith, when operationalized, still represents what in PR I called a distinguishable faith response to religious propositions, which needs to be taken into account alongside believing, disbelieving, and purely sceptical responses.

Now this altered and more ecumenical position of mine may appear to be nicely in line with Howard-Snyder's way of dealing with the concept of propositional faith, which, together with an epistemic suboptimality constraint, has three conditions – evaluative, conative, and cognitive – that, as he emphasizes, may be satisfied in various ways. In particular, the cognitive condition may be satisfied either doxastically or non-doxastically. There certainly is something attractive about Howard-Snyder's approach. But even setting aside my rejection of his conative (desire-based) condition, one has to wonder, at least initially, whether his approach accommodates too much and enables too few relevant discriminatory judgements. For example, how does the hope lauded by Andrew Chignell fit into Howard-Snyder's picture? One might think it represents a way of satisfying Howard-Snyder's first two conditions, but could Andrew argue that it satisfies the third as well and thus, all on its own, represents one of the ways of having propositional faith non-doxastically?

It's not obvious to me how Howard-Snyder would respond here, but what does seem obvious is that with a strong condition of voluntariness for non-doxastic faith and the presumption that belief will be replaced by something at least belief-like, such issues would be avoided. My approach has both, but it's not clear that Howard-Snyder's approach or the one suggested by Andrew has either. This seems to me to be a point in favour of my approach. In particular, I can show why hope ought not to be conflated with faith.

But perhaps Howard-Snyder will have his own way of doing so. We should look forward to the fuller development of his approach. And he has various criticisms of the ‘voluntary belief-like element’ I have proposed, which in PR is called mental assent. Howard-Snyder sees that something like this is needed to plug a conceptual hole. But to provide it, he would replace ‘assent’ with another six-letter word beginning with ‘a’: ‘assume’.

Now it is not clear to me that the concept of ‘belief-less assuming’ allows the hole in question to be satisfactorily plugged. But I have never been very happy with my notion of mental assent either. Very recently I have started to wonder, partly under the pressure of Howard-Snyder's questioning, whether the ‘assent’ or ‘consent’ I have had in mind is sufficiently accommodated by the imaginative element of propositional faith that I have emphasized alone, so long as it is remembered that what we have here is an intentionally produced and maintained series of imaginings or picturings, rather than just the involuntary occurring to one, from time to time, of a certain picture of things. In PR I wanted something additional, the quality of mental assent, because an imagined state of affairs may appear in one's thinking without one going along with it in the relevant way. But if one deliberately imagines the state of affairs, directing one's mind to it for mental guidance whenever relevant matters arise, perhaps training oneself to behave in this way regularly, is there anything over and above what we've described for the ‘going along with consentingly’ to describe? There may indeed be something like an initial consent, which really amounts to the formation of the relevant intention – an intention, we might say, to call up this state of affairs in relevant circumstances and be mentally guided by it. And if the mental ‘yeses’ – for example, the yeses of the runner as, nearing the finish line, he says to himself ‘Yes, I will make it! Yes, I will make it!’ – are functionally understood, that is, understood as amounting to a way of keeping the picture before one's mind and fulfilling one's intention, then they do not refer to anything over and above the intentional picturing either but rather to something that (perhaps contingently) is for a time needed to keep it going.

Notice that this would potentially give us a nice, and more complete, parallel with belief: propositional belief involves a disposition involuntarily to picture or see the world a certain way, whereas propositional faith involves doing so voluntarily and intentionally. When layering faith that p over one's sceptical or doubting state, although one is not being involuntarily represented-to in the manner of belief, one deliberately represents the world to oneself through the power of the imagination as including the truth of p. Moreover, one forms the intention to continue doing so and to be mentally guided by what one imagines on an ongoing basis when relevant things come up, that is, to think accordingly in relevant theoretical and practical contexts as a matter of policy; and – in so far as faith continues – one follows through on the policy.

This characterization can be seen as filling out – while also making more voluntary and belief-like – what Howard-Snyder describes as assuming. Interestingly, some of what it contains is already implicitly there in what Howard-Snyder has to say. Notice that in his ‘Northbound’ trek he pictures to himself arriving at the Canadian border. (By the way, I can understand why reaching Canada should be a matter for such profound aspiration.) And as he imagines his arrival there, isn't he saying to himself something like: ‘Yes, I will make it’? Doesn't he have to repeat that to himself when doubts intrude and the delightful picture of Canada gets blurry?

Howard-Snyder's inquiries into the nature of propositional faith continue, and so do mine, though at present I am more interested in the lineaments of the non-doxastic propositional faith attitude I take myself to have justified as (part of) a response to ultimism, both in the trilogy and in my more recent work on evolutionary religion. Clearly, and at the very least, this attitude (the imaginative disposition just described) represents one way of instantiating a robust religiousness non-doxastically – a peculiarly appropriate one, it seems to me, for anyone future bound. And with belief replaced by a healthy dollop of imagination, and the latter's role properly conceived, we can avoid such criticisms as those Howard-Snyder has elegantly raised, as well as what may be liabilities in his own present account.

Reply to Cuneo

With Terence Cuneo's graceful article on the hiddenness argument we move from the first volume of the trilogy (PR) to the second (WD). That the hiddenness argument takes up but two of the latter work's fourteen chapters is an indication of the fact that it does not presently loom large in my thinking. (A better indication is that it takes up just two of the trilogy's thirty-four chapters.) As Cuneo suggests, the main role of this argumentation in my work today is to serve as one strand in a multi-strand defeater of any theistic attempt to defeat my argument for a general religious scepticism by inferring the truth of ultimism from that of theism. And this it could provide even if we remained unsure that the hiddenness argument shows theism to be false, and unsure whether Cuneo is right in his critique.

Having said that, it is of course an interesting question in its own right whether the hiddenness argument succeeds when taken as an independent attempt to show that theism is false. Perhaps it can show, all on its own, that theism should be regarded as an unsuccessful attempt to provide a convincing elaboration of ultimism. That it can do this much is challenged by Cuneo in his article. And in this context our being unsure whether he's right would vindicate Cuneo!

But before getting to his arguments, I should briefly address a question some readers may want answered about a possible connection between the ideas of PR, earlier at issue, and the hiddenness argument now under consideration. If a robust concept of non-doxastic religious faith can be developed, as I have claimed, should this make us less inclined to say that a perfectly loving personal Ultimate would facilitate creaturely belief? The short answer is no. A slightly longer one might point out that even if we should expect God to give creatures the opportunity of cultivating the difficult, virtue-bearing imaginative dispositions I have described, the extension of non-doxastic faith into non-religious contexts (seen both in Howard-Snyder's work and in mine) and the many uncertainties of the world we live in show that there can be innumerable opportunities of just this sort even if the opportunity to have non-doxastic faith that there is a God is not given to us, as a corollary of God's relating to us in deepest love.

Now to Cuneo's points. Some of the discussion in the first part of his article is marred by subtle misunderstandings of my argument. For example, he conflates being in a position to participate in conscious relationship with God just by trying with being in a position to become consciously aware of God just by trying. The ‘trying’ referred to by the first premise of the hiddenness argument is something one does after becoming consciously aware of God and in pursuit of some relational goal (it might, for example, involve expressing gratitude to God for what one has experienced as help in time of trouble), not something one does in order to become consciously aware of God – with that end in view. And the relevant sort of trying might not be easy, as Cuneo suggests (perhaps it will be hard to relate oneself properly to a God one believes to exist), even if one never has to try at all to become consciously aware of God.

But setting aside points based on such misunderstandings, we are left with Cuneo's central point in the article, which is illustrated when he speaks of the non-theist's positive interaction with goodness and beauty in the world as indicating that he is able to ‘apprehend God's presence and activity’. Cuneo takes more seriously than many theistic critics of the hiddenness argument have done what he styles my ‘consciousness constraint’, but he supports in a new way the rejection of any attempt to apply it to God, even where no independent reason for Divine hiddenness is available. A perfectly loving God, we are told, might be unnoticeably present to many non-resistant non-believers and yet at all times open to intimate union with each of them.

The argument for this striking claim runs as follows. The consciousness constraint does indeed apply to ‘admirable and intimate’ love between persons ‘in ordinary conditions’ but the only good reason for thinking so is that ‘typically we have no other available means [no means other than conscious interaction] by which to form and sustain the bonds of union and intimacy’. And this reason does not apply to relations between God and finite persons. Here there are other means by which such bonds can be formed and sustained, such as our appropriate response to the ‘goodness, beauty, and need we find in the world’. Thus we lack any good reason to apply the consciousness constraint to love between God and finite persons.

Cuneo's article is interesting, and I almost wish I could accept a view so attractively presented. (Almost …) But to advance our understanding of the relevant issues here, we will, I think, need to recognize that this argument is flawed. It errs at several points but these errors can be reduced to one: the notion that conscious interaction is only contingently tied to union and intimacy of the sort we admire and seek in love.

Suppose a woman brings into the world a son – call him Cuneo. She is as deeply loving as mothers can be and so, naturally, wants to form and sustain bonds of union and intimacy between herself and Cuneo. She assumes that this means being present to Cuneo's conscious awareness, interacting with him in the rich medley of ways that mothers do typically interact with their children. But then she remembers that the home in which they live is entirely of her design: the myriad opportunities for exploration and discovery in a young life that its hallways and bookshelves and toy closets represent are all of her making, and she has fashioned each one with the flourishing of Cuneo and his siblings in mind. Furthermore, she just happens to be the governor of the small island province in which Cuneo will spend all his days – a remarkably active and productive governor: marks of her benevolent activity will be discernible everywhere Cuneo goes even after he leaves the cosy house in which he was born and ventures out into the world. What all this means, she realizes with a start, is that there is another way of forming and sustaining bonds of union and intimacy between herself and Cuneo. For he will be aware of her through her works everywhere he goes! Even without knowing it, he will have the opportunity to be united with her in an intimate personal relationship by having the opportunity to respond to all these things she has done in appropriate ways. It's up to him. Will he, for example, be grateful when he discovers toys in the closet, so carefully designed with his needs in mind? Will he appreciate the aesthetic qualities of the kitchen layout, and of the beautifully landscaped parklands that dot the island? Will he play well with her other offspring – his brothers and sisters? If Cuneo responds in these ways then he is responding to her, just as when creating his world she was relating to him. Thus she need not do everything possible to ensure that Cuneo is able to relate to her consciously. Even if Cuneo never becomes consciously aware of her existence, bonds of union and intimacy between them can still exist. Of course in ordinary conditions such is not possible between a mother and her son. But theirs are not ordinary conditions! Since as governor she is, after all, rather busy, Cuneo's mother therefore leaves it up to chance whether he will become consciously aware of her and able to interact consciously with her or not. As it happens, Cuneo never does become aware of her existence and of the fact that she is his mother, and never does interact with her consciously. Nevertheless, because of his fine responses to what she has done in his world, the two of them live on for many years in a loving relationship of union and intimacy.

I submit that if Cuneo (the author of the article to which I am replying) is right, then this story is coherent. But it is not coherent. The description of her behaviour entails that the mother of the story is not as deeply loving as mothers can be. More to the point, the phrase ‘bonds of union and intimacy’ is in the story no more than a freeloading expression: it invites an evaluation that none of the facts of the case would warrant. Another way of putting this point is to say that the existence, between persons A and B who are related in love, of anything worthy of the label union and intimacy must entail certain things – such as, at the very least, friendship in a literal sense between A and B – which simply do not exist in the story.

Now perhaps there are ways in which the mother of the story could be good – or at least benevolent or a benefactor – despite her behaviour. Perhaps one might even try to use claims about countervailing goods to argue that there is reason for union and intimacy to be for a time placed out of reach (though we should remember that Cuneo does not wish to rely on such a move). But the experience of loving union and intimacy with another involves more than being on the receiving end of goodness or benevolence or benefaction. And so does the experience of an openness to such things on the part of the Divine. Cuneo at several points refers to my suggestion that when we speak admiringly of someone's love, we have in mind their openness to conscious relationship. And he seeks to undermine this suggestion with stories of benefaction. But he doesn't tell you what comes along with and before this reference to admirability in my work. The full statement is as follows:

It may be easy to blur ‘loving’ and ‘good’, but the former, when treated discriminatingly, will reveal distinctive features of any God there may be.

Of course it is to be expected that theistic religious traditions emerging in circumstances of Divine hiddenness will see things differently. They will try to make sense of God as sometimes distant: what else can they do? By the same token we should expect to find in the Christian New Testament such stories as Cuneo gleans from it, which as he shows lend themselves to interpretations compatible with Divine hiddenness. But it doesn't follow from their existence that there is nothing philosophically problematic in the conception of a perfectly loving God who is hidden! And if reflection should suggest that this conception is problematic – that the stories are built on philosophical sand – then continued reliance on those stories will hinder philosophical enlightenment rather than promoting it.

Reply to Crisp

Where Cuneo challenges the hiddenness argument, Thomas M. Crisp takes on the argument from horrors developed in WD. I want to underline again that my main aim in that book was not to show any such argument to succeed as a stand-alone proof of atheism, or even as one part of a proof of atheism, but instead to weave various atheistic arguments into a defeater of theistic attempts to defeat my more general religious scepticism. But, just as before, I will go along with my commentator's desire to extract the argument from its original context and to assess its force independently.

So what does Crisp have to say about the argument? Well, after developing some scepticism of his own about my view on the deepest good of finite persons created by God, he plays Cleanthes to my Philo, arguing that the idea of a deity who is limited in certain respects can escape the argument from horrors even if theism as more standardly construed cannot. This interesting move, which sharpens some ideas from process thought unavailable to Cleanthes, comes last in his article. But I will start with it. For if standard theism can easily or at least acceptably, in both philosophical and religious terms, be exchanged for Crisp's ‘broad theism’, and if broad theism is not challenged by the argument from horrors, then we will have no reason to linger over that argument.

Let me first say, in answer to Crisp's main question, that, yes, I think his version of theism does indeed escape the problem of horrors. But that is only because it was tailor-made to do so! And the religious and philosophical costs incurred are great. Crisp's God has a number of essential properties. Two in particular are important: non-coercive love and everlasting creativity. Now these sound quite attractive, as candidates for what might belong to axiological ultimacy personified. But notice that the non-coercive love comes with an incapacity for coercion; and the everlasting creativity is restricted to matter. Moreover, matter, on this theory, has strongly indeterministic tendencies, and God has no choice as to its potentialities. All God can do is seek to influence things in the direction of the good. In a manner quite beyond divine control the materials God has to work with may swiftly become recalcitrant and – in part because of the considerable room there is in this picture for the operation of chance – horrors may result despite God's essentially loving nature.

Let me bring out more explicitly now the philosophical and religious worries attaching themselves to this picture. Is it really true to Hebrew and Christian scriptures, as Crisp says? Well, perhaps it is in harmony with some parts of the latter – but the God of the Hebrew Bible seems to wield power rather differently. It's not at all clear that the theory could be regarded as designed with faithfulness to these scriptures in mind. Moreover, the God of every scripture inspires worship, and philosophical talk about omni-attributes, which Crisp dismisses as an ‘add-on’, can be seen as simply teasing out the intellectual content of such an attitude. So perhaps a theory designed with theistic scriptures in mind would actually look quite different.

How about God as metaphysically ultimate? Have we given up on this idea if we accept Crisp's picture? It seems so. God and matter appear to be co-ultimate in this theory, and because of that, important philosophical questions which one might hope such a theory would address remain unanswered. For example, why does matter exist at all? (God doesn't freely choose it for reasons, so the traditional answer is unavailable.) Other questions unanswered here: why is matter indeterministic, and why does it have the ‘potentialities’ it does have? Of course, one can always say ‘That's just how things are.’ But then why not simplify the picture, metaphysically, by taking out God and leaving matter, and say the same thing?

A related problem is that the restriction of divine creativity to the shaping of matter seems unrationalized and a tad arbitrary. General religious and philosophical desiderata of the sorts already mentioned do not warrant veering in this direction. Why suppose every creatable universe must be material? God isn't material, on Crisp's theory. So why might God not emanate an immaterial world? What seems to motivate Crisp at this point is nothing but the need to deal with the problem of horrors in a material universe like ours.

Furthermore, Crisp's God, despite initial appearances, doesn't really seem to approach axiological ultimacy either. Can't we think of persons greater than this one? Now Crisp says he doesn't worry about this because, again, he regards the omni-God attributes as dispensable ‘philosophical add-ons’. But here I think he gives short shrift not only to the sentiments of worship, which beg to be expressed in ultimistic terms, but also to the most powerful religious experiences, which purport to put the experient in contact with a reality of unlimited richness. This last strikes me as a deeply religious consideration. But philosophy too is relevant: surely the bias of philosophers towards religious ideas that may bring fundamental understanding is, in their area of inquiry, an appropriate one and not something we should resist.

Finally, is Crisp's indeterministic theory at odds with how things are in actual fact? He seems to be plumping for a particular interpretation of quantum mechanics, and dependent on its not being disconfirmed. Someone disagreeing with such an interpretation will say that the actual material universe is much less given to indeterminism than the one Crisp has described. Indeed, perhaps by building so much indeterminism, so much unpredictability, so much chance into the universe in order to deal with horrors, Crisp's picture only suggests a new argument for the non-existence of God: if God exists, then the universe in which finite sufferers exist is one in which indeterminism has a prominent role; the actual universe is not of that sort; so God does not exist.

It may be that at least some of these problems can be solved; Crisp is nothing if not resourceful. But it seems clear to me that given these problems, neither philosophers nor theologians would be wise in rushing to embrace his alternative understanding of the Divine just because it can handle the problem of horrors.

Assuming, then, that traditional theism will retain some interest for theists, we may wonder what should be made of Crisp's attempt, earlier in the article, to save it from my version of the problem of horrors. Here's what he does. He conjoins two premises from my argument and calls the resulting conjunction the Deepest Good Thesis (DGT). Then he claims that the DGT is a ‘recondite philosophical claim’, one that concerns matters deep and profound that are much debated and ‘far removed from the everyday concerns of life’. This is followed by an argument, inspired by Plantinga, to the effect that atheists who accept the basic fact of evolution through natural selection should be in doubt about the reliability of any faculty producing such claims, and so should be agnostic about the DGT instead of believing it. Many issues arise here; I can chase only a few of them, and only for a little way.

First, the problem with the DGT cannot stem, as some may think, from the complexity of conjunction, since we can always decouple the conjuncts and assess their believability separately. Moreover, those two propositions can be stated quite simply and intuitively. What they come to is this: (1) Because God is unsurpassable greatness personified, persons who grow ever deeper into God realize their deepest good. (2) Because God is infinitely deep and rich and exclusively good, endless opportunities for persons to grow ever deeper into God arise even where God has prevented horrors altogether. My rational intuition tells me, of each proposition, that it is true. Of course, this provides a defeasible justification for believing it true. So in the relevant chapter of WD I examine defeaters, including ones of the sort alluded to by Crisp, which appeal to free will and the alleged spiritual power of an experience of horrors. I conclude that they fail, in part because a relationship structured by free will is only one of a limitlessly large number of ways, each as good as the next, of experiencing an infinitely deep and rich reality. Now Crisp doesn't respond to these latter arguments. At the same time he suggests that philosophical debate over matters they concern should give me pause. But if those arguments have been shown to be sound then this is not the case. Thus Crisp's suggestion seems to beg the question against what I say in those arguments, and is at the very least inadequately supported.

But what about Crisp's more general evolutionary argument? Perhaps it can save the day. Well, I have some broadly evolutionary arguments of my own in WD, focused on the deep future instead of the deep past, which I take to support a new pessimism about many of our beliefs but also new insights as to what inquiry may yet achieve. And I argue that at so early a stage of inquiry it is appropriate for all of us to rely on universal and unavoidable belief-producing faculties. What else are we to do? Crisp expresses agreement with me on this point in an extended note at the end of his article. And in our present context this agreement becomes seriously problematic. For the faculties I have mentioned include rational intuition – which is just what I rely on in connection with the DGT!

Now Crisp tries to remove the force of this concession by arguing in his note that the evolutionary atheist might still have good reasons for doubt as to the reliability of rational intuition ‘in certain domains’, and that he, Crisp, has provided such reasons in relation to the domain inhabited by recondite philosophical claims. But although narrower in his sceptical focus than Plantinga, Crisp is still rather undiscriminating here. The relevant domain seems to be all of philosophy. Notice what he is therefore committed to, given also the reasonable concession I have noted: for the purposes of inquiry it is appropriate for all of us to rely on universal and unavoidable belief-producing faculties – except when it comes to philosophical inquiry!

There is nothing to justify such arbitrariness. Moreover, we ought to notice that even in philosophical inquiry, should we choose to pursue it, we can distinguish the more clear from the less. In short, ‘philosophical’ is not coextensive with ‘recondite’. I would welcome Crisp's help in identifying the sharper criteria we need here, for the purposes of philosophical inquiry. But no such criteria are needed to discern that some of the propositions we use in consequential philosophical arguments, including arguments from evil, are as clear as day. And it will not surprise you to learn that I think the propositions conjoined in the DGT are among them.

Reply to Dole

With Andrew Dole's entertaining and usefully provocative piece we move from the second to the third volume of the trilogy (from WD to WI). This is at the same time a move from the new pessimism I have already mentioned, stimulated in part by a clear awareness of distantly future evolutionary possibilities, to a new optimism resulting from the discovery of a non-doxastic form of religion peculiarly appropriate to our own place in time. Dole's central suggestion could be expressed by saying that if I anticipate this new form of religion easily becoming popular and institutionally established, I'm getting a little carried away in my imaginings. Sceptical religion, as he puts it, is ‘an odd bird’. My basic reply is that it is indeed an odd bird and that I don't think what he describes will be easy (in part because I've hardly thought about it at all), but that if as a culture we succeed in getting our ideas about time in order, this odd bird might yet be made to fly.

Let me develop a bit the remark I just made parenthetically. The issue whether sceptical religion might successfully be institutionalized and thrive over the long haul is not one with which I myself was preoccupied in WI. Here a distinction that emerges in my reply to Andrew Chignell, between personal and institutional religion, is again relevant. I have been preoccupied with whether the former is or can be rationally justified (and again I emphasize that ‘personal’ is for me compatible with ‘social’ even if it does not extend as far as ‘institutional’). So I have been thinking about sceptical religion in the sense of religiousness. I have not thought much about whether what we see here might be turned into a religion. And you'll be happy to hear that I haven't thought at all about whether I might become the founder of such a religion!

Why then do I have the two chapters in WI in which the question appears to be what sceptical religion practised might look like, and in which I talk about sceptically religious communities? Because this was something needed at a certain stage of my reasoning to show that sceptical religion has certain properties it might appear not to have: the properties of clarity, substantiality, and feasibility. This result, in turn, provided the minor premise for an argument whose major premise is that if it possesses the mentioned properties, then sceptical faith or sceptical religion is the sort of religion that is justified, if any is, for a twenty-first-century sceptic wondering, after taking on board the results of PR and WD, what her religious options are. Later in WI, of course, I argue that such faith is justified, and in a very strong sense indeed. What follows is that reason invites us to practice sceptical religion.

Now perhaps a lot of people will pick up this ball and run with it, perhaps not. I have put such ideas forward as proposals to be debated in philosophy. If they survive debate, there will of course be more reason to think about real-life sceptical religion. But I have been inclined to think that I'd be comfortable, even at that later stage, with a disjunction of possibilities: (1) sceptical religion is picked up and institutionalized soon, or (2) it is picked up in the near or distant future after other forms of religion have lost vitality because of a general increase in scepticism, or (3) it is never picked up in a thorough or wide-ranging manner but is practised by a few individuals or groups here and there in the history of humanity – and their religious practice, unlike many or all others, is rationally justified.

Dole challenges this somewhat laissez-faire attitude of mine, and I must say he has a point: if one of the claimed benefits of sceptical religion is that it could help us dig ourselves out of some holes we've fallen into as a species, as I've been arguing more and more noticeably of late, then it could hardly be of no concern to me at all whether sceptical religion, or something like it, becomes, as Dole is inclined to put it, ‘historically actual’. Perhaps my reluctance or slowness here bears the imprint of my Mennonite background. I grew up ethnically a Mennonite and – particularly because of my father's influence – religiously an evangelical Christian. Then I discovered and for a time strongly identified with the Anabaptist tradition, flowing from the Radical Reformation, to which religious Mennonites belong. (My parents had never told me about it; I suspect they knew little of it.) This was all before I discovered philosophy. Perhaps it is understandable that someone influenced by a tradition content with minority status and relatively unconcerned about ecclesiastical distinctions might not think a large presence in the world, or complex institutional structure, all that critical to the success of a form of religion.

Whatever the facts on that may be, suppose I take Dole's point here. This gets me as far as interest in the question whether sceptical religion could become adequate as a religion (or, perhaps better: whether it could become a religion). Dole's title is still not quite right, since it suggests an intention the trilogy was not written to fulfil. Nor do I need to have that intention now. Much less must I believe that sceptical religion will become a religion in order to put forward my proposals reasonably. What might, however, be expected is hope that it will, in light of a belief I do have: that if it is more widely taken up, we may see a contribution to such benefits as Dole has mentioned.

Such hope is enough to warrant taking seriously Dole's pessimism about the institutional prospects of sceptical religion. Now that he's won me over to interest in his topic, I'm more inclined than I otherwise might be to point out that he has been taking something of a ‘glass half empty’ approach to that topic. The sensible thing, it seems to me, is to consider the various obstacles Dole and others have noticed or might notice and to think creatively about how they could be circumvented, with a strong emphasis on not crying over unspilt milk. It's still early days – evolutionarily and also for sceptical religion.

I have very little space in which to record such creative thinking here. Indeed, I think the main thing to be done just now is to thank Dole for exposing the shape of so many new inquiries to which the trilogy might lead, of which I had no inkling when I started writing it. But having said that, I do have a few remarks that may provide some small support for replacing Dole's pessimism with what I have already termed a ‘new optimism’.

(1) Dole wants to hold me to the idea that for a religion to attain what he calls subjective religious adequacy, there have to be activities which its adherents take to promote their ultimate good. And here he emphasizes how on my view we may just be wrong on many detailed matters concerning value, with the truth about such things available perhaps only to inquirers in the far distant future. The effect of such scepticism on the religious life, Dole appears to think, must be crippling or at least demoralizing.

When thinking about this challenge we should first notice that it's our ultimate good we need to be talking about – the ultimate good of a hominid species at a quite early stage of development. And perhaps that will be different in its fine-grained features from the ultimate good for possible creatures down the line, temporally speaking, who may have got closer to an understanding of things that still baffle us. Indeed, we should expect that uncertainty about many things will be woven into the nature of the best way of life for us. Our deepest good may involve searching without finding in relation to many topics.

And if such thoughts seem too weak to motivate, as Dole several times suggests, because of the possibility of error even about matters pertaining to our own good, then we need to remember that we could only turn out to be wrong if we were acting on specific instructions presupposing the truth of detailed religious propositions – instructions that could be countered by the final truth about value. But sceptical religion only involves doing what's appropriate to certain general propositions embedded in or entailed by ultimism. We can know what these are, and also something about the actions that – for us and at our stage of development – are appropriate if those propositions are true. So adherents of a sceptical religion will have reason to deny that what they are doing could be incorrect or unreliable or unhelpful. We lack the whole story, yes. But that has to be written into any religion of today. It's not as though any other form of religion will be seen by the aforementioned adherents as in a better epistemic position. Again, with sensitivity to our place in scientific time the entire context for our thinking about religion is changed.

(2) What about children? Is sceptical religion going to be inhospitable to them? This is an interesting question. Notice first that the behaviour of children is in fact one of the many features of ordinary life to which an adherent of a sceptical religion would seek to become more discerningly attuned, if my descriptions in WI of what a sceptically religious person might do are at all on the right track. But quite apart from this, we may respond to Dole's pessimism here simply by pointing to such historically established forms of religious life as those of the Unitarians or certain Buddhist groups. Sceptical religion provides at least as much of an avenue for the participation of children as do these forms of religion. (What can a child understand of the no-self doctrine, it might well be asked, or how can we expect a child to enter into deep and prolonged silent meditation?)

(3) Let me finish with a brief word on natural religion, of which Dole thinks sceptical religion is a species. Suppose it is. And suppose we take with full seriousness his point that previous forms of natural religion have not become ‘historically actual’. Still, we might say, sceptical religion has something none of its predecessors have had, which may help it succeed where they failed: this is, once again, the vital insight about our place in time.

Assume with me, just for the sake of our discussion, that evolutionary thinking, including a keen awareness of the Great Disparity between the few thousand years of thought and feeling we've invested in religion so far and the possibility of millions more to come (if not for us, then for species that follow us), spreads through most of humanity within the next two hundred years. Such a development would make scepticism about religion – as opposed to either belief or disbelief – more widely attractive, and a faith attitude compatible with scepticism correspondingly attractive. It would also increase the attractiveness of ultimism (properly understood) as an object of faith. For now its disjunctive nature will seem much more appealing. It will appear as a mosaic of possibilities to tantalize and enrich the imagination circling about a general core affirmation, recognized as more appropriate to an early stage of evolution than any detailed core affirmation at war with others.

Perhaps natural religion only needed more time.

Reply to Diller

I think Jeanine Diller may have been looking over my shoulder when in my reply to Chignell I emphasized accessibility and inclusiveness. In her challenging yet intellectually companionable piece, these desiderata are repeatedly stressed as absolutely central among those that any approach to philosophy of religion must bear in mind. And it is argued that ultimism, as I have construed it and suggested we might employ it, comes up short in relation to them.

Diller sees herself as standing up for religious diversity and for religiously and philosophically important yet non-ultimistic ideas. Because they manifestly exist, she argues, philosophers of religion should spread their net more widely than the ultimistic framework, as PR conceives it, would allow us to do. And even though, as Diller generously allows, ultimism makes for an interesting object of religious faith, it would still be good, in her view, to construe this option more flexibly than I do in WI. She would allow for the idea of a being greater than any in the actual world but not the greatest possible, or that of a being singly or doubly ultimate rather than triply ultimate, as in my conception, to count as identifying a way for ultimism to be true.

In response, let me point out that it is one thing to suggest that such ideas as those just mentioned, when incorporated into someone's faith stance, deserve attention from philosophers of religion and another to say that what they refer to deserves to be brought under the canopy of what I have called ultimism. I have some sympathy for the former claim (more on this in a moment) but I deny the latter. I deny it mainly for the following three reasons. (i) A central motivation for distinguishing ultimism from other propositions – and notice that it is a proposition and not a form of religion, as Diller sometimes suggests – was indeed to distinguish and identify clearly the idea I had in mind, whereas Diller's ‘Ultimism II’ is compatible with so much that it threatens to go quite out of focus. (ii) I believe the notion of a ‘limited ultimate’ is incoherent or badly named: in my view the word ‘ultimate’, especially in a religious context, should be reserved for what cannot be transcended in the relevant respect(s). (iii) Even if there are propositions not entailing it that deserve to be called religious, ultimism as I have characterized it neatly captures a centrally important religious possibility and with (but only with) its three-sidedness allows us to see quite clearly what makes it religious.

It may be, however, that by encouraging the idea that ultimism be loosened up, Diller is merely generously suggesting a way in which I could be right – or close to right – in saying that it provides a good framework for investigation in philosophy of religion. It's clear that she doesn't think I'm right as things stand. So let's have a look at this issue. Even if we don't call them forms of ultimism, should the claims Diller has mentioned, which she thinks I exclude, be considered to be among those within the purview of philosophy of religion?

One way to resolve the issue might involve distinguishing different understandings of ‘philosophy of religion’. It's clear that the phrase is used variously. On some understandings – perhaps Diller's would be among them – the field overlaps with theology at one end and certain parts of religious studies at the other (here it may be noteworthy that two of the views I am said to have excluded are called functional theology and end of being theology). So perhaps the problem is just that I have a narrower understanding of the field than Diller does – which is of course compatible with a very broad view of what should be studied, all things considered, including things falling into philosophy of religion differently conceived.

I myself would very much emphasize the ‘philosophy’ part of ‘philosophy of religion’. And philosophy, in ways I have no space to go into here, I conceive as aimed at the widest and most fundamental understanding. Now just on this point Diller makes one of the most interesting comments of her article: non-ultimistic truths ‘might be the deepest truths’ about fact and value. And relatedly: ‘Might not our deepest questions about value and reality be answered in limited ways?’ I suppose she means here that something like a limited deity, axiologically non-ultimate and maybe soteriologically challenged too, might nonetheless be metaphysically ultimate. Shouldn't religious-seeming options like this be brought to the philosophical table?

I can see the motivation for saying so, but here's why I think they should be sent to the back of the line instead, at least in philosophy. (Of course I am under no illusion about the efficaciousness of my opinion.) Human investigation into matters religious, as into other matters, has just begun – or at least we will see things this way when we have fully internalized findings about deep time resulting from investigations in science. And although ideas about limited deities have emerged in religion during the first few thousand years of its history, much more religiously and intellectually interesting and momentous religious ideas have also arisen, including ultimistic ones. Philosophically, it seems to me, one would have to have given up on the fully ultimistic idea – what Diller implicitly calls Ultimism I – to be focused on anything else.

Now in traditional western philosophy, where ultimism is represented largely by the dubious and much-discussed theism, it might seem that we do have to give up on it. From limited personal gods we have moved to the Unlimited Personal God, and according to the most recent reports, many have soured on the latter. But this is all before the new temporal orientation and the distinction between theism and simple or generic ultimism. If we see that we've just got started in religious investigation and that theism is merely a species from a broader genus, then I think we can also see there's no reason to scale back yet. Bishop and Bacon and others Diller has referenced appear not to notice this option – the option of, as it were, moving up to generic ultimism rather than down to a scaled back or limited version of theism. And at least within philosophy, and given the appropriate temporal perspective, the former is the move that seems most appropriate. (The idea of a limited god, for example, may in ways similar to what we saw when discussing Crisp's proposal raise philosophical questions that are silenced within an ultimistic framework.)

I suspect that given a new temporal orientation ultimism may come to appear most attractive in purely religious terms too. Take worship, for instance. Worship, Diller tells us, has in past religion often been purely calculative and instrumental, carried out by persons who may have regarded the object of their worship as something rather less than ultimate. I understand this use of the word ‘worship’. But there is another use at least equally important. On this to my mind more deeply religious understanding, worship is much more than a calculated response involving statements and bodily movements designed to elicit favourable divine action; this second worship involves the heart, the emotions (content perhaps more obviously signalled by the near synonym ‘devotion’). Though the calculated response is clearly present in institutional religious contexts, whether ancient Greek or modern Christian, so is the other. And typically the latter is an emotionally extravagant response that at least tends towards implicit or explicit ultimization. Such a tendency is important here because given its existence, when ultimization is extended and made explicit in philosophy we can say that we are only drawing out or taking to its logical conclusion something incipiently present, even if alongside less obviously ultimistic tendencies, in the forms of life most widely viewed as religious.

And what about religious experience? One reason for taking something like ultimism more seriously even within purely religious contexts than Diller appears to do comes straight out of much religious experience, including especially the most powerful such experiences (in particular much so-called mystical experience). It's precisely the sense of an absolutely limitless richness that most makes me want to apply the term ‘religious’ to an experience. There has to be some term for experiences and ideas that want to burst all limits; what would it be if not ‘religious’?

So a sceptical ultimistic approach can accommodate much that is important and indeed central in religion as we've had it thus far. And what is even more important, to my mind, is its embrace of the idea that much deeply significant religious diversity may not yet have come to light. Because of our early place in time, we need to be open to the possibility that our best religious ideas are still ahead of us. (Talk about inclusiveness!) And what could provide a better framework for their exploration than the most broad and deep and capacious and interesting idea religion has yet produced? I, of course, am not responsible for producing this idea. It has been with us for thousands of years. All I am doing is drawing attention to it, pushing aside the weeds of thought that have obscured it from our eyes.

I suggest that we need to think about philosophy of religion in a manner that allows us to be part of a trans-generational process of inquiry – a long process weaving its way through deep time that we who recognize our evolutionary immaturity will imagine to be unfolding in order to give life and hope to our present inquiries. And we should at any rate begin with the biggest and best religious idea we've got. In this context, it seems to me, the ultimistic proposal makes sense. We don't need a God from our time or for our time alone. We need a God for all time. Better than any alternative, this is what ultimism provides.

Reply to Morriston

Wes Morriston's article is challenging and penetrating. He vigorously criticizes three of my arguments, while saying he is ‘broadly sympathetic’ to my project. I hope that means he will welcome my next paragraph, in which I show how the broader resources of that project might permit me to absorb the criticisms, even should they be in every case sound.

The basic point is that WI contains many different ‘modes’ of faith, eighteen in all. These criss-cross the project of human living, drawing alternately on personal, moral, intellectual, and aesthetic considerations and identifying certain associated aims that we ought to pursue at an early stage of evolution. The more of these aims you accept, the harder it will be to say that their conjunction can be satisfied without sceptical religion. Also, some aims are reinforced by others. Finally, the overall claim of the book depends not on any single argument or conjunction of arguments but rather on the disjunction of all these arguments, with different disjuncts perhaps proving convincing for different people.

But now for a more direct response to Morriston's comments, starting with Anselm's Idea. He is right to want to ‘distinguish an idea of something that would be wonderful if it existed, and a wonderful idea of that same thing’. But I don't think the Anselmian alignment argument has got hold of the wrong side of that distinction. When one regards the idea of x as a wonderful idea of x, one is implicitly distinguishing between the idea, regarded as an idea, and its content; and one's attention has turned from the latter to the former. Perhaps one thinks the idea can be put to work in inquiry in valuable ways, or that it confers credit on its conceiver. ‘Great idea!’ one says. But nothing like that is present in my argument. Anselmian contemplation is directed to the content of Anselm's idea – to there being a metaphysically, axiologically, and soteriologically ultimate reality. (The difference here is similar to the one Andrew Chignell rightly insists on in connection with the nature of belief.) And this content, sufficiently penetrated, will lead a mind into wonder and awe, whether one thinks there is anything answering to it in the actual world or not. After all, it's not that the properties of metaphysical, axiological, and soteriological ultimacy are instantiated in the actual world, if they are, that invites wonder and awe of the kinds at issue here, but rather what is instantiated.

In response to my alignment argument, Morriston suggests that maybe it's only given this or that filling for ultimism that we will get my result. But here I think there is a misunderstanding, similar to one I detected in Andrew Dole's piece. When we act on ultimism, doing what it would be appropriate to do if it were true, guided by a willingness to imagine it as true, we are not going out on a limb about its details, making assumptions that could later be falsified even if it is true. What we are acting on is the most general content of the idea, and this could not later be disconfirmed by some detailed version of ultimism – since it is entailed by all of them.

What about the Pascalian discussion? Here again Morriston thinks my reasoning is bedevilled by details – in this case, details about differences among individual human beings. He makes some good and interesting points, and one answer to them involves conceding that some twenty-first-century sceptics may rightly focus on non-prudential considerations in WI, considering them either as alternatives to Pascal's Wager or as ways of buttressing it. I would especially recommend the argument I build on Leibniz's Ambition as a way of allaying specifically intellectual fears, such as the fear Morriston suggests of betting one's life on an ‘illusion’. An evolutionary sceptic thinking about sceptical religion is not contemplating some narrow belief system of today but is letting her imagination linger in the thought of an understanding of the world beheld by finite beings but so magnificently deep that all our present thinking, even if it contains much truth, is just the tip of the proverbial iceberg. If this turns out to be an illusion … well, better to have loved and lost than never to have loved at all.

But there is another answer too. Clearly there are plenty of things – for example, in respect of exercise – that I may find annoying and distracting and may even be tempted to regard as unnecessary but which are in fact good for me; my life would go better for me if I took them more seriously. Might not the same be true of sceptical religion, even if (as I do not believe) it may easily come to feel burdensome? Especially at an early stage of evolution we should be thinking of using all the resources available to us to grow, to enlarge ourselves, and to manage the various problems so often attached to our actual inclinations, even if this is sometimes felt as an unwelcome intrusion precisely because of the nature of those inclinations.

The third argument Morriston examines concerns Kant's Postulate. Unfortunately, he says almost nothing about the second half of it, in which appear certain additional premises, based on Kant's motivationalism rather than his presuppositionalism, which are intended to support the operational element of sceptical faith. As I point out in WI, since operational faith presupposes propositional, this reasoning gives us an independent justification for the latter, and so might compensate for weaknesses in the reasoning Morriston examines, should there be such. In the Kantian case, even more than in the others, there is really no substitute for detailed engagement with my arguments. Not having the space here to indicate many details, I will content myself with repeating some of the premises from the additional reasoning in question:

Here again, though in ways I could have brought out more fully, evolutionary considerations come to the fore, helping to make the case that whatever we may say about the admirable examples of moral commitment without religion that Morriston draws to our attention in connection with Billy Kwan and Bertrand Russell, philosophers cognizant of our immaturity will advocate the overlooked resources of sceptical religion for the benefit of morally serious seekers who wish to avoid demoralization as they face up to the severe challenges of the future.

I come finally to Morriston's suggestion that I should be more worried than I am about the idea that there are good hiddenness-based or horror-based reasons for concluding that ultimism, like theism, is false, or that we are justified in believing there to be no afterlife of the sort that ultimism appears to require. Since I will shortly be responding to Terence Penelhum's concerns about the afterlife, let me set that issue aside here. On the other matter, my basic reply is this: that it is something specifically about a personalist conception of the Ultimate that permits the argument from horrors or the argument from hiddenness to have such force. In the absence of specific insights about, for example, how empathy or love necessarily leads a person to behave, I don't feel I know my way around sufficiently well to have any confidence in either sort of reasoning.

The point is that it is very hard to get the sort of grip needed here on the huge disjunction to which ultimism is equivalent in the absence of more detailed content. (Notice that Morriston emphasizes a similar point in his own reasoning.) Now it is important and can be difficult to achieve a proper balance: when do you draw a conclusion and when do you say we should wait for more evidence? Many philosophers today would say that we are rushing things even if we declare theism false, as I have done. I would say that we know enough to do so. The details theistic ideas contain allow inferences about what most fundamentally has value and how it is realized if this filling for ultimism is realized – and also the inference to atheism – to be made. And I say we should get on with exploring other fillings for ultimism, leaving open the possibility that ultimism itself is true. The idea is that even at this early stage of investigation we should draw conclusions where we can, to help keep inquiry moving, but be very careful not to foreclose inquiry where we shouldn't. The distinction I have suggested between the epistemic status of ultimism and that of its personalist elaboration seems to me to get this balance about right.

Notice that ultimism isn't necessarily always going to be in the clear, epistemically speaking; it's just in the clear now – at the present stage of human investigation. And notice that there is no implication that nothing can presently be known about value. Let's not make that mistake. Again we need to strike the proper balance. And so I agree with advocates of a more general problem of evil that some propositions in this neighborhood are obviously true and escape the problem of Total Evidence set out in WD. Most fundamentally, we certainly must regard horrors as intrinsically very bad. How could one avoid thinking this without forsaking all thought? In my view one may also, as already suggested, justifiedly believe certain claims about horrors and love and empathy, using them to argue against traditional theism – seeing here the possibility of a critical ‘next step’ in inquiry about things religious. But ultimism, which tells us only that a Divine reality exists and not how, is another matter.

Reply to Penelhum

I come at last to the comments of Terence Penelhum. Reading the thoughts, in this forum, of my first teacher and mentor in the philosophy of religion, someone whose personal character and intellectual judgement I have always greatly respected and admired, is a rich experience. I owe a great deal to Terry, and I hope he can sense the considerable extent to which what he admires in the trilogy is his own doing.

But I must return comment. First, on implicit trust: I agree with Penelhum that implicit trust involves no difficult ‘choice taken … in the face of alternatives’. Indeed, this is what distinguishes implicit trust from explicit trust and allows the latter quite naturally to be paired with non-doxastic propositional faith. And I wonder whether he would agree with me that the alternative realized by implicit trust still involves a certain sort of behaviour or behavioural disposition (one that includes the very ‘not-thinking-to-question’ he refers to), a natural disposition to act on the belief that the object of trust is trustworthy. If so, there may be little disagreement between us on this matter.

Now it might seem odd to call such implicit trust voluntary precisely because it is not the sort of thing one chooses to have – or even could choose to have – in the face of explicitly considered alternatives. But the PR account (see, for example, pp. 147–148) does not require us to say that trust is voluntary in this sense. Trust is in a way accessible, achievable by choice, even in such a case because it can continue in a more explicit form when the implicit variety is lost, and because the transition from the latter to the former involves a choice. But the implicit sort of trust is voluntary too, in a couple of ways I have distinguished from accessibility: namely, by being terminable and vulnerable. If one chooses no longer to behave in the relevant way, one will lose the trust, no matter how firmly one continues to believe in trustworthiness; and in the absence of sustaining activity it will be lost, whether one chooses to lose it or not. All of this seems supportable by reference to examples from ordinary life. And if so, what I say about implicit trust may not involve too much in the way of ‘normative tidying-up’.

Penelhum is correct to say that sceptical faith is ‘not through and through beliefless’. But where exactly is belief to be found in it? Well, there is the evaluative belief that the truth of the proposition in question would be a good thing. But this, presumably, would have been held by Braithwaite too, from whose account Penelhum helpfully seeks to distance mine. Penelhum suggests that someone who has my sort of faith will moreover believe that while the proposition in question, with respect to which she has faith, is not clearly true, its denial is not clearly true either. And this certainly seems right – otherwise the label ‘sceptical’ would not apply. But should we characterize this in terms of belief or in terms of the absence of belief? Penelhum appears to do both, but I am wary of the former route, since many different beliefs (beliefs about the degree of evidential support, for example) would represent a sufficient condition for the truth of what's really important here: that the person of sceptical faith that p lacks both the belief that p and the belief that not-p. Now most often what I myself have done is to refer explicitly to the first of the two states just mentioned (lacking the belief that p) while presupposing that our person of faith is in the second state (lacking the belief that not-p). The contexts of discussion in PR and in WI, which, after all, was preceded by WD, were such as to allow for this. But I think Penelhum is right to call attention to the fact that where we bring non-doxastic faith into other contexts, it needs to be made clear that both beliefs are absent to prevent someone conflating what I am advocating with an account like Braithwaite's.

This takes us to another concern of Penelhum's: even if Braithwaite and other religious reductionists are premature in thinking religious claims false, we don't want to go to the other extreme where nothing at all can falsify the claim we take up in faith. Do we go to that extreme with so general a claim as ultimism? Now the first thing I want to say here is that I think it's a good thing, in relation to the hope for a reconciliation of faith and reason, that ultimism is less likely to be shown false – more stable – than propositions such as theism. But, as I suggested in response to Morriston, ultimism is not immune to counterargument. Indeed, both Morriston and Penelhum have suggested a way in which someone might falsify ultimism or at least force a revision somewhere among the propositional commitments of sceptical religion: by providing decisive evidence that human consciousness cannot survive the death of the brain or be recreated.

So let's think a bit more about the afterlife problem. Can we already show ultimism to be false, by showing there to be no afterlife? Here I would mention some of the properties discussed in chapter 1 of WD, which can justify doubt in relation to propositions possessing them – and all the more so for beings at an early stage of evolutionary development. Take in particular precision, detail, and profundity. The anti-afterlife belief that we are concerned with is certainly precise, having clear content. But contrary to what may at first seem to be the case, it is also rather detailed. Unlike other anti-afterlife beliefs one might imagine (for example, the belief that we do not live on thanks to the brain's being miraculously revivified, in every case, by God at exactly one second after complete brain death), it rules out every way in which there could be an afterlife. In fact it is equivalent to a large conjunction of denials, corresponding to all the ways in which the afterlife could be understood. And this detail contributes to its evolutionary vulnerability because each form of afterlife mentioned represents an alternative to the anti-afterlife belief – a way it could be false. Moreover, in the general anti-afterlife belief we have a certain degree of profundity: a fairly deep understanding of our nature and destiny comes with such a belief, and many other understandings are ruled out – for example, all ultimistic ones! What follows is that there may well be many alternatives to the anti-afterlife belief we don't know, perhaps tied to transcendent truths we can't know. We're ruling these out too when we say there is no afterlife. Are we really in a position to do so at the present stage of inquiry?

For how do we know that there aren't positions we can't understand supported by evidence we couldn't assess which entail that there is an afterlife? How can we rule out the existence of facts – whether discoverable by us or not – that cancel the force of those physical facts on which arguments against the afterlife are based? Now note carefully that I'm not saying it might be the case that mental events are not causally dependent on brain processes, as we all rightly believe. I'm saying that even if this is indeed undeniably the case, it might, because of what the correct solutions to problems of consciousness and the self and religion contain, also be a fact that in one way or another I am not thus dependent, or that even if I am, a new ‘platform’ for my mental activity can be acquired. Precisely because of what we know about our ignorance concerning consciousness (presently a hugely controversial subject) and what we don't know about how deep that ignorance goes (more generally, what we don't know concerning how ignorant we may be about our own nature and, not unrelatedly, about the ultimate structure of reality) – because of all this, denying the afterlife today instead of accepting scepticism on that vexed subject is more like denying that any model of string theory scientists will conceive can be made to work (something quite unjustified) than it is like denying that natural selection plays no role in evolution (something quite justified). Now if we had a clear picture of how the brain produces consciousness, assuming it does, or even a satisfying way of identifying what consciousness is, or if there were in this vicinity some apparently necessary truths to guide us, things might be different, but as things stand, scepticism is the order of the day.

These, at any rate, are some of the things that can be said (and that to my mind have weight) against the afterlife worry. But it is an important worry and I am glad that Morriston and Penelhum have raised it.

I come, finally, to Penelhum's intriguing thought experiment about how things might have gone rather differently in my work had I started with the arguments for religious scepticism to which I have just alluded. Is there something a bit odd – maybe even paradoxical – about religion after atheism if that religion is imbued with evolutionary scepticism?

I recognize that many may think so. And my atheism may furthermore represent a strategic liability if my critics, who after all are mostly theists, come to be preoccupied with the anti-theistic arguments raised in WD instead of deeply engaging the broader sceptical argumentation I have carefully developed. Perhaps the articles of Crisp and Cuneo already reflect just such a tendency. But all I can do here is to ask again that readers consider how the atheistic arguments are in fact used in my more recent work. This is not at all in the dogmatic way Penelhum suggests, and so the alleged paradox is immediately dissolved – at least if it is supposed to appear in the trilogy. One need not be convinced by those arguments to believe that atheism is true in order for their purpose in the trilogy to be fulfilled. That purpose was again to defeat any attempt to use confidence about theism as a defeater of my broader sceptical case. A justified confidence about atheism would of course be sufficient for such defeat but it is not necessary: doubt or agnosticism would be quite enough, and what I argue in WD is only that I have justified at least this much in the way of a non-theistic attitude.

It follows that one could accept all the arguments of the trilogy without being an atheist at all. That its author is an atheist should not be allowed to distract from this fact!

Now, having said that, I do independently believe that there is no paradox in an atheist such as myself also being an evolutionary sceptic. Here I would recur to such points as I made in response to Morriston about finding a proper balance in what we believe or disbelieve and what we find subject to doubt. Since evolutionary scepticism emerges within inquiry and as part of an attempt to further the goals of inquiry by means compatible with our present primitivity, one should expect that not all beliefs will be regarded as off limits by it; certainly the clearest of our results – and I hold that traditional atheism is among them – may be regarded as ones we can take with us into a future that in ways both intellectual and spiritual may be immeasurably richer than our past.

But here we enter the enormously difficult broader issues – issues, really, about prolegomena to inquiry rather than about prolegomena to a philosophy of religion – on which evolutionary scepticism invites us to reflect. So let me conclude by inviting readers to venture with me into this new terrain, with at any rate an openness to both philosophy and religion after atheism.