University of Virginia Library

AGNOSTICISM

I

Agnosticism is a philosophical and theological concept
which has been understood in various ways by different
philosophers and theologians. T. H. Huxley coined the
term in 1869, and its first home was in the disputes
about science and religion, naturalism and super-
naturalism, that reached a climax during the nineteenth
century. To be an agnostic is to hold that nothing can
be known or at least that it is very unlikely that any-
thing will be known or soundly believed concerning
whether God or any transcendent reality or state exists.

It is very natural for certain people conditioned in
certain ways to believe that there must be some power
“behind,” “beyond,” or “underlying” the universe
which is responsible for its order and all the incredible
features that are observed and studied by the sciences
even though these same people will readily grant that
we do not know that there is such a power or have
good grounds for believing that there is such a power.
While the admission of ignorance concerning things
divine is usually made by someone outside the circle
of faith, it can and indeed has been made by fideistic
Jews and Christians as well.

Some writers, e.g., Robert Flint and James Ward,
so construed “agnosticism” that (1) it was identified
with “philosophical skepticism” and (2) it allowed for
there being “theistic agnostics” and “Christian agnos-
tics.” However, the more typical employment of “ag-
nosticism” is such that it would not be correct to count
as agnostics either fideistic believers or Jews and Chris-
tians who claim that we can only gain knowledge of
God through some mystical awareness or “ineffable
knowledge.” It surely was this standard but more cir-
cumscribed sense of “agnosticism” that William James
had in mind when he made his famous remark in his
essay “The Will to Believe” that agnosticism was the
worst thing that “ever came out of the philosopher's
workshop.” Without implying or suggesting any sup-
port at all for James's value judgment, we shall construe
agnosticism in this rather more typical manner. Given
this construal (1) “theistic agnosticism” is a contra-
diction and thus one cannot be a Jew or a Christian
and be an agnostic and (2) also agnosticism is neutral
vis-à-vis the claim that there can be no philosophical
knowledge or even scientific or common-sense knowl-
edge. We shall then take agnosticism to be the more
limited claim that we either do not or cannot know
that God or any other transcendent reality or state
exists and thus we should suspend judgment concerning
the assertion that God exists. That is to say, the agnostic
neither affirms nor denies it. This, as should be evident
from the above characterization, can take further
specification and indeed later such specifications will
be supplied. But such a construal captures in its char-
acterization both what was essentially at issue in the
great agnostic debates in the nineteenth century and
the issue as it has come down to us.

II

T. H. Huxley was by training a biologist, but he had
strong philosophical interests and as a champion of
Darwinism he became a major intellectual figure in
the nineteenth century. In his “Science and Christian
Tradition” (in Collected Essays), Huxley remarks that
agnosticism is a method, a stance taken toward putative
religious truth-claims, the core of which is to refuse
to assent to religious doctrines for which there is no
adequate evidence, but to retain an open-mindedness
about the possibility of sometime attaining adequate
evidence. We ought never to assert that we know a
proposition to be true or indeed even to assent to that
proposition unless we have adequate evidence to sup-
port it.

After his youthful reading of the Scottish meta-
physician William Hamilton's Philosophy of the Un-
conditioned (1829), Huxley repeatedly returned to
questions about the limits of our possible knowledge
and came, as did Leslie Stephen, to the empiricist
conclusion that we cannot know anything about God
or any alleged states or realities “beyond phenomena.”
Whether there is a God, a world of demons, an immor-
tal soul, whether indeed “the spiritual world” is other
than human fantasy or projection, were all taken by
Huxley to be factual questions open to careful and
systematic empirical investigation. In short, however
humanly important such questions were, they were also
“matters of the intellect” and in such contexts the

central maxim of the method of agnosticism is to “fol-
low your reason as far as it will take you, without
regard to any other consideration. And negatively: In
matters of the intellect do not pretend that conclusions
are certain which are not demonstrated or demon-
strable” (Huxley, pp. 245-46). Operating in accordance
with such a method does not justify “the denial of the
existence of any Supernature; but simply the denial
of the validity of the evidence adduced in favour of
this, or that, extant form of Supernaturalism” (p. 126).
Huxley found that he could no more endorse materi-
alism, idealism, atheism, or pantheism than he could
theism; they all claimed too much about essentially
contested matters. Huxley felt that people espousing
such world views were too ready to claim a solution
to the “problem of existence,” while he remained
painfully aware that he had not succeeded in coming
by such a solution and in addition retained “a pretty
strong conviction that the problem was insoluble” (pp.
237-38).

This conviction is at the heart of his agnosticism.
Huxley was convinced that Kant and Hamilton had
established that reason fails us—and indeed must fail
us—when we try to establish that the world is finite
in space or time or indefinite in space or time, rational
or irrational, an ordered whole or simply manifesting
certain ordered features but not something properly
to be called an ordered whole. Answers to such ques-
tions reveal something about our attitudes but can
never provide us with propositions we can justifiably
claim to be true or even know to be false. Agnosticism
is a confession of honesty here. It is “the only position
for people who object to say that they know what they
are quite aware they do not know” (p. 210).

Such skepticism concerning the truth-claims of reli-
gion and metaphysics, including, of course, meta-
physical religiosity, should not be taken as a denial that
there can be reliable knowledge. Rather Huxley
argued, as John Dewey did far more systematically
later, that we can and do gain experimental and ex-
periential knowledge of nature, including human na-
ture, and that this, by contrast with so-called “super-
natural knowledge,” becomes increasingly more exten-
sive and reliable. And while remaining an agnostic,
Huxley saw in science—basically the scientific way of
fixing belief—a fundamental and well grounded chal-
lenge to the authority of the theory of the “spiritual
world.”

Whatever may have been the case in the seventeenth
century, there was in Huxley's time a state of war
between science and religion. Huxley took science to
be a challenge to claims of biblical infallibility and
revelation. The whole supernatural world view built
on the authority of the Bible and revelation must come
under scientific scrutiny and when this is done it be-
comes gradually apparent that the use of the scientific
method and appeals to scientific canons of criticism
give us a far more reliable method of settling belief
than do the scriptures and revelation.

To commit ourselves to the Bible as an infallible
authority is to commit ourselves to a world view in
which we must believe that devils were cast out of
a man and went into a herd of swine, that the deluge
was universal, that the world was made in six days, and
the like. Yet such claims are plainly and massively
contravened by our actual empirical knowledge such
that they are quite beyond the boundaries of respon-
sible belief. About such matters, Huxley argues, we
ought not to be at all agnostic. Moreover, we cannot
take them simply as myths, important for the biblical
and Christian understanding of the world, if we are
to take seriously biblical infallibility and the authority
of revelation. For the Jewish-Christian world view to
establish its validity, it must provide us with adequate
grounds for believing that there are demons. But there
is no good evidence for such alleged realities and to
believe in them is the grossest form of superstition
(Huxley, p. 215).

Even if we fall back on a severe Christology, we
are still in difficulties, for it is evident enough that Jesus
believed in demons and if we are to adopt a radical
Christology and take Jesus as our infallible guide to
the divine, we are going to have to accept such super-
stitious beliefs. Such beliefs affront not only our intel-
lect—our credibility concerning what it is reasonable
to believe—they also affront our moral sense as well
(p. 226). Yet once we give up the Gospel claim that
there are “demons who can be transferred from a man
to a pig,” the other stories of “demonic possession fall
under suspicion.” Once we start on this slide, once we
challenge the ultimate authority of the Bible, and
follow experimental and scientific procedures, the
ground for the whole Judeo-Christian world view is
undermined.

Huxley obviously thinks its credibility and proba-
bility is of a very low order; an order which would
make Christian or Jewish belief quite impossible for
a reasonable and tolerably well informed man. Those
who claim to know that there are such unseen and
indeed utterly unseeable realities, are very likely peo-
ple who have taken “cunning phrases for answers,”
where real answers are “not merely actually impossi-
ble, but theoretically inconceivable.” Yet as an agnostic
one must always—even for such problematical trans-
cendental claims—remain open to conviction where
evidence can be brought to establish the truth of such
transcendent religious claims.

Leslie Stephen in his neglected An Agnostic's

Apology (1893) remarks that he uses “agnostic” in a
sense close to that of T. H. Huxley. To be an agnostic,
according to Stephen, is to reject what he calls “Dog-
matic Atheism,” i.e., “the doctrine that there is no God,
whatever is meant by God...”; it is, instead, (1) to
affirm “what no one denies,” namely “that there are
limits to the sphere of human intelligence” and (2) also
to affirm the controversial empiricist thesis “that those
limits are such as to exclude at least what Lewes called
'Metempirical knowledge'” (p. 1). (“Metempirical
knowledge” is meant to designate all forms of knowl-
edge of a transcendent, numinal, nonempirical sort.)

Stephen makes apparent the empiricist commit-
ments of his conception of agnosticism in charac-
terizing gnosticism, the view agnosticism is deliberately
set against. To be a gnostic is to believe that “we can
attain truths not capable of verification and not needing
verification by actual experiment or observation” (ibid.,
pp. 1-2). In gaining such a knowledge gnostics in
opposition to both Hume and Kant claim that by the
use of our reason we can attain a knowledge that
transcends “the narrow limits of experience” (p. 1). But
the agnostic, firmly in the empiricist tradition, denies
that there can be any knowledge of the world, includ-
ing anything about its origin and destiny, which tran-
scends experience and comprehends “the sorry scheme
of things entire.” Such putative knowledge, Stephen
maintains, is illusory and not something “essential to
the highest interests of mankind,” providing us, as
speculative metaphysicians believe, with the solution
to “the dark riddle of the universe” (p. 2).

In a manner that anticipates the challenge to the
claims of religion and metaphysics made by the logical
empiricists, Stephen says that in addition to the prob-
lem of whether they can establish the truth or probable
truth of “religious truth-claims” there is the further
consideration—actually a logically prior question—of
whether such putative claims “have any meaning”
(p. 3).

It should be noted that Stephen does not begin “An
Agnostic's Apology” by discussing semantical diffi-
culties in putative religious truth-claims but starts with
problems connected with what W. K. Clifford was later
to call “the ethics of belief.” We indeed would all
want—if we could do it honestly—to accept the claim
that “evil is transitory... good eternal” and that the
“world is really an embodiment of love and wisdom,
however dark it may appear to us” (p. 2). But the rub
is that many of us cannot believe that and in a question
of such inestimable human value, we have “the most
sacred obligations to recognize the facts” and make
our judgments in accordance with the facts. But the
facts do not give us grounds for confidence in the
viability of Judeo-Christian beliefs. Rather we are
strongly inclined when we inspect these beliefs to
believe they are wish fulfillments. And while it may
indeed be true that for the moment dreams may be
pleasanter than realities, it is also true that if we are
bent on attaining a more permanent measure of happi-
ness, it “must be won by adapting our lives to the
realities,” for we know from experience that illusory
consolations “are the bitterest of mockeries” (ibid.).
The religious platitudes “Pain is not an evil,” “Death
is not a separation,” and “Sickness is but a blessing
in disguise” have tortured sufferers far more than “the
gloomiest speculations of avowed pessimists” (ibid.).

However, the problem of meaning cuts to a deeper
conceptual level than do such arguments about the
ethics of belief. Where Judeo-Christianity does not
have a fideistic basis, it is committed to what Stephen
calls gnosticism. But does not such a doctrine fail “to
recognize the limits of possible knowledge” and in
trying to transcend these limits does it not in effect
commit the gnostic to pseudo-propositions which are
devoid of literal meaning? Logical empiricists later
answered this question in the affirmative and while it
is not crystal clear that Stephen's answer is quite that
definite, it would appear that this is what he wants
to maintain. And if that is what Stephen is maintaining,
there can, of course, be no knowledge of the divine.

Stephen raises this key question concerning the
intelligibility of such gnostic God-talk, but he does
little with it. Instead he focuses on some key questions
concerning attempts by theologians to undermine
agnosticism. He first points out that an appeal to rev-
elation is no answer to the agnostic's denial that we
have knowledge of transcendent realities or states, for
in claiming to rely exclusively on revelation these
theologians acknowledge that “natural man can know
nothing of the Divine nature.” But this Stephen replies,
is not only to grant but in effect to assert the agnostic's
fundamental principle (p. 5). He points out that H. L.
Mansel in effect and in substance affirms agnosticism
and that Cardinal Newman with his appeal to the
testimony of conscience does not provide a reliable
argument on which to base a belief in God nor does
he undermine the agnostic's position, for “the voice
of conscience has been very differently interpreted.”
Some of these interpretations, secular though they be,
have all the appearances of being at least as valid as
Newman's, for all that Newman or anyone else has
shown. Moreover, on any reasonable reading of a prin-
ciple of parsimony, they are far simpler than Newman's
interpretation. Thus Newman's arguments in reality
prove, as do Mansel's, that a man ought to be an
agnostic concerning such ultimate questions where
reason remains his guide and where he does not make
an appeal to the authority of the Church. They, of

course, would have us accept the authority of the
Church, but how can we reasonably do so when there
are so many Churches, so many conflicting authorities,
and so many putative revelations? Where reason can
only lead us to agnosticism concerning religious mat-
ters, we can have no ground for accepting one Church,
one religious authority, or one putative revelation
rather than another. We simply have no way of know-
ing which course is the better course. Agnosticism,
Stephen concludes, is the only reasonable and viable
alternative.

Like Huxley, and like Hume before him, Stephen
is skeptical of the a priori arguments of metaphysics
and natural theology. “There is not a single proof of
natural theology,” he asserts, “of which the negative
has not been maintained as vigorously as the
affirmative” (p. 9). In such a context, where there is
no substantial agreement, but just endless and irre-
solvable philosophical controversy, it is the duty of a
reasonable man to profess ignorance (p. 9). In trying
to escape the bounds of sense—in trying to gain some
metempirical knowledge—philosophers continue to
contradict flatly the first principles of their prede-
cessors and no vantage point is attained where we can
objectively assess these endemic metaphysical conflicts
that divide philosophers. To escape utter skepticism,
we must be agnostics and argue that such metaphysical
and theological controversies lead to “transcending the
limits of reason” (p. 10). But the only widely accepted
characterization of these limits “comes in substance
to an exclusion of ontology” and an adherence to
empirically based truth-claims as the only legitimate
truth-claims.

It will not help, Stephen argues, to maintain that
the Numinous, i.e., the divine, is essentially mysterious
and that religious understanding—a seeing through a
glass darkly—is a knowledge of something which is
irreducibly and inescapably mysterious. In such talk
in such contexts, there is linguistic legerdemain: we
call our doubts mysteries and what is now being ap-
pealed to as “the mystery of faith” is but the theolog-
ical phrase for agnosticism (p. 22).

Stephen argues that one could believe knowledge
of the standard types was quite possible and indeed
actual and remain skeptical about metaphysics. It is
just such a position that many (perhaps most) contem-
porary philosophers would take. In taking this position
himself, Stephen came to believe that metaphysical
claims are “nothing but the bare husks of meaningless
words.” To gain genuine knowledge, we must firmly
put aside such meaningless metaphysical claims and
recognize the more limited extent of our knowledge
claims. A firm recognition here will enable us to avoid
utter skepticism because we come to see that within
the limits of the experiential “we have been able
to discover certain reliable truths” and with them “we
shall find sufficient guidance for the needs of life”
(p. 26). So while we remain religious skeptics and
skeptical of the claims of transcendental metaphysics,
we are not generally skeptical about man's capacity
to attain reliable knowledge. Yet it remains the case
that nothing is known or can be known, of the alleged
“ultimate reality”—the Infinite and Absolute—of tra-
ditional metaphysics and natural theology (p. 26). And
thus nothing can be known of God.

III

Before moving on to a consideration of some twen-
tieth-century formulations of agnosticism and to a
critical examination of all forms of agnosticism, let us
consider briefly a question that the above charac-
terization of Huxley and Stephen certainly should give
rise to. Given the correctness of the above criticisms
of Judaism and Christianity, do we not have good
grounds for rejecting these religions and is not this in
effect an espousal of atheism rather than agnosticism?

We should answer differently for Huxley than we
do for Stephen. Huxley's arguments, if correct, would
give us good grounds for rejecting Christianity and
Judaism; but they are not sufficient by themselves for
jettisoning a belief in God, though they would require
us to suspend judgment about the putative knowledge-
claim that God exists and created the world. But it must
be remembered that agnosticism is the general claim
that we do not know and (more typically) cannot know
or have good grounds for believing that there is a God.
But to accept this is not to accept the claim that there
is no God, unless we accept the premiss that what
cannot even in principle be known cannot exist. This
was not a premiss to which Huxley and Stephen were
committed. Rather they accepted the standard agnostic
view that since we cannot know or have good reasons
for believing that God exists we should suspend judg-
ment concerning his existence or nonexistence. More-
over, as we shall see, forms of Jewish and Christian
fideism when linked with modern biblical scholarship
could accept at least most of Huxley's arguments and
still defend an acceptance of the Jewish or Christian
faith.

Stephen's key arguments are more epistemologically
oriented and are more definitely committed to an
empiricist account of meaning and the limits of con-
ceivability. As we shall see in examining the conten-
tions of some contemporary critics of religion, it is
more difficult to see what, given the correctness of
Stephen's own account, it could mean to affirm, deny,
or even doubt the existence of God. The very concept
of God on such an account becomes problematical. And

this makes what it would be to be an agnostic, an
atheist, or a theist problematical.

The cultural context in which we speak of religion
is very different in the twentieth century than it was
in the nineteenth (cf. MacIntyre, Ricoeur). For most
twentieth-century people with even a minimal amount
of education, the authority of science has cut much
deeper than it did in previous centuries. The cosmo-
logical claims in the biblical stories are no longer taken
at face value by the overwhelming majority of edu-
cated people both religious and non-religious. Theolo-
gians working from within the circle of faith have
carried out an extensive program of de-mythologizing
such biblical claims. Thus it is evident that in one quite
obvious respect the nineteenth-century agnostics have
clearly been victorious. There is no longer any serious
attempt to defend the truth of the cosmological claims
in the type of biblical stories that Huxley discusses.

However, what has not received such wide accept-
ance is the claim that the acceptance of such a de-
mythologizing undermines Judaism and Christianity
and drives an honest man in the direction of agnos-
ticism or atheism. Many would claim that such de-
mythologizing only purifies Judaism and Christianity
of extraneous cultural material. The first thing to ask
is whether or not a steady recognition of the fact that
these biblical stories are false supports agnosticism as
strongly as Huxley thinks it does.

Here the new historical perspective on the Bible is
a crucial factor. The very concept of the authority of
the Bible undergoes a sea change with the new look
in historical scholarship. It is and has been widely
acknowledged both now and in the nineteenth century
that Judaism and Christianity are both integrally linked
with certain historical claims. They are not sufficient
to establish the truth of either of these religions, but
they are necessary. Yet modern historical research—to
put it minimally—places many of these historical
claims in an equivocal light and makes it quite im-
possible to accept claims about the literal infallibility
of the Bible. Conservative evangelicalists (funda-
mentalists) try to resist this tide and in reality still
battle with Huxley. They reject the basic findings of
modern biblical scholarship and in contrast to mod-
ernists treat the Bible not as a fallible and myth-laden
account of God's self-revelation in history but as a fully
inspired and infallible historical record. Conservative
evangelicalists agree with modernists that revelation
consists in God's self-disclosure to man, but they further
believe that the Bible is an infallible testimony of God's
self-unveiling. Modernists by contrast believe that we
must discover what the crucial historical but yet divine
events and realities are like by a painstaking historical
investigation of the biblical material. This involves all
the techniques of modern historical research. The vari-
ous accounts in the Bible must be sifted by methodical
inquiry and independently acquired knowledge of the
culture and the times must be used whenever possible.

Conservative evangelicalism is still strong as a cul-
tural phenomenon in North America, though it is
steadily losing strength. However it is not a serious
influence in the major seminaries and modernism has
thoroughly won the day in the intellectually respect-
able centers of Jewish and Christian learning. Huxley's
arguments do come into conflict with conservative
evangelicalism and his arguments about the plain fal-
sity, utter incoherence, and sometimes questionable
morality of the miracle stories and stories of Jesus'
actions would have to be met by such conservative
evangelicalists. But the modernists would be on
Huxley's side here. So, for a large and respectable
element of the Jewish and Christian community,
Huxley's arguments, which lead him to reject Christi-
anity and accept agnosticism, are accepted but not
taken as at all undermining the foundations of Judaism
or Christianity.

Huxley's sort of endeavor, like the more systematic
endeavors of David Strauss, simply helps Christians rid
the world of the historically contingent cultural trap-
pings of the biblical writers. Once this has been cut
away, modernists argue, the true import of the biblical
message can be seen as something of decisive relevance
that transcends the vicissitudes of time.

However, this is not all that should be said vis-à-vis
the conflict between science and religion and agnos-
ticism. It is often said that the conflict between science
and religion came to a head in the nineteenth century
and now has been transcended. Science, it is averred,
is now seen to be neutral concerning materialism or
any other metaphysical thesis and theology—the en-
terprise of attempting to provide ever deeper, clearer,
and more reasonable statements and explications of the
truths of religion—is more sophisticated and less vul-
nerable to attacks by science or scientifically oriented
thinkers. Still it may be the case that there remain some
conflicts between science and religion which have not
been overcome even with a sophisticated analysis of
religion, where that analysis takes the religions of the
world and Christianity and Judaism in particular to
be making truth-claims.

Let us consider how such difficulties might arise.
Most Christians, for example, would want to claim as
something central to their religion that Christ rose from
the dead and that there is a life after the death of our
earthly bodies. These claims seem at least to run
athwart our scientific understanding of the world so
that it is difficult to know how we could both accept
scientific method as the most reliable method of settling

disputes about the facts and accept these central
Christian claims. Moreover, given what science teaches
us about the world, these things could not happen or
have happened. Yet it is also true that the by now
widely accepted new historical perspective on the
Bible recognizes and indeed stresses mythical and po-
etical strands in the biblical stories. And surely it is
in this non-literal way that the stories about demons,
Jonah in the whale's belly, and Noah and his ark are
to be taken, but how far is this to be carried with the
other biblical claims? Are we to extend it to such
central Christian claims as “Christ rose from the
Dead,” “Man shall survive the death of his earthly
body,” “God is in Christ”? If we do, it becomes com-
pletely unclear as to what it could mean to speak of
either the truth or falsity of the Christian religion. If
we do not, then it would seem that some central Chris-
tian truth-claims do clash with scientific claims and
orientations so that there is after all a conflict between
science and religion.

Given such a dilemma, the agnostic or atheist could
then go on to claim that either these key religious
utterances do not function propositionally as truth-
claims at all or there is indeed such a clash. But if
there is such a clash, the scientific claims are clearly
the claims to be preferred, for of all the rival ways
of fixing belief, the scientific way of fixing belief is
clearly the most reliable. Thus if there are good empir-
ical, scientific reasons (as there are) for thinking that
people who die are not resurrected, that when our
earthly bodies die we die, and that there is no evidence
at all, and indeed not even any clear meaning to the
claim that there are “resurrection bodies” and a “res-
urrection world” utterly distinct from the cosmos, we
have the strongest of reasons for not accepting the
Christian claim that “Christ rose from the Dead.” The
scientific beliefs in conflict with that belief are ones
that it would be foolish to jettison. But it is only
by a sacrifice of our scientific way of conceiving of
things that we could assent to such a central religious
claim. Thus it is fair to say that our scientific under-
standing drives us in the direction of either atheism
or agnosticism.

Some contemporary theologians have responded to
such contentions by arguing that there are good con-
ceptual reasons why there could not be, appearances
to the contrary notwithstanding, such a conflict.
“Christ” is not equivalent to “Jesus” but to “the son
of God” and God is not a physical reality. Christianity
centers on a belief in a deity who is beyond the world,
who is creator of the world. But such a reality is in
principle, since it is transcendent to the cosmos, not
capable of being investigated scientifically but must
be understood in some other way. God in his proper
non-anthropomorphic forms is beyond the reach of
evidence. Only crude anthropomorphic forms of
Christian belief could be disproved by modern scien-
tific investigations.

To believe that Christ rose from the dead is to be
committed to a belief in miracles. But, it has been
forcefully argued by Ninian Smart, this does not com-
mit us to something which is anti-scientific or that can
be ruled out a priori (Smart [1964], Ch. II; [1966], pp.
44-45). A miracle is an event of divine significance
which is an exception to at least one law of nature.
Scientific laws are not, it is important to remember,
falsified by single exceptions but only by a class of
experimentally repeatable events. Thus we can believe
in the miracle of Christ's resurrection without clashing
with anything sanctioned by science. It is a dogma,
the critic of agnosticism could continue, to think that
everything that can be known can be known by the
method of science or by simple observation. A thor-
oughly scientific mind quite devoid of credulity could
remain committed to Judaism or Christianity, believe
in God, and accept such crucial miracle stories without
abandoning a scientific attitude, i.e., he could accept
all the findings of science and accept its authority as
the most efficient method for ascertaining what is the
case when ascertaining what is the case comes to
predicting and retrodicting classes of experimentally
repeatable events or processes.

Christians as well as agnostics can and do recognize
the obscurity and mysteriousness of religious claims.
The Christian should go on to say that a nonmysterious
God, a God whose reality is evident, would not be the
God of Judeo-Christianity—the God to be accepted
on faith with fear and trembling. It is only for a God
who moves in mysterious ways, that the characteristic
Jewish and Christian attitudes of discipleship, adora-
tion, and faith are appropriate. If the existence of God
and what it was to act in accordance with His will
were perfectly evident or clearly establishable by hard
intellectual work, faith would lose its force and ration-
ale. Faith involves risk, trust, and commitment. Judaism
or Christianity is not something one simply must be-
lieve in if one will only think the matter through as
clearly and honestly as possible.

What is evident is that the agnosticism of a Huxley
and a Stephen at least—and a Bertrand Russell as
well—rests on a philosophical view not dictated by
science. James Ward saw this around the turn of the
century and argued in his Naturalism and Agnosticism
that agnosticism “is an inherently unstable position”
unless it is supplemented by some general philosophical
view such as materialism or idealism (p. 21). Yet it is

just such overall views that Huxley and Stephen were
anxious to avoid and along Humean lines viewed with
a thoroughgoing skepticism.

In sum, the claim is that only if such an overall
philosophical view is justified is it the case that there
may be good grounds for being an agnostic rather than
a Christian or a Jew. The overall position necessary
for such a justification is either a position of empiricism
or materialism and if it is the former it must be a form
of empiricism which in Karl Popper's terms is also a
scientism. By this we mean the claim that there are
no facts which science cannot explore: that what can-
not at least in principle be known by the method of
science cannot be known. Where alternatively scien-
tism is part of a reductive materialist metaphysics,
there is a commitment to what has been called an
“existence-monism,” namely, the view that there is
only one sort of level or order of existence and that
is spatiotemporal existence. That is to say, such an
existence-monist believes that to exist is to have a place
in space-time. In support of this, he may point out
that we can always ask about a thing that is supposed
to exist where it exists. This, it is claimed, indicates
how we in reality operate on materialist assumptions.
And note that if that question is not apposite, “exists”
and its equivalents are not being employed in their
standard senses, but are being used in a secondary sense
as in “Ghosts and gremlins exist merely in one's mind.”
Besides existence-monism there is the even more per-
vasive and distinctively empiricist position—a position
shared by the logical empiricists, by Bertrand Russell,
and by John Dewey—referred to as “methodological-
monism”: to wit “that all statements of fact are such
that they can be investigated scientifically, i.e., that
they can in principle be falsified by observation”
(Smart [1966], p. 8).

However, critics of agnosticism have responded, as
has Ninian Smart, by pointing out that these philo-
sophical positions are vulnerable to a variety of fairly
obvious and long-standing criticisms. Perhaps these
criticisms can be and have been met, but these positions
are highly controversial. If agnosticism is tied to them,
do we not have as good grounds for being skeptical
of agnosticism as the agnostics have for being skeptical
of the claims of religion.

Some samplings of the grounds for being skeptical
about the philosophical underpinnings for agnosticism
are these. When I suddenly remember that I left my
key in my car, it makes sense to speak of the space-time
location of my car but, it is at least plausibly argued,
not of the space-time location of my sudden thought.
Moreover numbers exist but it hardly makes sense to
ask where they exist. It is not the case that for all
standard uses of “exist” that to exist is to have a place
in space-time. Methodological-monism is also beset
with difficulties. There are in science theoretically
unobservable entities and “from quite early times, the
central concepts of religion, such as God and nirvana
already include the notion that what they stand for
cannot literally be observed” (Smart [1962], p. 8).
Moreover it is not evident that we could falsify state-
ments such as “There are some graylings in Michigan”
or “Every human being has some neurotic traits” or
“Photons really exist, they are not simply scientific
fictions.” Yet we do recognize them (or so at least it
would seem) as intelligible statements of fact. Such
considerations lead Ninian Smart to claim confidently
in his The Teacher and Christian Belief (London, 1966)
that “it remains merely a dogma to claim that all facts
are facts about moons and flowers and humans and
other denizens of the cosmos. There need be no general
embargo upon belief in a transcendent reality, pro-
vided such belief is not merely based on uncontrolled
speculation” (p. 51). Smart goes on to conclude that
“the exclusion of transcendent fact rests on a mere
decision” (p. 52). So it would appear, from what has
been said above, that agnosticism has no solid rational
foundation.

The dialectic of the argument over agnosticism is
not nearly at an end and it shall be the burden of the
argument here to establish that agnosticism still has
much to be said for it. First of all, even granting, for
the reasons outlined above, that neither the develop-
ment of science nor an appeal to scientism or empiri-
cism establishes agnosticism, there are other consid-
erations which give it strong support. David Hume's
Dialogues on Natural Religion (1779) and Immanuel
Kant's Critique of Pure Reason (1781) make it quite
evident that none of the proofs for the existence of
God work, i.e., they are not sound or reliable argu-
ments. Furthermore it should be noted that their ar-
guments do not for the most part depend for their force
on empiricist assumptions and they most certainly do
not depend on the development of science.

The most rigorous contemporary work in the philos-
ophy of religion has not always supported the detailed
arguments of Hume and Kant but it has for the most
part supported their overall conclusions on this issue.
Alvin Plantinga, for example, in his God and Other
Minds (1967) rejects rather thoroughly the principles
and assumptions of both existence-monism and metho-
dological-monism and he subjects the particulars of
Hume's and Kant's views to careful criticism, yet in
the very course of giving a defense of what he takes
to be the rationality of Christian belief, he argues that
none of the attempts at a demonstration of the exist-

ence of God have succeeded. He is echoed in this claim
by such important contemporary analytical theologians
as John Hick and Diogenes Allen. This lack of validated
knowledge of the divine or lack of such warranted
belief strengthens the hand of the agnostics, though
it is also compatible with fideism or a revelationist view
such as Barth's, which holds that man on his own can
know nothing of God but must rely utterly on God's
self-disclosure.

IV

In the twentieth century a distinct element comes
to the fore which counts in favor of agnosticism but
also gives it a particular twist. This new turn leads
to a reformulation of agnosticism. It states agnosticism
in such a manner that it becomes evident how it is
a relevant response to one of the major elements in
contemporary philosophical perplexities over religion.

We have hitherto been talking as if God-talk is used
in certain central contexts to make statements of whose
truth-value we are in doubt. That is, there is no doubt
that they have a truth-value but there is a doubt which
truth-value they actually have. Theists think that at
least some of the key Jewish or Christian claims are
true, atheists think they are false, and traditional agnos-
tics, as H. H. Price puts it in his Belief (London, 1969),
suspend “judgement on the ground that we do not have
sufficient evidence to decide the question and so far
as he [the agnostic] can tell there is no likelihood that
we ever shall have” (p. 455). But in the twentieth
century with certain analytic philosophers the question
has come to the fore about whether these key religious
utterances have any truth-value at all.

A. J. Ayer defending the modern variety of em-
piricism called “logical empiricism” argued in his
Language, Truth and Logic (London, 1935) that such
key religious utterances are devoid of cognitive mean-
ing. Such considerations lead Ayer to deny that he or
anyone taking such a position could be either a theist,
an atheist, or even an agnostic. In a well known passage
Ayer comments that it is very important not to confuse
his view with agnosticism or atheism, for, as he puts
it,

(p. 219).

Ayer goes on to remark that the theist's putative claims
are neither valid nor invalid; they say nothing at all
and thus the theist cannot rightly be “accused of saying
anything false, or anything for which he has insufficient
grounds” (ibid., p. 219). It is only when the Christian,
so to speak, turns meta-theologian and claims that in
asserting the existence of a Transcendent God he is
expressing a genuine proposition “that we are entitled
to disagree with him” (ibid.).

The central point Ayer is making is that such reli-
gious utterances do not assert anything and thus they
can be neither doubted, believed, nor even asserted
to be false. With such considerations pushed to the
front, the key question becomes whether such religious
utterances have any informative content at all.

There is something very strange here. Ayer, as we
have seen, does not regard his position as atheistical
or agnostic, for since such key religious utterances
could not even be false, they could not be intelligibly
denied and since they make no claim to be intelligibly
questioned, they could not be sensibly doubted. But,
as Susan Stebbing rightly observed, “the plain man
would not find it easy to see the difference between
Mr. Ayer's non-atheism and the fool's atheism” (Steb-
bing, p. 264). But before we say “so much the worse
for the plain man,” we should remember that to believe
that such key religious utterances are unbelievable
because nonsensical is even a more basic rejection of
religious belief than simply asserting the falsity of the
putative truth-claims of Christianity, but allowing for
the possibility that they might be true.

Because of this altered conceptualization of the situ-
ation, Price, Edwards, and Nielsen have characterized
both agnosticism and atheism in a broader and more
adequate way which takes into account these problems
about meaning. A contemporary agnostic who is alert
to such questions about meaning would maintain that
judgments concerning putatively assertive God-talk
should be suspended for either of two reasons, depend-
ing on the exact nature of the God-talk in question:
(1) the claims, though genuine truth-claims, are without
sufficient evidence to warrant either their belief or

categorical rejection, or (2) their meaning is so problem-
atical that it is doubtful whether there is something
there which is sufficiently intelligible or coherent to
be believed. Where God is conceived somewhat an-
thropomorphically the first condition obtains and
where God is conceived non-anthropomorphically the
second condition obtains. The contemporary agnostic
believes that “God” in the most typical religious
employments is so indeterminate in meaning that he
must simply suspend judgment about whether there is
anything that it stands for which can intelligibly be
believed. His position, as Price points out, is like the
traditional agnostic's in being neutral between theism
and atheism (p. 454). He believes that neither such
positive judgment is justified, but unlike a contem-
porary atheist, on the one hand, he is not so confident
of the unintelligibility or incoherence of religious ut-
terances that he feels that religious belief is irrational
and is to be rejected, but, on the other hand, he does
not believe one is justified in taking these problematic
utterances as being obscurely revelatory of Divine
Truth. Neither atheism nor any of the several forms
of fideism is acceptable to him.

The contemporary agnostic sensitive to problems
about the logical status of religious utterances simply
stresses that the reasonable and on the whole justified
course of action here is simply to suspend judgment.
His doubts are primarily doubts about the possibility
of there being anything to doubt, but, second-order as
they are, they have an effect similar to the effect of
classical agnosticism and they lead to a similar attitude
toward religion. There is neither the classical atheistic
denial that there is anything to the claims of religion
nor is there the fideistic avowal that in spite of all their
obscurity and seeming unintelligibility that there still
is something there worthy of belief. Instead there is
a genuine suspension of judgment.

The thing to ask is whether the doubts leading to
a suspension of judgment are actually sufficient to
justify such a suspension or, everything considered, (1)
would a leap of faith be more justified or (2) would
the overcoming of doubt in the direction of atheism
be more reasonable? Or is it the case that there is no
way of making a rational decision here or of reasonably
deciding what one ought to do or believe?

It may indeed be true, as many a sophisticated theo-
logian has argued, that religious commitment is per-
fectly compatible with a high degree of ignorance
about God and the nature—whatever that may
mean—of “ultimate reality.” But, if this is the case and
if our ignorance here is as invincible as much contem-
porary philosophical argumentation would have us
believe, natural theology seems at least to be thor-
oughly undermined. In trying to establish whether the
world is contingent or non-contingent, whether there
is or can be something “beyond the world” upon which
the world in some sense depends, or whether there is
or could be an unlimited reality which is still in some
sense personal, theological reasonings have been no-
toriously unsuccessful. About the best that has been
done is to establish that it is not entirely evident that
these questions are meaningless or utterly unan-
swerable.

Here a Barthian turn away from natural theology
is equally fruitless. To say that man can by his own
endeavors know nothing of God but simply must await
an unpredictable and rationally inexplicable self-
disclosure of God—the core notion of God revealing
himself to man—is of no help, for when we look at
religions in an honest anthropological light, we will
see, when all the world is our stage, that we have
multitudes of conflicting alleged revelations with no
means at all of deciding, without the aid of natural
theology or philosophical analysis, which, if any, of
these putative revelations are genuine revelations. It
is true enough that if something is actually a divine
revelation, it cannot be assessed by man, but must
simply be accepted. But the agnostic reminds the reve-
lationist that we have a multitude of conflicting candi-
date revelations with no means of reasonably deciding
which one to accept. In such a context a reasonable
man will remain agnostic concerning such matters. To
simply accept the authorative claims of a Church in
such a circumstance is to fly in the face of reason.

The most crucial problem raised by the so-called
truth-claims of Judaism and Christianity is that of
conceivability—to borrow a term that Herbert Spencer
used in the nineteenth century and thereby suggesting
that there are more lines of continuity between the
old agnosticism and the new than this essay has indi-
cated. The incredibility—to use Spencer's contrasting
term—of these central religious claims is tied, at least
in part, to their inconceivability. “God” is not supposed
to refer to a being among beings; by definition God
is no finite object or process in the world. But then
how is the referring to be done? What are we really
talking about when we speak of God? How do we or
can we fix the reference range of “God”? God surely
cannot be identified in the same manner we identify
the sole realities compatible with existence-monism.
There can be no picking God out as we would a dis-
crete entity in space-time. Alternatively there are theo-
logians who will say that when we come to recognize
that it is just a brute fact that there is that indefinitely
immense collection of finite and contingent masses or
conglomerations of things, we use the phrase “the
world” to refer to, and when we recognize it could
have been the case—eternally the case—that there was

no world at all, we can come quite naturally to feel
puzzled about why there is a world at all.

Is there anything that would account for the exist-
ence of all finite reality and not itself be a reality that
needed to be similarly explained? In speaking of God
we are speaking of such a reality, if indeed there is
such a reality. We are concerned with a reality not
simply—as the world might be—infinite in space and
time, but a reality such that it would not make sense
to ask why it exists. Such a reality could not be a
physical reality.

In sum, we have, if we reflect at all, a developing
sense of the contingency of the world. The word “God”
in part means, in Jewish and Christian discourses,
whatever it is that is non-contingent upon which all
these contingent realities continuously depend. God is
the completeness that would fill in the essential incom-
pleteness of the world. We have feelings of de-
pendency, creatureliness, finitude and in having those
feelings, it is argued, we have some sense of that which
is without limit. “God” refers to such alleged ultimate
realities and to something richer as well. But surely
this, the critic of agnosticism will reply, sufficiently
fixes the reference range of “God,” such that it would
be a mistake to assert that “God” is a term supposedly
used to refer to a referent but nothing coherently
specifiable counts as a possible referent for “God,”
where “God” has a non-anthropomorphic employment.

Surely such a referent is not something which can
be clearly conceived, but, as we have seen, a non-
mysterious God would not be the God of Judeo-
Christianity. But has language gone on a holiday? We
certainly, given our religious conditioning, have a feel-
ing that we understand what we are saying here. But
do we? Perhaps, as Axel Hägerström thought, “contin-
gent thing,” “finite thing,” and “finite reality” are
pleonastic. For anything at all that exists, we seem to
be able to ask, without being linguistically or con-
ceptually deviant, why it exists. “The world” or “the
cosmos” does not stand for an entity or a class of things,
but is an umbrella term for all those things and their
structural relations that religious people call “finite
things” and many others just call “things.” What are
we talking about when we say there is something
infinite and utterly different from these “finite realities”
and that this “utterly other reality” is neither physical
nor temporal nor purely conceptual nor simply imagi-
nary, but, while being unique and radically distinct
from all these things, continuously sustains all these
“finite things” and is a mysterious something upon
which they are utterly dependent? Surely this is very
odd talk and “sustains” and “dependent” have no un-
problematical use in this context.

These difficulties and a host of difficulties like them
make it doubtful whether the discourse used to spell
out the reference range of “God” is sufficiently intelli-
gible to make such God-talk coherent. An agnostic of
the contemporary sort is a man who suspends judg-
ment, oscillating between rejecting God-talk as an
irrational form of discourse containing at crucial junc-
tures incoherent or rationally unjustifiable putative
truth-claims and accepting this discourse as something
which, obscure as it is, makes a sufficiently intelligible
and humanly important reference to be worthy of
belief.

One reading of the situation is that the network of
fundamental concepts constitutive of nonanthropo-
morphic God-talk in Judeo-Christianity is so problem-
atical that the most reasonable thing to do is to opt
for atheism, particularly when we realize that we do
not need these religions or any religion to make sense
of our lives or to buttress morality. But agnosticism,
particularly of the contemporary kind specified here,
need not be an evasion and perhaps is the most reason-
able alternative for the individual who wishes, concern-
ing an appraisal of competing world views and ways of
life, to operate on a principle of maximum caution.

BIBLIOGRAPHY

Two extensive discussions are in Robert Flint, Agnosticism
(London, 1903); and in R. A. Armstrong, Agnosticism and
Theism in the Nineteenth Century (London, 1905). See also
James Ward, Naturalism and Agnosticism (London, 1899).
The central works from Hume and Kant relevant here are
David Hume, Enquiry concerning Human Understanding
(1748), and Dialogues concerning Natural Religion (London,
1779); Immanuel Kant, Kritik der reinen Vernunft (1781),
and Die Religion innerhalb der Grenzen der blossen Vernunft
(1793). For the paradigmatic nineteenth-century statements
of agnosticism see T. H. Huxley, Collected Essays, 9 vols.
(London, 1894), Vol. V; and Leslie Stephen, An Agnostic's
Apology and Other Essays (London, 1893), and English
Thought in the Eighteenth Century (London, 1876).

The following works are central to the nineteenth-century
debate over agnosticism: Sir William Hamilton, “Philosophy
of the Unconditioned,” The Edinburgh Review (1829); H. L.
Mansel, The Limits of Religious Thought (London, 1858);
J. S. Mill, Three Essays on Religion (London, 1874); and
Herbert Spencer, First Principles (London, 1862). Noel
Annan, Leslie Stephen (London, 1952); William Irvine,
Thomas Henry Huxley (London, 1960); John Holloway, The
Victorian Sage (New York, 1953); Basil Willey, Nineteenth
Century Studies (London, 1950); and J. A. Passmore, A
Hundred Years of Philosophy (London, 1957), provide basic
secondary sources. For material carrying over to the twen-
tieth-century debate see R. Garrigou-Lagrange, Dieu, son
existence et sa nature; solution thomiste des antinomies
agnostiques (Paris, 1915); and J. M. Cameron, The Night
Battle (London, 1962). For some contemporary defenses of
agnosticism see Ronald W. Hepburn, Christianity and Para-

dox (New York, 1966); Bertrand Russell, Why I am Not a
Christian (London, 1957); H. J. Blackman, ed., Objections
To Humanism (London, 1963); Religion and Humanism, no
editor, various authors—Ronald Hepburn, David Jenkins,
Howard Root, Renford Bambrough, Ninian Smart (London,
1964); William James, The Will to Believe and Other Es-
says... (New York, 1897), attacked agnosticism.

The following books by contemporary philosophers or
analytically oriented philosophical theologians make argu-
ments relevant to our discussion. A. J. Ayer, Language, Truth
and Logic (London, 1935); Axel Hägerström, Philosophy and
Religion, trans. Robert T. Sandin (London, 1964); John Hick,
Faith and Knowledge, 2nd ed. (Ithaca, 1966); R. B.
Braithwaite, An Empiricist's View of the Nature of Religious
Belief (Cambridge, 1955); Diogenes Allen, The Reasona-
bleness of Faith (Washington and Cleveland, 1968); Ninian
Smart, The Teacher and Christian Belief (London, 1966);
idem, Philosophers and Religious Truth (London, 1964);
idem, Theology, Philosophy and Natural Sciences (Bir-
mingham, England, 1962). Alasdair MacIntyre, Secular-
ization and Moral Change (London, 1967); idem and Paul
Ricoeur, The Religious Significance of Atheism (New York,
1969); H. H. Price, Belief (London, 1969); L. Susan Steb-
bing, “Critical Notice, Language, Truth and Logic,” Mind,
new series, 45 (1936); Kai Nielsen, “In Defense of Athe-
ism,” in Perspectives in Education, Religion and the Arts,
eds. Howard Kiefer and Milton Munitz (New York, 1970);
Paul Holmer, “Atheism and Theism,” Lutheran World, 13
(1966); Alvin Plantinga, God and Other Minds (Ithaca,
1967); George Mavrodes, Belief in God (New York, 1970).

Some good critical and historical commentary on Hume
occurs in Bernard Williams, “Hume on Religion,” in David
Hume: A Symposium, ed. D. F. Pears (London, 1963); in
the essays by James Noxon, William H. Capitan, and George
J. Nathan, reprinted in V. C. Chapell, ed., Hume: A Collec-
tion of Critical Essays (New York, 1966); and in Norman
Kemp Smith's masterful and indispensable introduction to
Hume's Dialogues. See David Hume, Dialogues Concerning
Natural Religion, ed. and introduction by Norman Kemp
Smith (Edinburgh, 1947). For Kant see W. H. Walsh, “Kant's
Moral Theology,” Proceedings of the British Academy, 49
(1963).

KAI NIELSEN

[See also Gnosticism; God; Positivism; Skepticism.]