THE EPISTEMOLOGY OF MODEST ATHEISM

J. L. SCHELLENBERG
Mount Saint Vincent University

Abstract. Distinguishing between the old atheism, the new atheism, and modest
atheism, and also between belief and acceptance, and belief and acceptance
tokens and types, I defend the disjunctive view that either modest atheistic
belief or modest atheistic acceptance, construed as type, is today epistemically
justified in the context of philosophical inquiry. Central to my defence is
a deductive version of the hiddenness argument and an emphasis on the early
stage of philosophical inquiry that we presently occupy.

I begin with distinctions between what I shall call the old atheism, the
new atheism, and modest atheism. The old atheism, exemplified by J. L.
Mackie in his book The Miracle of Theism (1982) and by hundreds of others,
especially from the Enlightenment on, has the following three features.
It is narrowly personalist (that is, concerned only with a conception of
God as person, or something like a person); it is commonly supported
by philosophical arguments; and it is purely negative (that is to say,
restricted to denying the existence of a personal God).
The new atheism, exemplified by Richard Dawkins in his book The
God Delusion (2006) and by at least three others (though many more are
cheering the ‘four horsemen’ on), lacks each of these features. For it is quite
generally opposed to the idea of transcendent or supernatural realities
and thus has a broader than personalist focus; it tends, moreover, to
depend on appeals, implicit or explicit, to a certain positive metaphysics,
namely, scientific naturalism (hereafter: naturalism); and because of the
previous point, it cannot be said to restrict itself to a purely negative
claim.
In my view, there are a number of things wrong with the new atheism,
and its epistemological approach – to the extent that it has one – is flawed.
Ideology-infused bellicosity too often takes the place of careful reasoning

EUROPEAN JOURNAL FOR PHILOSOPHY OF RELIGION 7/1 (SPRING 2015), PP. 51-69
52 J. L. SCHELLENBERG

here. If atheism depends on such a flawed epistemology, then atheism

should be rejected by philosophers. Happily (or unhappily, depending
on your view), atheism does not depend on it, and so we cannot turn
that conditional into a sound instance of modus ponens. Most obviously,
there is also the less flamboyant but more formidable old atheism, whose
arguments against theism and in support of its own negative claim have,
over the past few centuries, been something of a thorn in the side for
theistic philosophers.
When I was young, I identified with the old atheism, and sought to
enlarge its store of arguments with new philosophical arguments from
hiddenness, horrors, and free will (yes, there is a free will offence as well
as a free will defence). But in the last decade or so I have moved to a more
nuanced position which I shall here call modest atheism.
At first glance modest atheism may seem somewhat less than modest,
for it does not reject any of the three features distinguishing old atheism,
instead adding to them. (It might therefore also be called old atheism
plus.) But what it adds makes for an overall stance that includes a modest,
even sceptical strain. Modest atheism supposes to be false a certain
precise affirmative proposition about the existence of an ultimate divine
reality influential in both western philosophy and western religion – that
the divine exists as person and actor – while regarding it as epistemically
possible (by which I mean ‘not justifiedly deniable’) that some other
affirmative proposition about the existence of a religious ultimate,
perhaps one unknown or even unknowable to us today, should one
day prove to be true. While it closes the book on personal theism, it is
open – and explicitly open – to the discovery of other forms of divine
reality. And it is so (quite ironically given the preoccupations of the new
atheism, whose opposition in the name of science to all things religious
we have already noted) at least partly in light of what science teaches us
about our place in evolutionary time, a place which, when we make the
appropriate transition from human to scientific timescales, we will see
to be at the very beginning of intelligent inquiry on our planet, which
our self-important species has grown accustomed to treating as though
it were the end.
I take it that the old atheism is not thus open. Indeed, here we hit
on an assumption apparently shared by the old and the new atheists.
This assumption is that there is truth in religion only if something like
personal theism is true. If this assumption is not being made by the old
atheists, then how shall we explain the fact that, after reaching atheism,
 THE EPISTEMOLOGY OF MODEST ATHEISM 53

they venture no further in religious investigation? Most old atheists,

just like new atheists, are naturalists, even if they don’t crudely conflate
atheism with naturalism at the conceptual level. And if this assumption
is not being made by the new atheists, then how shall we explain the fact
that they do tend to conflate atheism with naturalism?
Modest atheism, therefore, is unlike each of the other atheisms in
its openness to nontheistic religious discoveries, perhaps ones occurring
only in the far future. It does not rule them out. That is to say, it is more
modest.
In this essay I want to defend the claim that modest atheism is in
good shape, epistemologically. But to prepare the way for this defence
some more distinctions are needed. We need to distinguish between
atheism (of any kind) as proposition, as belief, and as acceptance, and
then also between belief and acceptance tokens and types.
Much of what I have said so far could be understood on the assumption
that atheism is a proposition or claim of some kind – in the case of modest
atheism, the proposition that no personal God exists but some other
depiction of the divine may someday prove correct. In philosophical
discussions of whether atheism is true this propositional interpretation is
clearly being applied. For only propositions are literally true or false. But
we might also – and sometimes do – ask whether so-and-so’s atheism is
well grounded or justified, or speak (as I did earlier) of a person’s state of
mind as exemplifying atheism. And here we are usually instead thinking
about a certain individual’s belief that an atheistic proposition is true. But
there is yet a third possibility, which tends to be overlooked in philosophy
today but will become more salient as the distinction in epistemology
between belief and acceptance is further clarified and utilized – a process
which, perhaps optimistically, I think is well underway. This is that the
atheism of a philosopher such as myself, or of any person, may amount
to an acceptance of the relevant proposition rather than belief of it. In
close but not quite complete conformity with what L. Jonathan Cohen
says about that distinction in his excellent book on the subject (1992),
I suggest that the term ‘acceptance’ is most helpfully used to name what
is described when we speak of in a fully voluntary manner forming
and maintaining a policy of treating a proposition as true, using it as
a basis for inference. A corollary is that the term ‘acceptance’ ought
to be distinguished from ‘belief,’ which rather names a less than fully
voluntary disposition (or set of dispositions) such as the involuntary
disposition Cohen himself identifies with belief: namely, the disposition
54 J. L. SCHELLENBERG

to feel a proposition true in relevant circumstances. It seems clear that

sometimes the ‘position’ of a philosopher on this issue or that should
be understood in terms of that philosopher’s accepting a certain
proposition rather than in terms of belief. And perhaps at an early stage
of investigation, of the sort I have said we will see ourselves to be in
when we fully absorb scientific timescales, there will often be occasion
for acceptance even if not for belief.
Suppose, then, that we have seen the differences between atheism
as proposition, as belief, and as acceptance. Suppose also that we have
noticed that the conditions of belief ’s justification might be different
from those attaching to the justification of acceptance (more on this
later), and accordingly that the epistemology of atheism as belief might
be different from the epistemology of atheism as acceptance. There is
still – and finally – the distinction between belief and acceptance tokens
and types to take note of.
This is really a distinction between different senses of the expressions
‘belief that p’ and ‘acceptance that p’. Sometimes it is a certain way of
believing or accepting, the belief or acceptance that p, that we have in
mind when we use such an expression, and to use it correctly we need
not presuppose that this belief or acceptance is realized in anyone
(even if its appropriateness to this or that mental or social context is
discussed); but in another sense what we may have in mind is his or her
belief or acceptance that p, and in evaluating the belief or acceptance
thus understood we evaluate the person who exemplifies it by way of
assessing their relevant dispositions (the dispositions involved in their
coming to, or not ceasing to, include in their mental repertoire the belief
or acceptance in question). In the former abstract case what we have
is a belief or acceptance type; in the latter concrete case it is a belief or
acceptance token.
In considering the justification of a belief or acceptance type in
connection with the existence of God what we are looking for is
a worthiness of instantiation that abstract discussion of whether belief
or acceptance is best among available responses (either the best or
a best, and either way such as cannot be exceeded) will help us discern;
such discussion, in my view, is the task of philosophers, and I shall be
engaging in it here. Whereas in evaluating belief or acceptance tokens
the relevant desideratum is what we may call responsibility, which
amounts to something like the proper fulfilment of all relevant duties
and the exercise of intellectual virtue in the formation and maintenance
 THE EPISTEMOLOGY OF MODEST ATHEISM 55

of belief or acceptance by the relevant believer or accepter. There is this

connection between the two levels of evaluation that may be noted: if
one declares a certain response type to be unjustified within a certain
context and so unworthy of being instantiated, then one will also think
that, other things being equal, investigation of the most responsible
and virtuous sort will in that context lead to such a response type not
being instantiated by the investigator, and so the investigator will, in the
token sense, not be justified in exhibiting that response. But there are
obviously many variables that can prevent facts about type and token
justifications from matching up here, such as persistent controversy over
type justification, the amounts of information particular investigators
have about the results of formal inquiry, and so on.
So when I say that I mean to defend the thesis that modest atheism is
in good shape, epistemologically, do I have in mind atheism as belief or
acceptance, and will it be belief or acceptance as token or type? What I’ve
just said about the task of philosophers leads me to a focus on response
types. And the distinction between belief and acceptance affords the
defence of modest atheism some flexibility here, which I intend to make
use of. The view I shall defend is accordingly the following: that either
modest atheistic belief or modest atheistic acceptance, construed as type,
is today justified (i.e., worthy of being instantiated) within the context of
philosophical inquiry about things religious. Since this is the view I mean
to defend, it may from here on be assumed that when I speak of atheistic
belief or acceptance I am speaking of a certain type of stance. I myself think
modest atheistic belief is justified, but I shall argue only for the weaker
disjunctive claim. And the final result I’m aiming at is indeed a successful
defence – in a short paper I cannot hope to establish my conclusion but
I do intend to advance its cause, to put it in a more favourable light.
Even so, I have my work cut out for me. How will the defence proceed?
I want now to suggest that here too modest atheism can do something
to earn its name, for it is also in a way modest in the sort of reasoning
it uses to support itself. At first, you will want to say, once again, that
what I count as modesty here is really immodesty! For the reasoning
I have in mind is deductive reasoning – and haven’t we long since come
to appreciate that deductive arguments for the nonexistence of God are
hopelessly over-optimistic? Hasn’t atheistic philosophy of religion in the
past few decades come to focus, much more modestly and appropriately,
on inductive reasoning, such as the probabilistic reasoning of William
Rowe or Paul Draper (Howard-Snyder, ed. 1996)?
56 J. L. SCHELLENBERG

My reason for focusing on deductive arguments, and daring to call

what I am doing modest, is in part bound up with what is required to justify
belief on big issues such as those of philosophy at an evolutionarily early
stage of inquiry such as our own. I would not say that no philosophical
beliefs at all can presently be justified; that way an inquiry-stultifying
scepticism looms. But I do say that a justification for philosophical
belief is, at an early stage of inquiry, much easier to provide where
compelling grounds, such as those embodied by an apparently sound
deductive argument, are available. In the absence of compelling grounds,
in particular where probabilistic arguments alone are given, I think
inquirers should be much more reluctant to claim justification for
philosophical belief as opposed to, say, an increase in the probability that
a certain philosophical claim is true. My modesty consists, in part, in this
reluctance.
But my immodesty, you may now say, consists in my suggestion that
the high bar I have set up can be reached in the case of atheism! Well, is
it any more modest to assume that no other good deductive arguments
against the existence of God will be discovered than have already been
discovered at a relatively early point in the evolution of intelligence, say,
1982? This, it appears, is what any pre-emptive argument of the sort
suggested by the critic must assume. In any case, here it is important
to recall my disjunctive approach, which will be satisfied even if only
acceptance of the proposition put forward by the modest atheist can in
some way be justified. Deductive arguments, as we will see, may have
a role to play in relation to acceptance too.
So which deductive argument(s) for the nonexistence of God will
I advance? Well, when I saw the line-up of topics and speakers gracing
the conference on the epistemology of atheism for which this paper
was written, I noticed that the hiddenness argument was to receive
considerable discussion. Since – as I note perhaps without modesty but
I think truthfully – I was responsible for getting that way of reasoning
into discussion a couple of decades ago (Schellenberg 1993), I decided
that I might as well jump on the bandwagon and convey how I view the
hiddenness argument at present, using this to develop my case. But there
is another reason for focusing on the hiddenness argument here. Since it
is a relatively new argument, driven in part by secularization processes
that have had some considerable influence in the last few centuries, it fits
nicely with my emphasis on how we are still very much at the beginning
 THE EPISTEMOLOGY OF MODEST ATHEISM 57

of things where religion and its discussion are concerned and nowhere
near the end.
But what matters here is whether the hiddenness argument is a good
argument, capable of justifying atheistic belief or acceptance. So let’s
have a look. There are various ways of formulating the argument. The
way I have chosen to use is the following:
(1) If no perfectly loving God exists, then God does not exist.
(2) If a perfectly loving God exists, then there exists a God who is
always open to personal relationship with any finite person.
(3) If there exists a God who is always open to personal relationship
with any finite person, then no finite person is ever nonresistantly
in a state of nonbelief in relation to the proposition that God
exists.
(4) If a perfectly loving God exists, then no finite person is ever
nonresistantly in a state of nonbelief in relation to the proposition
that God exists (from 2 & 3).
(5) Some finite persons are or have been nonresistantly in a state of
nonbelief in relation to the proposition that God exists.
(6) No perfectly loving God exists (from 4 & 5).
(7) God does not exist (from 1 & 6).
The phraseology here is a bit loose in places, but if one wants to convey
a clear first impression it helps not to weigh the argument down with
numerous explanatory clauses, and we can tighten things up as we go
along.
The first thing to notice about the argument is that the inferences at
steps (4), (6), and (7) are clearly deductively valid. So we can focus on
whether the premises should win our belief or acceptance.
Shall we go along with premise (1): if no perfectly loving God exists,
then God does not exist? Well, the idea of a person-like God – which as
I’ve noted is the idea of God that the modest atheist, like the old atheist,
is concerned with – represents one way in which the religious idea of
an ultimate reality has been interpreted by human beings. Notice also
that it is only or mainly as a candidate for metaphysical and axiological
ultimacy that God comes to have a place in the discussions of western
philosophy. Now, as one might expect, given that word ‘ultimate’, God is
commonly regarded as having all knowledge and all power – or at least
as much as it makes sense to suppose a person like God could have. For
58 J. L. SCHELLENBERG

the same reason of ultimacy, God is said to be the source of our existence
and perfectly good. But God is also commonly said to be perfectly loving
toward created beings. And this attribute is at least as obviously essential
as the others. For the best love, love of the sort we rightly admire, is one
of the most impressive features any person, man or woman, can display.
Perhaps it is one of the results of recent cultural evolution that we can
now see this more clearly than humans once did. How could a candidate
for ‘greatest possible person’ be anything but a fraud if it weren’t always
possessed of the greatest possible love? Whatever stunning attributes it
displayed, we would then be able to imagine an even greater person, who
was perfectly loving. It seems incumbent, therefore, on everyone who
today reflects on the existence of God to acknowledge that if God exists,
God is perfectly loving. Christians of course have specially emphasized
this attribute but for all theists and atheists in philosophy there is good
reason to do so.
Let’s move on, then, to premise (2): if a perfectly loving God exists,
then there exists a God who is always open to personal relationship with
any finite person. That phrase ‘personal relationship’ should at this point
be tightened up a bit: what I have in mind is a conscious and (positively)
meaningful relationship. I should also acknowledge a point that, once
acknowledged, will remain tacit: namely, that the scope of premise
(2) is restricted to finite persons who are relevantly capable, where the
relevant capacities are cognitive and affective capacities sufficient to be
able at the time in question to be in a meaningful conscious relationship
with God – such things as a capacity then to feel the presence of God,
recognizing it as such; a capacity to exhibit attitudes of trust, gratitude,
and obedience to God, and so on.
Now, some theists might be inclined to resist this premise because
of a prior commitment to a religious scripture or creed incompatible
with it or in tension with it. Isn’t the God of the Bible, for example, often
portrayed as somewhat distant relationally? But none of this can be
relevant here where we are considering what the modest atheist must do
to rise above epistemological suspicion. The modest atheist, who like the
old atheist is working within a philosophical frame of reference, cannot be
limited by theological assumptions which have been formed because of
the need to find room for God in our world. It shouldn’t need to be said,
but in the present circumstances of inquiry in philosophy of religion,
which is filled with believing philosophers, it has to be emphasized
that philosophers cannot assume because of some consensus in their
 THE EPISTEMOLOGY OF MODEST ATHEISM 59

communities that God exists and so has only those qualities compatible
with creating a world like ours, but must seek to be guided by reason
when considering what a God would be like.
So what does careful reasoning, which seeks to be attentive to all that
we humans have learned, tell us? Well, it makes it clear that people who
admirably love you (and thus any who perfectly love you) are invariably
open to a kind of personal relationship with you in which the two of
you can interact meaningfully and consciously with each other. Indeed,
since they love you in this way, they want to be close to you, and close in
a way you can appreciate, so you can turn to them for advice or draw on
their support or just feel them present with you when that’s needed. (Of
course if they admirably love you, they’ll value being with you for its own
sake, too.) Now, it’s true that they won’t force any of this on you, which
is one reason why I only used the word ‘open’ when stating this premise.
There is even room here for a sort of withdrawal within relationship. But
if they aren’t at least open to such relationship, it would be a mistake to
say they admirably love you.
To see this with full clarity, imagine that you’re listening to a friend,
who’s describing his parents: ‘Wow, are they ever great – I wish everyone
could have parents like mine, who are so wonderfully loving! Granted,
they don’t want anything to do with me. They’re never around. Sometimes
I find myself looking for them – once, I have to admit, I even called
out for them when I was sick – but to no avail. Apparently they’re just
not open to a relationship with me right now. But it’s so good that they
love me as much and as beautifully as they do!’ If you heard your friend
talking like this, you’d think he was seriously confused. And you’d be
right. His parents, if your friend’s description of them is correct, could
certainly be lots of other things – even impressive things, like the best
corporate lawyer in the country and the President – but their attitude
toward their son, whatever it is, does not include an admirable love.
I expect you’ll see how all of this can be applied to God. A careful look
at the concept of love should lead us to affirm that God is always open to
personal relationship with each of us (or with each of the beings a God
would or might create, whoever they are), if God exists and is perfectly
loving – which is to say that premise (2) is true.
Now, so far I’ve been treating the concept of ‘openness’ to relationship
as one we all understand, and at an intuitive level I’m sure this is the
case. But as we move on to premise (3) – if there exists a God who is
always open to personal relationship with any finite person, then no
60 J. L. SCHELLENBERG

finite person is ever nonresistantly in a state of nonbelief in relation to

the proposition that God exists – we will need to probe this concept
a bit more carefully. So let’s have a look at that word ‘open’ and how it
behaves, logically speaking. In particular, let’s note a sufficient condition
of someone not being open in the relevant way:
Not Open
If a person A, without having brought about this condition through
resistance of personal relationship with person B, is at some time in
a state of nonbelief in relation to the proposition that B exists, where
B at that time knows this and could ensure that A’s nonbelief is at that
time changed to belief, then it is not the case that B is open at the time
in question to having a personal relationship with A then.
Indeed, in such circumstances B (if B exists) is consciously preventing
such a relationship from existing at that time. And if anything is obvious,
it is that you cannot be open to a relationship in the relevant way while
consciously preventing it! We can apply this to God, who of course
possesses all relevant knowledge and ability: if any finite person is ever
nonresistantly in a state of nonbelief in relation to God’s existence, then
there is no God always open to personal relationship with each finite
person. Another way of putting that point gives us its contrapositive,
which is premise (3): if there exists a God who is always open to personal
relationship with each finite person, then no finite person is ever
nonresistantly in a state of nonbelief in relation to the proposition that
God exists.
So premise (3) seems quite clearly believable too. What about the last
premise that needs to be checked, premise (5): some finite persons – and
of course I have in mind human beings – are or have been nonresistantly
in a state of nonbelief in relation to the proposition that God exists?
Here again theology may tempt some theistic thinkers to suppose that
our modest atheist is stepping outside the bounds of what should be
believed. For might not any one of us be secretly resistant to a holy and
demanding God, blinded to the motives that grip us? If so, then perhaps
those who don’t believe in God are, in a way, hiding from God. Might
the proponent of the hiddenness argument have managed to get things
backward in this way?
Notice first that she needn’t be thinking about herself: perhaps
other nonbelievers strike her as displaying nonresistance by the same
standards that leave her questioning her own. Indeed, how could
 THE EPISTEMOLOGY OF MODEST ATHEISM 61

an investigator help noticing that some people who don’t believe in God
still have an admirable track record of investigation, and emotionally
are, if anything, biased in favour of God? Some people who find that
the evidence of argument and experience has taken belief in God away
in midstream, as it were, in the midst of a strenuous religious life, would
love to believe in God. What reason could someone have to say that they
are resisting a relationship with God? It strains – and indeed breaks –
credulity. The evidence of nonresistance here can pile up in such a way
that an honest inquirer judges it to be stronger than any counterevidence.
Even if in such circumstances one thinks belief is unjustified because
of new and unsuspected evidence that only future inquiry may reveal,
clearly acceptance of a premise like (5) is justified given that the available
evidence strongly supports it.
But even this is a weaker stance than is justified when we consider that
we needn’t stay focused on people who have thought about the existence
of God and so have come within the range of motives for resisting it.
Behind them, as it were, stretching into places far distant from any
affected by Western culture, and also into times long ago, before humans
had so much as conceived of an all knowing, all good and loving creator
of the universe, we find evidence of individuals and communities who,
though capable of possessing it, lacked belief in God, and obviously
without ever having blinded themselves by resisting God in any way.
How could there be resistance in such a case? The critic of (5) needs you
to look away from all this evidence for nonresistant nonbelief. But to do
so would be to fall prey to blindness of another kind.
The four premises of the hiddenness argument therefore seem clearly
true. Since, as we have already seen, its three inferences are clearly valid,
it follows that the argument seems clearly sound. Shall we therefore
pronounce in favour of a belief or acceptance type of response to modest
atheism?
Many philosophers will think that there could still be good reasons
not to do so. For example, it may be said that there are powerful defeating
objections showing one or another of the hiddenness argument’s
premises to be false or undercutting the justification for believing or
accepting some such premise, or that there are equally strong arguments
for theism to be weighed on the other side. Of course there isn’t time here
to examine closely all the reasoning I’ve just alluded to, but I think some
illuminating general comments may still be made.
62 J. L. SCHELLENBERG

Let’s start with the last idea mentioned, that of equally strong
arguments for theism. Even the most respected and spirited defenders
of theism – take Richard Swinburne, for example – would shrink from
a claim of the sort I have made on behalf of atheism: they would deny
that there is a sound deductive proof of the truth of theism. And it is
not hard to see why things should be harder here for theism than for
atheism. If the existence of God requires that there be a person who is all
powerful, all-knowing, perfectly good and loving, as well as the creator
of the universe, then arguments for theism have the task of showing that
all those conditions are present. But an argument for atheism need only
show that one such condition is absent. (It may, for example, endeavour
to show, as does the hiddenness argument, that perfect love is absent.)
And the latter task might be expected to be the easier one. Indeed, it
is notorious that the so-called theistic proofs are quite incapable of
proving the existence of a being with the whole collection of properties
possessed by the theistic God. (Even the ontological argument suffers
from this incapacity, for although it purports to prove the existence of
a greatest possible being, there is nothing in it to imply that a greatest
possible being would be a greatest possible person, with such properties
as knowledge and love.) Atheism does not have an analogous problem,
and so we have the argumentative asymmetry.
What about the alleged defeating objections to premises of the
hiddenness argument (or to our belief or acceptance of them)? I myself
have done a thorough investigation of these objections and have found
them all wanting. Indeed, I have created many new objections, in order
to test the argument – with the same result. Now, it is of course true
that others, especially philosophers who are theists, may disagree with
me about one or another objection. But there is an important point
to be noted here: namely, that such disagreement is frequently not
philosophically grounded. Oftentimes I appear to be met by philosophers
who are operating as theologians rather than as philosophers when they
question the hiddenness argument.
Perhaps the clearest example of what I am talking about here appears
near the end of a paper by the American philosophers Ted Poston and
Trent Dougherty (2007: 196):
In the final analysis Schellenberg’s argument fails because it envisions
God as requiring too much: explicit, highly confident belief at all times.
Fortunately, God is more generous. The Christian tradition attests that
God will accept far less, he will ‘meet us where we are’.
 THE EPISTEMOLOGY OF MODEST ATHEISM 63

Now, at first this may seem a relevant criticism. But listen to it carefully.
When I first did so, I was truly startled: How could it be thought that the
hiddenness argument depicts God as requiring belief of anyone and thus
as ungenerous – with the implication that if the hiddenness argument
were right about God, God would be leaving some – namely nonresistant
nonbelievers – out in the cold, since they are unable to come up with
what is required? After all, what the hiddenness argument clearly says
is that God would generously offer belief – and the explicit relationship
made possible thereby – to all, and so there would be no nonresistant
nonbelievers in a world created by God. But then I saw that there is in
Poston and Dougherty a tendency to assume that God exists and that
whatever is being said about God must apply to the actual world – even
when that comes in the context of an atheistic argument! Why else would
it be supposed, when someone like me claims that God would favour
explicit relationship, that those in the actual world who don’t have what
it takes to participate in such relationship are going to be left out? How
could what God wants be too much unless creatures are unable to deliver
it, and how could they be thought unable to deliver it unless we are
thinking about our world instead of the world the hiddenness argument
says would exist if God existed, in which all who are nonresistant believe?
Sadly, many allegedly philosophical objections to the hiddenness
argument display a tendency similar to the one I claim to have found in
Poston and Dougherty, even if not so brazenly. Within a philosophical
context they can have no weight at all. In a philosophical context, where
we have to let the voice of authority grow dim and think for ourselves
about what a perfect personal being would be like, we may notice points
that undermine the allegedly undermining objections to the hiddenness
argument.
This holds also for a strategy quite popular today, known as ‘sceptical
theism,’ which questions how we could justifiedly rule out the existence
of unknown goods for the sake of which God is hidden. Accepting this
move in the absence of some special theological bias or preconception
seems to require forgetting what theism has got us talking about in the
first place – an ultimate person. Consider by way of analogy a single man
who marries and has children: Does this behaviour not rightly constrain
the goods he is willing to pursue, at least insofar as he is a loving husband
and father? Though when he was on his own he spent time with many
female friends and was otherwise preoccupied with his own wide-
ranging pursuits, travelling to far-flung regions of the earth for months
64 J. L. SCHELLENBERG

at a time, shifting from place to place and from one activity to another,
now things are different – and quite naturally and rightly so. Now he has
a family to help provide for, to support in emotional and financial ways.
He can’t just take off for Greece or France for long periods at a time to
indulge his own interests. Better, he has new interests which lead him
happily to say no when invitations to do such things arise. Similarly with
God, if God is to be regarded as a loving person – an ultimately loving
person – who has created vulnerable finite persons to be the object of
Divine love. The ‘God’ described by sceptical theists who may, for all we
know, have purposes quite unrelated to us that require hiddenness from
us is not an ultimately loving being at all. If construed personally, such
a God is comparable to a limited or delinquent father or mother who
simply can’t or won’t live up to the demands taken on board when the
commitments of marriage and family are entered into.
I suggest, therefore, that on the basis of such considerations as
I have briefly aired a great deal can be done to warrant, in the context
of philosophical inquiry, setting aside our two counter-suggestions –
concerning equally strong arguments for theism and crippling objections
to the hiddenness argument – without entering into many details of the
associated reasoning.
But here’s another counter-suggestion. Perhaps it will be suggested
that there is also non-propositional experiential evidence to be considered
here. Might not people who find themselves in the grip of suitably
powerful experiences apparently of God have grounds for resisting the
hiddenness argument – perhaps for saying that something is wrong with
it, even if they know not what and though they lack any reasoning to offer
against it? Recently philosophers of religion have been much concerned
with questions of this sort, often defending an affirmative answer
(Swinburne 2004, Alston 1991, Plantinga 2000). But the most that could
conceivably be shown by this means is that theistic religious experience
brings a non-atheistic response to the question of God’s existence up
to a level of worthiness for those inquirers who find themselves in the
relevant experiential circumstances. It could not be shown that it brings
an atheistic response down to a level of unworthiness for those who lack
such non-propositional evidence.
Obviously there is no space here for a proper discussion of the
epistemology of religious experience. But again some general comments
suggest themselves which show that what I’ve found conceivable here is
not actually to be expected, given the facts on the ground. For example,
 THE EPISTEMOLOGY OF MODEST ATHEISM 65

experiences apparently of God, to do the epistemic work required of them

here, would have to be more forceful and also more discriminating than
religious experiences often are. By ‘discriminating’ I mean they would
need to clearly have theistic as opposed to any other religious content.
Now when we have perceptual experiences of other human persons
our experiences commonly are discriminating in the relevant way: I see
from the phenomenological details of my experience that it is John Greco
before me and not Paul Draper or Roger Pouivet. Religious perceptual
experiences are often much more fluid and malleable. It will, I suspect,
be much easier in many cases to get someone to back down from the
claim that the omni-God of traditional personalist theism was present to
her to the claim that something powerfully transcendent was present to
her than it will be to get me to back down from the claim that I saw John
Greco to the claim that some human being or other stood before me. And
if their degree of modesty about such things is tailored to our possible
evolutionary immaturity, which here as elsewhere we are called to take
into account, I think even philosophical inquirers in the grip of religious
experience may accordingly often find its epistemic force less obvious
and relevant than would be required to support the judgment we are
considering. And we have not yet said anything about the problem,
which arises for those who reject the argument I’ve just given, of religious
experiential diversity.
So without too much discussion of details we can see that alleged
proofs of theism, objections to the hiddenness argument, and suggestions
about theistic religious experience may not gain much traction among
those who earnestly and as philosophers investigate the question whether
there is a God and, in that context, wonder what force the hiddenness
argument should be regarded as having.
So what exactly am I proposing – that a belief response to modest
atheism is justified for philosophers thinking about the existence of God,
or that acceptance is? I will provide some more defence for each of the
disjuncts of the disjunction suggested here in turn, hoping to impress
each relevant investigator with at least one of my arguments and thus
successfully to defend the disjunction in relation to everyone.
Let’s start with belief. Each of the premises and also each of the
inference claims of the hiddenness argument can be made to appear
worthy of belief, and the counter-suggestions we have considered seem
not obviously capable of diminishing this justification for the belief that
the theistic God does not exist. Now, of course, we have been unable
66 J. L. SCHELLENBERG

here to examine many details of those counter-arguments; but nor

have we been able to examine other important deductive arguments for
atheism, which, as I have maintained elsewhere (Schellenberg 2007), can
be combined with the hiddenness argument to produce an even more
forceful case for atheistic belief. A definitive outcome is, for these reasons,
not in the cards today but I still conclude that atheistic belief in the part
of philosophical inquiry concerning God has been made defensible,
or more defensible, by my arguments in this paper – especially since it
is a modest atheism that I have in mind, open to the idea that other
conceptions of the divine demand inquiry.
But precisely this modesty, and its rootedness in scientific facts about
our place in time, an objector may now wish to query more closely in
an attempt to overturn my conclusion about atheistic belief. Are we not
‘in over our heads’ when we reach a belief about the existence of God,
given the vast diversity of arguments from perhaps better equipped
future inquirers that we are in the nature of the case unable to sample?
Elsewhere, I have defended such reasoning in relation to the broader
idea that there is no ultimate divine reality (Schellenberg 2007). Why
isn’t it equally applicable to the narrower but equally profound claim that
there is no personal God?
Well, modest atheism, let us remember, denies only the existence of
a person-like ultimate: an omnipotent, omniscient, perfectly good and
loving creator of the universe. It makes the negative claim that there is
no such divine being. This is not nearly so ambitious or profound a claim
as the positive claim that there is such a being, or even as the negative
claim that there is no religious reality of any kind, for it has many fewer
metaphysical consequences. Think of how thorough a story of the overall
nature of things you could tell, knowing that there is a God! But if all
you know is that there isn’t a God, you’ve just ruled out one way things
could be. Indeed, you’ve only ruled out one religious way things could
be; many other religious ways things could be, with similar metaphysical
implications, remain. And so there’s no justification, given only modest
atheism, for an endorsement of such profound metaphysical claims
as that of naturalism – though many immodest theists are mistakenly
inclined to see the latter as following from atheism.
It should also be noted that by seeking only to refute traditional
theism, we remain ‘close to home’ and need to mobilize no more than
certain concepts and considerations we already possess. For the basic idea
of a personal God, as traditionally understood, extrapolates from certain
 THE EPISTEMOLOGY OF MODEST ATHEISM 67

basic facts about ourselves – our limited power, knowledge, goodness,

love – and thus from human qualities we do already know something
about even at the present stage of our development. All my claims in this
paper about how such a God could be revealed to us are unaffected by
the awareness that many other conceptions of the divine remain to be
explored and may indeed be outside the range of our current powers of
conceptualization.
Here’s another thought worth considering in this context. (It’s related
to the last one in that it simplifies the atheist’s job even further.) Some
of my arguments can make use of insights that draw on recent findings,
for example in psychology and feminist thought, where we find a natural
connection between admirable love and commitment to relationship.
Thus their claims need to be considered as contenders for the status of
propositions quite ‘clear’ in themselves but only now becoming clear
to us: that is, as representing the forward edge of new and positive
evolutionary developments. If this can be shown for atheism but not
for theism, then once more we see how arguments justifying modest
atheistic belief may be available even given only our present resources,
though the arguments for theism fall short.
Having said all that in defence of modest atheistic belief in
philosophy, I think another interesting argument that can be made,
if that defence fails, is for atheism’s acceptability. Of course we would
expect philosophers who believe atheistically to also act upon this belief
in inquiry, but arguments for acceptance are arguments for something
like acting on the modest atheistic claim even when you don’t yet believe
it. And here, as Cohen makes clear (1992), pragmatic considerations
concerning the needs of inquiry may be importantly relevant.
So consider these facts. (1) Inquiry about religion in western
philosophy has been going on for thousands of years, and for most of
that time has almost obsessively focused on theism and things theistic,
hardly ever venturing into the potentially vast regions beyond. (2) The
latest report (Chalmers and Bourget 2013) has it that 73% of philosophers
today favour atheism. Now, the figure would surely be different if we
restricted our concern to the opinions of so-called philosophers of
religion, who are predominantly believing theists. But while it might be
argued that this is deserving of notice on the grounds that philosophers
of religion are the experts on religion in philosophy, we would need to
set against this point the fact that most of these so-called philosophers of
religion, again, have not ventured beyond theism in their investigations,
68 J. L. SCHELLENBERG

and also the fact that (3) many of them are deeply motivated by loyalty
to their religious communities, and so should perhaps be regarded as
doing theology – even if philosophical theology – rather than philosophy
(Draper & Nichols 2013, Schellenberg 2009). Finally, we need to note –
as, in part, a consequence of points already made – that (4) acceptance
of atheism does not in any way imply (as those suppose who erroneously
think one must accept either theism or naturalism) that we are closing
the door to the truth of religious claims. Indeed, we are opening it more
widely than has ever been done before!
What should a philosopher say who seeks to be sensitive to all these
facts – while sensitive also to our temporal position and unwilling to
endorse atheistic belief – and who notices that the latest arguments for
atheism are as apparently forceful as the hiddenness argument? I think
she should favour the acceptance of atheism.
Now it can be difficult to achieve a proper balance: When do you
accept a conclusion and when do you say we should wait for more
evidence? Many philosophers today would say that we are rushing things
if we accept that theism is false. I would suggest 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 the idea of
an ultimate divine reality is realized – and also the inference to atheism –
to be made. And I say we should get on with exploring other fillings for
the idea of a divine reality, leaving open the possibility that the latter is
true and so neither believing nor accepting that it is false. I have named
the more general proposition here, the proposition more general than
theism, ‘ultimism’. The idea is that even at this early stage of religious
investigation we should draw conclusions where we can, to help keep
inquiry moving, while being very careful not to foreclose inquiry where
we shouldn’t. The distinction I have suggested between the epistemic
status of ultimism, which says only that there is a metaphysically,
axiologically, and soteriologically ultimate reality of some kind, and that
of its personalist elaboration seems to me to get this balance about right
and to respond appropriately to the needs of inquiry concerning religion
in philosophy. But if so, then even if modest atheistic belief is thought to
be unjustified in the precincts of philosophy, atheistic acceptance can still
be justified.
A concluding summary, then, might run as follows. Epistemological
considerations can be seen to favour an atheistic response in philosophy
to questions about God’s existence when we consider our subject
 THE EPISTEMOLOGY OF MODEST ATHEISM 69

carefully, distinguishing between belief and acceptance, and in a context

governed by scientific timescales. The sort of atheism that is thus justified
is a modest atheism. It claims to have extinguished the light from, at most,
one of the many facets of the concept of a Divine reality. And it makes
this claim only after discovering arguments of seemingly compelling
force. These arguments can be used to defend modest atheistic belief as
the preferable response to questions about a personal God in the context
of philosophical inquiry. And even if this conclusion were to be left
unsecured, perhaps because of the demands of deep time scepticism,
there would still be reason to take such arguments as justifying the
acceptance of modest atheism at the present stage of religious inquiry.
It seems, therefore, that a modern and modest atheism can acquit itself
admirably at the bar of epistemological reason.

BIBLIOGRAPHY
Alston, William P. 1991. Perceiving God: The Epistemology of Religious Experience
(Ithaca: Cornell University Press)
Cohen, L. Jonathan. 2006. An Essay on Belief and Acceptance (Oxford: Clarendon
Press)
Dawkins, Richard. 2006. The God Delusion (New York: Houghton Mifflin)
Draper, Paul & Ryan Nichols. 2013. ‘Diagnosing Bias in Philosophy of Religion’,
The Monist, 96: 422-448
Howard-Snyder, Daniel, ed. 1996. The Evidential Argument from Evil
(Indianapolis: Indiana University Press)
Mackie, J. L. 1982. The Miracle of Theism (Oxford: Clarendon Press)
Plantinga, Alvin. 2000. Warranted Christian Belief (Oxford: Oxford University
Press)
Poston, Ted & Trent Dougherty. 2007. ‘Divine Hiddenness and the Nature of
Belief ’, Religious Studies, 43 : 183-198
Schellenberg, J. L. 1993. Divine Hiddenness and Human Reason (Ithaca: Cornell
University Press)
Schellenberg, J. L. 2007. The Wisdom to Doubt: A Justification of Religious
Skepticism (Ithaca: Cornell University Press)
Schellenberg, J. L. 2009. ‘Philosophy of Religion: A State of the Subject Report’,
Toronto Journal of Theology, 25: 95-110
Swinburne, Richard. 2004. The Existence of God, 2nd. ed. (Oxford: Oxford
University Press)