My thanks to Mark McCreary for his paper,Footnote ¹ and to the editor of this journal for inviting me to respond. McCreary raises an interesting and important issue, and in explaining why his arguments do not succeed, I hope also, more positively, to advance our understanding of that issue. The issue is whether one can consistently be as sceptical about present human intellectual capacities as I am in the first two parts of my recent book, The Wisdom to Doubt (WD),Footnote ² and as confident as I have suggested we have reason to be about the central premises of certain arguments for atheism, such as the hiddenness argument.

But before addressing this issue, a couple of preliminary remarks. First, McCreary's case must be taken as a case for the inappropriateness of holding instances of both of the two kinds of argument alluded to in the previous paragraph to be sound: he has no case for the failure of hiddenness arguments per se. I shall charitably ignore the part of his paper in which, by comparing the force of his critique with that of attempted refutations of the hiddenness argument already in print, he suggests otherwise.

Second, let me clarify something in WD to which a number of readers, including McCreary, appear to have given insufficient attention: namely, that my principal goal in Part III, where theism is first addressed, is not to prove the truth of atheism but to show theists that they are not immune from the scepticism defended in Parts I and II (see WD, 191–193). Arguments can have various uses, and my main goal for the seven atheistic arguments developed in Part III was that they should (in a variety of ways, developed in chapter 14) show that any attempt to use theistic propositions to infer the truth of ultimism and thus escape a general religious scepticism would be seriously premature. Showing by means of those arguments that atheism is true is sufficient for this result, but it is not necessary: producing conditions in which scepticism about theism properly would flourish is sufficient too. It follows that, even if McCreary is right in his inconsistency claim and I ought to be sceptical about the hiddenness argument, the project of WD is not undermined.

But he is not right. Let's first notice how his reasoning fails to show that he is. Pointing to the fact that I have, in chapter 1 of WD, identified four marks – precision, detail, profundity, and attractiveness – that may be discernible in a proposition, and that are (in certain conditions) sufficient to justify scepticism about whether the available evidence bearing on a proposition is representative, McCreary pounces: are these marks not visible in a proposition I must believe as I develop my hiddenness arguments? Here is the proposition McCreary has in mind:

This proposition is supposed to be both properly attributable to me and so full of precise and detailed and profound and attractive content as clearly to flout the sceptical imperatives I myself have laid down. But unfortunately, McCreary here only succeeds in exemplifying some of the evidence – human tendencies toward oversight (or neglect) – supporting the mode of religious scepticism he is attempting to apply!

Notice first that this proposition is no more than an example of what in introductory logic classes is sometimes called the straw-person fallacy: McCreary represents me as having made an inference that is different from, and more easily queried, than the one I actually have made. Consider his ‘because’ – which I have highlighted – and what comes after. What should be there is this (or its equivalent): ‘because if a God of perfect love exists, only their resistance to God could prevent capable individuals from being in a position to enter into explicit relationship with God’. From this (in conjunction with an uncontroversially true observation about belief's role in being in such a position, as I have defined the latter), the content coming immediately before the ‘because’ – that if a perfectly loving God exists, only their resistance to God could prevent capable individuals from believing that God exists – clearly deductively follows.

Earlier in his paper (209) McCreary himself shows that, according to my reasoning, the proposition about God's opposition to non-resistant non-belief deductively follows from the content here mentioned, not directly or indirectly from any claim about our greatest good, which – as I have often allowed and even argued – would provide much more tenuous support. And yet here, where it is necessary to have an inference that is more questionable, we are diverted onto a different track! (I am not saying that the diversion is conscious or intentional.)

So there's nothing about McCreary's ‘because’ to disturb the careful interpreter, for the deductive inference I have actually made in the hiddenness argument is a different one, and about as secure as they come. But what about the basis for that inference – in particular, the claim that if a God of perfect love exists, only their resistance to God could prevent capable individuals from being in a position to enter into explicit relationship with God? Might McCreary still reasonably complain about detail or profundity here?

This is where the qualification I built into my sceptical argument – signalled above by the parenthetical expression ‘in certain conditions’ – comes into play. McCreary doesn't tell you about this all-important qualification until long after his introduction of the four scepticism-related properties; he suggests indeed (210–211) that I take the four properties to signal a reason for scepticism wherever they are found. (Again I do not assume that this was conscious or intentional.) But in chapter 1 of WD I explicitly deny that in just any case where these properties are present, scepticism must be the result. There can be ‘clear cases’ where the support for believing a proposition is strong enough to assure us of representativeness in the evidence. It is after making this point that I assemble my sufficient condition of reasonable doubt as to representativeness. When I state that condition in terms of the four properties, I accordingly do so in such a way as to leave aside clear cases: ‘I want now to develop certain points previously adumbrated into a sufficient condition for the justification of doubt with respect to the representativeness of evidence in certain nonclear cases’ (WD, 34, my emphasis).

Though McCreary's exposition initially misleads on this matter, he should therefore immediately have asked – assuming he had in hand the proposition I do in fact endorse, about the connection between divine love and openness to explicit relationship – whether the relevant proposition can reasonably be advanced as a clear case. If so, the four-properties argument, as developed in WD, won't touch it.

And not just that. In chapter 1 of WD, after making the point about the immunity from scepticism of ‘clear cases’, I add this, in a footnote (33): ‘More general considerations about what might guide us in judging certain beliefs to be justified, whether thus clearly and evidently or otherwise, in opposition to a completely global scepticism, are introduced and discussed in Chapter 8’ (my emphasis). So what are these considerations, which, even if he misses this footnote, the reader will encounter in chapter 8? In a nutshell: considerations concerning such belief-forming and belief-correcting practices ‘as those we call sensory, introspective, memorial, and (rationally) intuitive’ (WD, 170). The point in chapter 8 is that these practices are ‘universal and unavoidable’, and that it is therefore ‘not imaginable that we should make a beginning in investigation, even an investigation that eventually takes us beyond some of their deliverances, except by means of them’ (WD, 170, 171). And this – though I won't fill out the story here – gives us ‘an answer to a completely general scepticism and a good reason to make use of our basic and unavoidable belief-forming and revising mechanisms’ (WD, 172).

So what McCreary really has to ask, to avoid jumping to tempting conclusions with his inconsistency claim, is whether my claim about the connection between divine love and openness to explicit relationship might be advanced by me as a clear case or is at any rate the output, for those, like me, who advance the hiddenness argument, of one of those unavoidable belief-forming practices. And by now it will be easy to see that the answer is yes.

Consider the first possibility. Later on in his paper (214–215) McCreary entertains the notion that consistency in my reasoning might be preserved by the ‘clear-case’ argument, and he concludes that it cannot be, because the proposition he thinks I am committed to is not obviously true. But having seen that the proposition he says I am committed to is not correctly thus construed, we can see that his reasoning here fails. And when we look at what I am committed to, namely, the proposition about a conceptual connection between divine love and openness to explicit relationship, we see precisely a proposition that I am advancing as a clear case. Though it has not yet won wide acceptance as a conceptual truth – perhaps in part because of the operation of cultural factors discussed in WD (203–204), which McCreary does not consider applying to himself – anyone familiar with the literature will know that I have repeatedly argued that the proposition we are dealing with here is a conceptual truth about divine love and should be regarded as such, just as, say, the proposition that a perfectly good God would resist unjustified evil is regarded by us as a conceptual truth.

What does McCreary say about all this? Referring precisely to my repeated arguments and detailed defences of my views, he says the following: ‘All of the work that went into constructing and defending his critical premises demonstrates that they are not “simple beliefs of reason” or “clear truths of reason”’ (221). There is a failure of comprehension here. What McCreary fails to notice is that where I have given many arguments in connection with the same point, these are not always arguments directly for that point, on which it depends for its credibility. Precisely because of the resistance many in contemporary philosophy of religion apparently feel toward non-traditional pictures of the divine – especially where these lead to currently unloved views like atheism – many of my arguments are designed to prepare the way for conceptual insight; and in other cases justification is purposefully overdetermined, so that it will be harder to deny the view in question. How ironic, then, to find these arguments treated as evidence that the point I am trying to expose is not a conceptual point!

Though McCreary does at least entertain the ‘clear-case’ argument I have been considering here, he overlooks or neglects altogether the point about unavoidable belief-forming practices, which alone refutes his inconsistency claim. For someone like me who believes that a God of perfect love is always open to explicit relationship, this belief is the output of one of those practices – again, it presents itself as true to rational intuition. And, as such, it may consistently be held together with a more general religious scepticism by one whose religious scepticism is developed in the context of an acceptance, for the sake of inquiry, of just such belief-forming practices. (This point is already made in WD, 298–299).

So McCreary's inconsistency reasoning is, for a variety of clear reasons, unsuccessful. In conclusion, let me show, more positively, just how one can be an atheist and a sceptic, too, which may work to prevent future misunderstanding and – this is the important thing – promote fruitful discussion of this neglected joint option. The key is to notice the stark contrast between traditional theism's idea of a personal God and the general religious proposition I call ultimism – the claim that what is ultimate in reality is also of unsurpassably great value and the source of an ultimate good for humanity and the world. The religious scepticism I have defended concerns the latter claim, not the former.

Now, because for many of us traditional theism fills our whole field of vision, it can appear that there is little to choose between these two ideas: the ultimate just is the personal God. But we need more assiduously to cultivate a recognition of how things really are here. If we do, we may see not only that various actual versions of ultimism other than theism already exist, but that in our possibly millions-of-years-long future, many others may be revealed. Instead of ultimism being reducible to one of these claims, it is equivalent to their disjunction; and given how large and open-ended that disjunction must appear to hominids like us at an early stage of evolution, it shouldn't be hard to understand why someone might remain in doubt about the disjunction, perhaps on grounds of possibly unrepresentative evidence, even while disbelieving this or that disjunct.

In other words, ultimism is one thing, but specific versions of ultimism are another. Now consider theism, one of those versions (or disjuncts). Precisely because of the familiarity of the idea of a person, there can be specific awarenesses about theism's idea of the ultimate-as-person, accessible to us now and usable in argument against the existence of such a being, even if we are not yet in a position to rule out (or in) what is referred to by the more general notion of ultimism. Atheistic arguments such as the hiddenness argument – especially if restricted to seeming conceptual truths and deductive processes of reasoning, which are among the items we have to work with in order to make a beginning in intellectual or spiritual investigation at all – may therefore consistently be accepted even where the reach of arguments seeking to disestablish ultimism is judged to exceed their grasp.

Indeed, recognizing that such religious investigation as humans have so far been able to muster may be occurring at a very early stage in the total investigative life of the species, one might very well use such arguments as the hiddenness argument and the argument from horrors to suggest that we ought to get past the stage of our spiritual evolution in which we were so completely preoccupied with the familiar idea of a personal God. I have recently been making precisely this proposal. Of course it is a proposal, meant to be debated. But since one can in this way use atheism to clear a path for non-theistic religious ideas, in the general context of a larger scepticism about ultimism, it is evident that atheism is compatible with religious scepticism.