PHILOSOPHY OF RELIGION SERIES

Editor's Note

The philosophy of religion is one of several very

active branches of philosophy today, and the present
series is designed b o t h to consolidate the gains of the
past and to direct attention upon the problems of the
future. Between them these volumes will cover every
aspect of the subject, introducing it to the reader in
the state in which it is today, including its open ends
and growing points. Thus the series is designed to be
used as a comprehensive textbook for students. But it
is also offered as a contribution t o present-day
discussion; and each author will accordingly go
beyond the scope of an introduction to formulate his
own position in the light of contemporary debates.

JOHN HICK
 Philosophy of Religion Series
General Editor: John Hick, H. G. Wood Professor of Theology,
University of Birmingham

Published
John Hick (Birmingham University)
Arguments for the Existence of God
H. P. Owen (King's College, London) Concepts of Deity

Forthcoming titles
Ninian Smart (Lancaster University) The Phenomenon of Religion
William Christian (Yale University)
Oppositions of Religious Doctrines:
An Approach to Claims of Different Religions
M. J. Charlesworth (Melbourne University)
Philosophy of Religion: The Historic Approaches
Basil Mitchell (Oriel College, Oxford)
The Language of Religion
Terence Penelhum (Calgary University)
Problems of Religious Knowledge
Nelson Pike (California University)
Religious Experience and Mysticism
Donald Evans (Toronto University) Religion and Morality
Kai Nielsen (New York University)
Contemporary Critiques of Religion
Dennis Nineham (Keble College, Oxford) Faith and History
H. D. Lewis (King's College, London) The Self and Immortality
Arguments for the
Existence of God

JOHN HICK

HERDER AND HERDER

 1971
HERDER AND HERDER N E W Y O R K
232 Madison Avenue, New York 10016

Library of Congress Catalog Card Number: 75-150305

© 1971 by John Hick

Manufactured in Great Britain

Contents

Editor's Note i

Introduction vii

1. THE DESIGN ARGUMENT

(a) Introductory j
(b) The eighteenth-century design argument 2
(c) Hume's critique 7
(d) Modern restatements: (1) Lecomte du Noiiy 14

2. TELEOLOGY AND PROBABILITY

(a) Modern restatements of the design argument:
(2) F. R. Tennant 18
(b) Modern restatements: (3) Richard Taylor 22
(c) Theism and probability 27
(d) 'Design' as question rather than answer 33

3. THE COSMOLOGICAL ARGUMENT

(a) The Thomist arguments 37
(b) The self-explanatory and the non-self-explanatory 46

4. MORAL ARGUMENTS
(a) God as a postulate of practical reason 53
(b) The post-Kantian type of moral argument 56
(c) A reformulated moral argument 59

5. THE ONTOLOGICAL ARGUMENT: FIRST FORM

(a) Introductory 68
(b) Anselm's Proslogion 2 argument 70
(c) Gaunilo's criticism 76
(d) Cartesian reformulation and Kantian criticism 79

6. THE ONTOLOGICAL ARGUMENT: SECOND FORM

(a) 'Necessary being' 84
(b) The argument in Anselm 87
 (c) Norman Malcolm 91
(d) Charles Hartshorne 93
(e) A new Scotist modal argument 97

7. RATIONAL THEISTIC BELIEF WITHOUT PROOFS

(a) The religious rejection of the theistic arguments 101
(b) Can there be rational theistic belief without proofs? 107
(c) Religious and perceptual belief 113
(d) The problem of conflicting religious beliefs 117

Notes 121
Select Bibliography 136
Index 147

VI
Introduction

In this book we are concerned with philosophical arguments

for the existence of God, that is, with theistic arguments
which have their place in the history of Western philosophy
— these being primarily the cosmological type of reasoning,
which goes back to Plato; the teleological, going back to
the Stoics; the ontological proof, originated by Anselm; and
the family of moral arguments of which Immanuel Kant
was the founding father. This restriction to Western
thought corresponds to the fact that the notion of proving
the reality of the divine by philosophical argument was
contributed to the stock of human thought by the ancient
Greeks and has borne its fruit, beyond the thought of
Plato, Aristotle and the Stoics, within the Western develop-
ments of Christianity and Islam.
This means that the arguments with which we are to be
concerned are theistic in intent: the God whose existence
they seek to prove, or to show to be probable, is the God of
ethical monotheism (i.e. of Judaism, Christianity and Islam).
We may briefly characterise this conception of the divine as
follows. God is the unique infinite personal Spirit who has
created out of nothing everything other than himself; he is
eternal and uncreated; omnipotent and omniscient; and his
attitude to his human creatures is one of grace and love. This
is the being whose existence is sought to be established by
the arguments we are to examine.
Something should however perhaps be said at this point
about the phrase 'God exists', the propriety of which has
been challenged on b o t h philosophical and theological
grounds. We are not at the moment concerned with the
question whether God exists, but with the suitability of the
locution 'God exists', used either in affirmation or denial.
The philosophical objection is that 'x exists' is a logically
misleading way of saying something else, namely that the
description or definition indicated by the term x is in-
stantiated. Thus the correct question is not whether a being
vii
called God does or does not have the property of existence,
but whether the definition of 'God' has or lacks an instance.
This well-known Russellian analysis of 'exists' (1) entails that
'God exists' and 'the existence (or the reality) of God' are
solecisms. They remain however very convenient solecisms,
and they can be rendered harmless by stipulating that 'God
exists' is to be construed as shorthand for 'There is an
individual, and only one, who is omniscient, omnipotent,
e t c ' With this understanding it is permissible to retain the
traditional phrase even whilst acknowledging its strict
impropriety.
The theological objection to speaking of the existence of
God is of a quite different kind and is formulated as follows
by Paul Tillich: 'Thus the question of the existence of God
can be neither asked nor answered. If asked, it is a question
about that which by its very nature is above existence, and
therefore the answer — whether negative or affirmative —
implicitly denies the nature of God. It is as atheistic to affirm
the existence of God as it is to deny it. God is being-itself,
not a being.' (2) This is in effect a theological-semantic
recommendation that the term 'existence' be applied only to
entities within the created realm, with the result that it
becomes improper to assert of a postulated creator of this
realm that he exists. The recommendation operates as an
emphatic rejection of any notion of God as a finite object
alongside others in the universe. But we still want to be able
to distinguish between there being and there not being an
ultimate reality which is not a part of the universe and to
which we may properly direct our 'ultimate concern'; and the
term 'the existence of God' enables us to do this. Once again,
then, it seems on the whole preferable to retain the tradi-
tional phrase than to risk concealing important issues by
rejecting it.
The traditional theistic arguments are commonly dis-
tinguished as being either a priori or a posteriori. An a
posteriori argument is one which relies on a premise derived
from (hence after or posterior to) experience. Accordingly
a posteriori arguments for the existence of God infer a deity
from evidences within our human experience. An a priori
argument, on the other hand, operates from a basis which is
logically prior to and independent of experience. It rests
viii
upon purely logical considerations and (if it succeeds)
achieves the kind of certainty exhibited by mathematical
truths.
In point of fact only one strictly a priori theistic proof has
been offered — the ontological argument of Anselm,
Descartes and others. This claims on a priori grounds that the
idea of 'the most perfect and real conceivable being' is the
idea of a being which must and therefore does exist; for a
Non-existent could never be the most perfect and real con-
ceivable being. This argument has been the subject of a great
deal of discussion during the last couple of decades, including
the rediscovery of a second form of the proof which some
claim to be immune to the traditional criticisms. Chapters 5
and 6 are devoted to the ontological argument in b o t h its
forms.
As regards the a posteriori theistic proofs it is necessary to
distinguish between, on the one hand, those which profess to
constitute strict apodictic demonstrations or 'knock-down
arguments' and those, on the other hand, which are of the
nature of probability arguments, seeking to persuade us that
theism is the most reasonable of the available alternatives.
In considering any attempted strict proof of divine
existence it is necessary to be clear what ' p r o o f means in
such a context. We must in fact distinguish several different
senses of 'prove'. For there are two senses in which we may
speak of something being proved in which it is a non-
controversial statement that the existence of God can be
proved; and these need to be mentioned and set aside in
order to isolate the central problem, which concerns a third
sense.
The existence of God can undoubtedly be proved if a
proof is equated with a formally valid argument. For it is a
familiar logical truism that a valid argument can be con-
structed with any proposition as its conclusion. Given any
proposition, q, it is possible to supply other propositions
such that it would be inconsistent to affirm these and to
deny q. The propositions thus supplied constitute premises
from which q follows as a conclusion. One can easily
construct a proof in this sense for the existence of God. For
example: If Birmingham exists, God exists; Birmingham
exists; therefore God exists. The argument is formally
ix
impeccable — one cannot rationally affirm the premises and
deny the conclusion.
However this first sense of 'prove' is referred to here only
to be dismissed as an inconvenient and confusing usage. It is
much better to follow the more normal practice and to
distinguish between an argument being valid and its con-
clusion being true. The validity of an argument is a purely
formal characteristic of the relation between its constituent
propositions and does not guarantee the truth of any of
them. It guarantees that if the premises are true the
conclusion is true also; but it cannot guarantee that the
premises, and therefore the conclusion, are true.
A second sense of 'prove' is that in which a conclusion is
said to be proved, not merely if it follows from premises, b u t
if it follows from true premises. We may consider this second
sense in relation to and in distinction from a third in which
these conditions are supplemented by the yet further require-
ment that the premises are known or acknowledged to be
true. There might, in sense number two, be all manner of
valid arguments in which true premises lead to true con-
clusions but which do not prove anything to anyone because
no one acknowledges their premises as being true. In this
sense, all that can be said is that there may be a proof of
God's existence if God exists but not if he does not! But this
is so trivial a point that no one will be concerned to dispute
it. It is surely the third sense, in which to prove something
means to prove it to someone, that is really in question when
we ask whether the existence of God can be proved.
The sense of 'prove', then, which most concerns us is that
in which we speak of proving a certain conclusion to an
individual or a .group. Here it is required not only that the
conclusion follows from the premises, and not only that the
premises from which it follows are true, but also that they
are acknowledged to be true by those to whom we are
seeking to prove the conclusion. It is at this point that a basic
philosophical objection emerges to all strict theistic proofs of
the a posteriori type — namely that they necessarily beg the
question, in that a person who accepts their premises already
acknowledges the reality of God. For theistic arguments of
this type rely upon some connection between God and the
world. In order to provide a basis for a strict proof of God's
x
existence the connection must be such as to warrant the
proposition, 'If the world (or some particular aspect of it)
exists, God exists'. But clearly anyone who accepts this
premise already either acknowledges the existence of God or
else is unable to reason at all. And it is idle to offer a demon-
stration to one who does not need it or is incapable of using
it. Those, on the other hand, who do not accept the premise
will never be led by this route to affirm the reality of God.
But might one not become converted to this premise by
considering the implications of an additional premise — that
the world must be ultimately explicable by reference to some
reality beyond itself, and is not a sheer inexplicable 'brute
fact' that can only be accepted as such? As Father Copleston
says in the course of his debate with Bertrand Russell, 'my
point is that what we call the world is intrinsically un-
intelligible, apart from the existence of God'. (3) The first
cause argument and the argument from contingency b o t h
employ this principle either explicitly or implicitly. Their
logical form is that of a dilemma: either there is a God or the
world is ultimately unintelligible. The one argument urges
that either there is an endless and therefore meaningless
regress of causes or else the causal series must finally be
anchored in an uncaused first cause, which is God. The other
argument claims that each item in nature points beyond itself
for its sufficient explanation, and urges that either the regress
of explanations runs out to infinity, with the result that
nothing is ever finally explained, or else it must terminate in
a self-existent being which neither needs nor is capable of
further explanation; and this is God. Clearly the force of
these arguments depends upon the decisive ruling out of one
alternative, namely that the world is ultimately inexplicable,
so that we shall be driven by force of logic to the other
conclusion, that God exists. But the difficulty of finding an
agreed premise arises again at this point. For it is precisely
this excluding of the non-theistic alternative that apparently
is not and cannot be accomplished by logical considerations
alone. It rests upon a fundamental act of faith, faith in the
ultimate 'rationality' of existence; and this is part of the
larger faith which the atheist refuses. He believes on the
contrary that the universe is devoid of ultimate meaning or
purpose and that the question as to why there is anything at
xi
all has no meaning and therefore no answer. This cosmo-
logical type of reasoning will be examined more fully in
Chapter 3.
Whilst the first cause argument and the argument from the
contingency of the world profess to be strictly demon-
strative, the other a posteriori arguments, particularly the
design and moral arguments, attempt to establish a high
probability rather than a logical certainty. They direct
attention to some aspect of the world or of human
experience — for example, the order and beauty of nature
and its apparently purposive character, or man's ethical
experience and appreciation of Values — and conclude that
this is most adequately explained by postulating a divine
creator or a transcendent ground of value. It is not claimed
that the intellectual move from these starting-points to God
proceeds on the ironclad rails of logical entailment. There can
be no strict deduction of an infinite deity from the
character of finite things. Rather these function as significant
signs and clues, pointing with varying degrees of particularity
and force to the reality of God. Formulated as arguments
directed to the non-believer, such inferences accordingly
centre upon the notion of probability. Their general form is:
in view of this or that characteristic of the world it is more
probable that there is a God than that there is not. The
design argument — since the eighteenth century the one to
have made the widest appeal — and the notion of probability
in relation to theistic arguments are discussed in Chapters 1
and 2.
The moral argument, as it has generally been formulated,
fails to impress the atheist because it clashes ab initio with
the naturalistic type of ethical theory which he holds. The
argument is thus foiled at the outset by the difficulty of
finding an agreed premise. But in Chapter 4 I suggest that it is
possible to start from a naturalistic ethical theory and to
show that some of the acts which an atheist most highly
respects and praises are incapable of rational justification on
the basis of his own theory — so that the naturalistic picture
of man and his place in the universe is insufficient. This does
not prove the existence of God; but by raising a question to
which God could be the answer it shows such an existence to
be possible,
xii
 And this, I suggest, is what all of the traditional theistic
arguments do. They establish the possibility of God by
pointing to aspects of human experience which pose
problems in reflecting upon which we find ourselves faced
with the alternatives of accepting sheer formless and
meaningless mystery or the worship-eliciting mystery of God.
We may choose the former; but the latter always remains as
an open possibility. The way in which the different theistic
arguments fail to establish the existence of God but succeed
m establishing the possibility of God will be the theme 01 tne
succeeding chapters.
At various times in the past these traditional arguments
have been regarded as being of major religious importance;
for they answered the question, Is it reasonable to believe in
God? But the prevailing contemporary theological view of
them (of which the opinion I have expressed above is a
variant) is not only that they do not prove the existence of
God but also that it was a mistake to think that any such
proof could ever be constructed. In the last chapter I there-
fore enter into the contemporary discussion concerning the
rationality of belief in the reality of an unproved God.
But even if the theistic arguments are not religiously
important they remain philosophically important. They are
classic pieces of reasoning, inextricably intertwined with the
rest of the history of Western philosophy. Anyone who has
grappled with the concepts of necessity, existence, perfection
and deity involved in the ontological argument; with the
notions of causality and explanation involved in the cosmo-
logical argument; with the ideas of order and purpose which
occur in the teleological argument; and with the question of
the metaphysical implications of ethics raised by the moral
arguments, has already received a philosophical education.
A number of friends have read and criticised one or other
or several of the following chapters, and I have as a result
been able to make the book much less inadequate that it
would otherwise have been.
The various basic writings, from Plato to the present day,
which are discussed here are collected together in my book of
readings 'The Existence of God' (New York: Macmillan, and
London: Collier-Macmillan, 1964), which thus forms a
companion volume to the present work. (4)
xiii
1 The Design Argument

(a) Introductory

The teleological or design argument, proceeding from the

world as exhibiting intelligible order to a divine intelligence
as the source of that order, is very ancient. (1) It perhaps has
its earliest roots in the thought of Plato, who argued that the
physical universe is unintelligible apart from mind, which
moves and orders it. (2) However that idea has developed
more readily, from the point of view of our medieval and
modern distinctions, into the 'cosmological' than into the
'teleological' proof. The classic exposition in the ancient
world of a form of reasoning which is continuous with what
we today call the design argument comes from the Stoic
philosopher Lucilius Balbus (reported in Cicero's 'De Natura
Deorum'), who describes the strongest ground of theistic
belief as one which 'is drawn from the regularity of the
motion and revolution of the heavens, the distinctness,
variety, beauty, and order of the sun, moon, and all the stars,
the appearance only of which is sufficient to convince us
they are not the effects of chance'. (3)
After its first appearance in ancient Greece the argument
had its second life in medieval Europe. It occurs as the fifth
of Aquinas's 'Five Ways' and was employed by a number of
other theologians in the latter Middle Ages. But its greatest
flourishing was in the seventeenth and eighteenth centuries
when the development of the descriptive sciences, particularly
zoology, botany, astronomy and anatomy enabled the tele-
ologist to illustrate his case with hundreds of examples of
adaptation and therefore, he assumed, of design in nature.
Much of the literature developing this theme was produced in
England, two notable series of volumes propagating the
gospel of an evident divine arrangement of the natural world
being the 'Boyle Lectures' (4) and the 'Bridgewater Treatises
on the Power, Wisdom and Goodness of God as Manifest in
the Creation' (8 vols, London, 1833-40). Indeed the design
1
argument, as a feature of the thought-world of 'eighteenth-
century optimism', moved through christian apologetics and
philosophical preaching in this country with an impetus
which was to carry it on into the Victorian age t o do battle
with Darwinism.
It was likewise in the eighteenth century that the design
argument met its most formidable philosophical critic, David
Hume. However, the argument went on being reformulated
long after Hume, one of the most widely used versions
coming in William Paley's 'Natural Theology, or Evidences of
the Existence and Attributes of the Deity, collected from the
Appearances of Nature', published in 1802. This was a
standard work of christian apologetics in the English-speaking
world throughout the nineteenth century, and as recently as
the 1950s one could meet elderly ministers, educated
towards the end of last century, who had been brought up on
Paley. Paley's arguments are now defunct; b u t the design
argument itself has not died with the passing years. It was
revived in the last generation by two British philosophers,
F. R. Tennant (5) and A. E. Taylor (6), and by the French
physicist Pierre Lecomte du Noiiy; (7) and within the last few
years one aspect of it has been interestingly reformulated by
the American philosopher Richard Taylor. (8)
It is impossible in the space available t o describe and
discuss every historically important formulation of the
argument; and indeed in view of the largely repetitive
character of the relevant literature this would prove neither
interesting nor useful. I shall instead present the eighteenth-
century version of the argument, Hume's critique and the
way in which Darwin's work has reinforced some of Hume's
arguments; and then the much more recent versions offered
by Tennant, du Noiiy and Richard Taylor.

(b) The eighteenth-century design argument

For the eighteenth-century argument we can hardly do better

than begin with Paley's famous version of the analogy of the
watch. When the eighteenth-century apologists compared the
workings of the solar system and of terrestrial nature with
that of a machine, the example of mechanical design that
2
most readily came to mind was the clock which stood on the
mantlepiece in every reader's drawing-room — for the readers
of popular philosophy and theology were of course the
gentry, from the sale of whose libraries by their descendants
we in these latter days have acquired our leather-bound
copies of Ray and Derham and Paley and the rest.
The passage introducing 'Paley's Watch' occurs at the
beginning of his Natural Theology:

In crossing a heath, suppose I pitched my foot against a

stone, and were asked how the stone came to be there; I
might possibly answer, that, for anything I knew to the
contrary, it had lain there for ever: nor would it perhaps
be very easy to show the absurdity of this answer. But
suppose I had found a watch upon the ground and it
should be inquired how the watch happened to be in that
place; I should hardly think of the answer which I had
before given, that, for anything I knew, the watch might
have been always there. Yet why should not this answer
serve for the watch as well as for the stone? Why is it not
as admissable in the second case, as in the first? For this
reason, and for no other, viz. that, when we come to
inspect the watch, we perceive (what we could not
discover in the stone) that its several parts are framed and
put together for a purpose, e.g. that they are so formed
and adjusted as to produce motion, and that motion so
regulated as to point out the hour of the day; that, if the
different parts had been differently shaped from what they
are, or placed after any other manner, or in any other
order, than that in which they are placed, either no motion
at all would have been carried on in the machine, or none
which would have answered the use that is now served by
it. (9)

Paley then develops particular aspects of his analogy. I am

referring here to the first four of his eight points in chapter 1
(the last four adding nothing new of any significance),
together with the argument in chapter 2. First, it does not
weaken our inference from watch to watchmaker if we have
never seen a watch being made: the object itself convinces us
that it is a product of intelligence. Nor, second, does it
3
materially weaken this inference if the watch does not appear
to work perfectly: we can still appreciate that it must have
come about by design instead of being produced by the
random operation of such natural forces as wind and rain.
Nor again, third, would it cast doubt upon the inference if
there should be parts of the watch whose function we cannot
discover, or even parts which so far as we can tell have no
function: despite this the watch could not have come about
without a watchmaker. Fourth — and here Paley may have
had in mind an argument derived from Hume's Dialogues,
published twenty-three years earlier (10) — no one in his senses
could 'think the existence of the watch, with its various
machinery, accounted for, by being told it was one out of
possible combinations of material forms; that whatever he
had found in the place where he found the watch, must have
contained some internal configuration or other; and that this
configuration might be the structure now exhibited, viz. of
the works of a watch, as well as a different structure'. (11)
Fifth — and this again may refer to an argument of Hume's —
no one would be satisfied 'to be answered, that there existed
in things a principle of order, which had disposed the parts of
the watch into their present form and situation. He never
knew a watch made by the principle of order; nor can he
even form to himself an idea of what is meant by a principle
of order, distinct from the intelligence of the watch-
maker.' (12) And sixth, if we found that the watch included
a mechanism for manufacturing further watches, so that the
watch we had found had probably itself been made by a
previous watch, and so on back in time, this would not
counter our inference to an original watchmaker: on the
contrary it would only increase our admiration for the
genius of the designer of such complex self-producing
machines.
This being so, Paley is now able to apply his analogy
directly to the world, claiming that 'every indication of con-
trivance, every manifestation of design, which existed in the
watch, exists in the works of nature; with the difference, on
the side of nature, of being greater and more, and that in a
degree which exceeds all computation. I mean, that the con-
trivances of nature surpass the contrivances of art, in the
complexity, subtility, and curiosity of the mechanism; and
4
still more, if possible, do they go beyond them in number
and variety; yet, in a multitude of cases, are not less evidently
mechanical, not less evidently contrivances, not less evidently
accommodated to their end, or suited to their office, than are
the most perfect productions of human ingenuity.' (13)
In the course of his book Paley discusses in detail many
instances of the suitablility of the bodily structure of
animals, birds and fishes to the conditions of their life on the
earth, in the air, or in the sea. For example, 'there is precisely
the same proof that the eye was made for vision, as there is
that the telescope was made for assisting it.' (14)
And if Paley had lived more recently he would doubtless
have compared the eye, to even greater effect, with a cine
camera taking a colour film. There are similar arrangements
of adjustable lenses and light- and colour-sensitive surfaces,
with a bonus in the case of the eye of an automatic
mechanism for washing the lense.
But instead of staying with Paley, let us further sample the
flavour of the eighteenth-century design argument by
opening William Derham's works. His Physico-Theology: or,
A Demonstration of the Being and Attributes of God, from
his Works of Creation, first published in 1713 and containing
the Boyle Lectures for 1711 and 1712, went through many
editions during the following decades. After citing dozens of
examples of adaptation in animal life Derham sums up:

And upon the whole matter, what less can be concluded,

than that there is a Being infinitely wise, potent, and kind,
who is able to contrive and make this glorious scene of
things, which I have thus given only a glance of? For what
less than infinite, could stock so vast a globe with such a
noble set of animals? All so contrived, as to minister to
one another's help some way or other, and most of them
servicable to man peculiarly, the top of this lower world,
and who was made, as it were, on purpose to observe and
survey, and set forth the glory of the infinite Creator,
manifested in his works! Who? What, but the great God,
could so admirably provide for the whole animal world,
everything servicable to it, or that can be wished for, either
to conserve its species, or to minister to the being or well-
being of individuals! Particularly, who could feed so
5
 spacious a world, who could please so large a number of
palates, or suit so many palates to so great a variety of
food, but the infinite Conservator of the world! And who
but the same great He, could provide such commodious
clothing for every animal; such proper houses, nests, and
habitations; such suitable armature and weapons; such
subtilty, artifice, and sagacity, as every creature is more or
less armed and furnished with, to fence off the injuries of
the weather, to rescue itself from dangers, to preserve itself
from the annoyances of its enemies; and, in a word, to
conserve itself, and its species! What b u t an infinite super-
intending Power could so equally balance the several
species of animals, and conserve the numbers of the
individuals of every species so even, as not to over or
under-people the terraqueous globe! (15)

The next year, in 1714, Derham published a companion

volume, his Astro-Theology: or, A Demonstration of the
Being and Attributes of God, from a Survey of the Heavens,
from which I shall quote only one characteristic piece of
reasoning:

When we see divers pieces of curious device and work-

manship to bear the same marks of art, to have the same
masterly strokes of painting, clockwork, architecture, etc.
we conclude with great reason such pieces were made by
the same skilful hand. So when we see the same
commodious spherical figure to be imparted to the earth,
and all the heavenly bodies, we have as good reason to
conclude them to be pieces of the same hand, contrivances
and works of the same skilful architect. For if the universe
had been a work of chance, all the several globes would
have been of several forms, one of this, another of a quite
different figure: one square, another multangular, another
long, and another of another shape. . . . (16)

I shall only add to this a much more recent argument, of

which Derham would certainly have approved, found in a
book called 'Footprints of God'. The author refers to the
ozone layer in the atmosphere which filters out enough of
the burning ultra-violet rays of the sun to make life as we
6
know it possible on the surface of the earth. 'The Ozone gas
layer is a mighty proof of the Creator's forethought. Could
anyone possibly attribute this device to a chance evolu-
tionary process? A wall which prevents death to every living
thing, just the right thickness, and exactly the correct
defense, gives every evidence of plan.' (17)

(c) Hume's critique

When David Hume (1711-76), with the possible exception of

Duns Scotus Scotland's greatest philosopher, and a major
figure in the history of British empiricism, wrote a critique of
contemporary natural theology it was inevitable that his
attack should centre upon the design argument. In order to
see why his book is in the form of a dialogue — modelled
upon Cicero's 'De Natura Deorum' — we have to take note of
Hume's personal and social situation. The design argument
was deeply entrenched in the minds of his contemporaries,
and it would be hard to reconcile a direct rejection of it with
the tacit intellectual 'concordat' which governed Hume's
friendly relations with many of the leading moderate church-
men of his day. For although well known as 'Hume the
Sceptic', he was a respected member of a distinguished
Edinburgh circle in one of that city's greatest periods. The
Athens of the North provided as familiar friends of Hume's
Adam Smith, the economist; Hugh Blair, leading liberal
churchman; William Robertson, Principal of the University;
Lord Kames and Lord Elibank, judges; Allan Ramsay, the
portrait painter; and a number of other influential men
within whose society Hume played a leading part. In these
circles Hume was regarded as being, like all decent men, a
believer in God — though in the remote deity of philosophy
rather than the God of popular religion. However the
argument developed in the 'Dialogues' puts this theistic or
deistic assumption under considerable strain. Not only is the
great eighteenth-century bulwark of belief in God, the design
argument, subjected to powerful criticism, but in the second,
fourth and twelfth dialogues (further sharpened in this
direction in Hume's final revision) (18) it is suggested that
the concept of God must, in philosophic minds, be so
7
qualified that theism and atheism become in the end merely
alternative ways of describing the same facts. How was Hume
to present such unpalatable views without isolating himself
from the society of his friends and acquaintances? For
although in the end Hume did not himself publish the
'Dialogues', but left the manuscript to be printed after his
death by his nephew, he had completed it (though making
some additions during the months before his death) at least
fifteen years earlier, and much of it twenty-five years earlier;
and he had originally intended to publish it then, but was
dissuaded by his friends. Intending, then, t o make his
criticisms of natural theology public, he preferred not to
utter them directly in his own voice but to allow them to
emerge by their own force through a dialogue in which all the
major contemporary points of view were represented. Indeed
ostensibly it is not the sceptic, Philo, but the upholder of the
design argument, Cleanthes, (19) with whom the reporter of
the discussion expresses his agreement in the last paragraph.
And this device has succeeded in misleading many readers,
including some philosophical commentators. But it seems to
me that the evidence of Hume's other writings is decisively in
favour of the view argued by Kemp Smith in his critical
edition of the 'Dialogues' that 'Philo, from start to finish,
represents Hume, and that Cleanthes can be regarded as
Hume's mouthpiece only in those passages in which he is
explicitly agreeing with Philo, or in those other passages in
which, while refuting Demea, he is being used to prepare the
way for one or other of Philo's independent conclusions'.
(20) Adopting this interpretation I shall not scruple to speak
of Hume as making this or that point when, within the
dialogue, it is Philo who makes it.
Hume, then, develops a number of telling counter-
arguments, which I shall here conflate into five, against the
inference from the apparently designed character of the
world to a divine designer.
The first is the weakness of the analogy between the world
and a human artefact, such as a house, a clock, a ship, a
knitting-loom or a city. If the world were sufficiently like a
known product of human design we should be entitled to
infer that, like it, the world is a product of purposive activity.
'But surely you will not affirm, that the universe bears such a
8
resemblance to a house, that we can with the same certainty
infer a similar cause, or that the analogy is here entire and
perfect. The dissimilitude is so striking, that the utmost you
can here pretend to is a guess, a conjecture, a presumption
concerning a similar cause.' (21) (Indeed if the world were
sufficiently house-like there would be n o need to build
houses! We only make houses — or clocks — because the
world is not really very like one.) Further emphasising the
tenuousness of the world—artefact analogy, Hume later
points out that one could equally well make a case for
regarding the world, not as a vast machine, but as a vast
crustacean-like organism, or again as a vast floating vegetable.
The point he is making is that all such comparisons, including
the one relied upon in the design argument, are alike weak
and unverifiable: 'we have no data to establish any system of
cosmogony. Our experience, so imperfect in itself, and so
limited both in extent and duration, can afford us no
probable conjecture concerning the whole of things. But if
we must needs fix on some hypothesis; by what rule, pray,
ought we to determine our choice? Is there any other rule
than the greater similarity of the objects compared? And
does not a plant or an animal, which springs from vegetation
or generation, bear a stronger resemblance to the world, than
does any artificial machine, which arises from reason and
design?' (22)

Second, the design argument explains the order found in

nature by tracing its cause to a prior order existing in the
mind of the creator. This reasoning assumes that a mental
order — the order in the divine mind — is not in need of
explanation whereas a physical order is. But, Hume asks:

Allowing that we were to take the operations of one part

of nature upon another for the foundation of our
judgement concerning the origin of the whole (which can
never be admitted); yet why select so minute, so weak, so
bounded a principle [in the eighteenth-century sense of
'cause'] as the reason and design of animals is found to be
upon this planet? What peculiar privilege has this little
agitation of the brain which we call thought, that we must
thus make it the model of the whole universe? Our
9
 partiality in our own favour does indeed present it on all
occasions: But sound philosophy ought carefully to guard
against so natural an illusion. (23)

By what right, then, would we be satisfied by finding the

order of the material world prefigured in a prior world of
ideas? 'Have we not the same reason to trace that ideal world
into another ideal world, or new intelligent principle? But if
we stop, and go no farther; why go so far? Why not stop at
the material world?' (24)
Third, it is not enough for the protagonist of the design
argument simply to point, as did the eighteenth-century
apologists, to the ordered state of the universe. For any
universe, in order to exist at all, must be a sufficiently
coherent and stable complex of elements; and any such
system is capable of being regarded as the expression of
deliberate design. Any world will look designed, however it
has come into existence. 'It is in vain, therefore, to insist
upon the uses of the parts in animals or vegetables, and their
curious adjustment to each other. I would fain known how
an animal could subsist, unless its parts were so adjusted?'
(25) In other words, it is not sufficient, as warrant for an
inference from the world to God, to show that the world is
an orderly and self-sustaining system. It must also be shown
that this order could n o t have come about except by divine
activity. But this is just what cannot be done; for it is in
principle possible to give a speculative account of the ordered
state of the universe on purely naturalistic principles.
Suppose — as in what Hume calls the Epicurean hypothesis —
that the universe consists of a finite number of atoms moving
about at random. If there is any possible combination of
these atoms that constitutes a self-maintaining order it is
virtually inevitable that in infinite time they will sooner or
later fall into this combination. Perhaps, then, we are in such
a universe and in such a state of temporary or long-lived
order which has arisen b y chance out of a situation of ran-
domness.
Since Hume's day the hypothesis of a self-regulating
development of order has been strongly confirmed in the
biological sphere. For some time before the publication of
Charles Darwin's epoch-making 'Origin of Species' in 1859 a
10
number of biologists had surmised that the various species
inhabiting the globe were descended from one or only a few
original kinds of simpler organism. But until Darwin showed
the detailed stages of this development the theory of
evolution remained speculative. Darwin turned the
speculation into a concrete and convincing model of the
history of life on our planet. He showed that the mechanism
of evolution was a process of natural selection operating on
the stream of descent by inheritance with variations provided
by mutations. During the last forty or so years this picture
has been further amplified by new discoveries in genetics. To
quote Julian Huxley, the evolutionary process 'results
immediately and automatically from the basic property of
living matter — that of self-copying, but with occasional
errors. Self-copying leads to multiplication and competition;
the errors in self-copying are what we call mutations, and
mutations will inevitably confer different degrees of bio-
logical advantage or disadvantage on their possessors. The
consequence will be differential reproduction down the
generations — in other words, natural selection.' (26) And
'selective advantages so small as to be undetectable in any
one generation, are capable, when operating on the scale of
geological time, of producing all the observed phenomena of
biological evolution.' (27) Thus new species have gradually
evolved and life has grown from its simplest forms to its
present multiplicity of kinds, including man.
In its relation to the design argument this evolutionary
picture can be seen as spelling out Hume's 'Epicurean
hypothesis' in one very important sphere — that of the fitted-
ness of the different parts of a living organism to minister to
the organism's survival, and the adaptation of organisms to
the environment. Thus the Darwinian theory of the origin of
species by means of natural selection offers a causal
explanation of those features of animate life which had been
appealed to by the pre-Darwinian teleologists as evidence of a
divine designer.
Fourth, the inference to God from the appearances of
nature can never entitle us to affirm the infinite and perfect
creator depicted in Christian theology. For a finite and
imperfect world, as effect, cannot entail an infinite and
perfect creator as its cause. We are only authorised, at best,
11
to postulate as much in the divine cause as is required to
account for the observed effect. In a figure which Hume used
in the 'Enquiries' as well as the 'Dialogues', if one end of a
pair of scales is visible and the other concealed and the visible
end, containing ten ounces, is weighed down we can infer
that the other, which outweighs it, contains more than ten
ounces; b u t we cannot infer that it contains a hundred
ounces, and still less that it contains an infinitely heavy
weight. (28)
So the God of the design argument must not be affirmed
to be infinite in goodness, power, wisdom or skill:

If we survey a ship, what an exalted idea must we form of

the ingenuity of the carpenter, who framed so compli-
cated, useful, and beautiful a machine? And what surprise
must we entertain, when we find him a stupid mechanic,
who imitated others, and copied an art, which, through a
long succession of ages, after multiplied trials, mistakes,
corrections, deliberations, and controversies, had been
gradually improving? Many worlds might have been
botched and bungled, throughout an eternity, ere this
system was struck out: Much labour lost: Many fruitless
trials made: And a slow, but continued improvement
carried on during infinite ages in the art of world-making.
(29)
Nor, again, may we assume that the divine source of the
world is unitary: 'A great number of men join in building a
house or ship, in rearing a city, in framing a commonwealth:
Why may not several Deities combine in contriving and
framing a world? This is only so much greater similarity to
human affairs.' (30) Why not, then, troops of gods and
godesses, as in the ancient Greek pantheon? And Hume (in
the person of Philo) concludes:

In a word, Cleanthes, a man, who follows your hypothesis,

is able, perhaps, to assert, or conjecture, that the universe,
sometime, arose from something like design: But beyond
that position he cannot ascertain one single circumstance,
and is left afterwards to fix every point of his theology, by
the utmost licence of fancy and hypothesis. This world,
for aught he knows, is very faulty and imperfect,
12
 compared to a superior standard; and was only the first
rude essay of some infant Deity, who afterwards aban-
doned it, ashamed of his lame performance; it is the work
only of some dependent, inferior Deity; and is the object
of derision to his superiors: it is the production of old age
and dotage in some superannuated Deity; and ever since
his death, has run on at adventures, from the first impulse
and active force, which it received from him. (31)

A fifth acute point of Hume's questions the use of the

notion of probability in relation to the existence of God, in
view of the fact that the universe is unique:

When two species of objects have always been observed to

be conjoined together, I can infer, by custom, the
existence of one wherever I see the existence of the other:
And this I call an argument from experience. But how this
argument can have place, where the objects, as in the
present case, are single, individual, without parallel, or
specific resemblance, may be difficult to explain. And will
any man tell me with a serious countenance, that an
orderly universe must arise from some thought and art,
like the human; because we have experience of it? To
ascertain this reasoning, it were requisite, that we had
experience of the origin of worlds. (32)

Hume does not elaborate this argument against probability

judgements concerning the universe. But it is extremely
important, striking as it does at the root of both the design
argument and the other members of the family of probability
arguments for God. One might however take Hume's point as
being directed, not against probability judgements in the
strict sense b u t more broadly against analogical reasoning
concerning the origin of the universe. Understood in this
way, Hume's objection would fail. For the fact that the
universe is unique does not rule out the possibility of
similarities between it and some of its parts, and therefore
between its cause and the cause of those parts. As Alvin
Plantinga points out in the course of a valuable discussion of
this passage of Hume's, 'The mere fact that a thing is unique
does not of course entail that it has no property in common
13
with anything else.' (33) If, for example, the universe could
be shown t o be sufficiently like something — say, a clock —
which is known to be a product of purposive intelligence,
there would be a valid analogical inference to the production
of the universe itself b y a purposive intelligence. But of
course this is just what Hume denies in his main attack on the
design argument (see pp. 8-9 above). I therefore think it is
proper to understand Hume's reference to the uniqueness of
the universe as related to probability judgements rather than
as related to analogical reasoning in general. Seeing it in this
light I shall devote part of the next chapter to it (see section
(b)) and will therefore not pursue the matter further at this
point.
It is not I think necessary to say anything about Immanuel
Kant's influential criticism, in the 'Critique of Pure Reason'
(chapter III of the Transcendental Dialectic), of what he
called the physico-theological proof, for he did not add any
significant new arguments to those which he had encountered
in Hume's 'Dialogues', read by him in the manuscript of J. G.
Hamann's German translation. (34)

(d) Modern restatements: (1) Lecomte du Noiiy

Although chronologically the contribution of the French

physicist Pierre Lecomte du Noiiy comes after that of F. R.
Tennant, yet logically it occupies an intermediate point
between the positions of Paley and Tennant; and I therefore
propose to discuss du Noiiy first.
He argued that the improbability of the chance formation
of the basic chemical constituents of living organisms is so
great that 'it is totally impossible to account scientifically for
all phenomena pertaining t o Life, its development and pro-
gressive evolution.' (35) From this du Noiiy concludes that
we must postulate the intervention of a parascientific 'anti-
chance' factor in the evolution of the universe; and 'For a
man of science there is no difference between the meaning of
the words "anti-chance" and " G o d " . ' (36) In short, 'our
entire organised, living universe becomes incomprehensible
without the hypothesis of God.' (37)
In seeking to establish this conclusion du Noiiy discusses
14
the coming to be of a molecule of protein, this being a basic
constituent of living substances. A molecule of protein is
itself a fairly complex formation, although it is only one of
the many and various ingredients of organic life. Du Noliy
tries to calculate the probability of the appearance by chance
alone of a single molecule of protein. For the purposes of the
calculation he is supposing this molecule to contain only
2,000 atoms (this being true of the simplest proteins) and
these atoms to be of only two different kinds (whereas in
reality they are of at least four), ordered in a very simple
pattern with dissymmetry of 0-9. These are deliberately
simplifying assumptions which have the effect of over- rather
than underestimating the probability under investigation. Du
Nouy reports that according to the calculation of Charles-
Eugene Guye the probability of such a molecule coming
about by chance would be

2-02 X 10" 3 2 1 or 2-02 X ~j^

10
He continues:

The volume of substance necessary for such a probability

to take place is beyond all imagination. It would be that of
a sphere with a radius so great that light would take 1 0 8 2
years to cover this distance. The volume is incomparably
greater than that of the whole universe including the
farthest galaxies, whose light takes only 2 X 10 6 (two
million) years to reach us. In brief, we would have to
imagine a volume more than one sextillion, sextillion, sex-
tillion times greater than the Einsteinian universe (Charles-
Eugene Guye).
The probability for a single molecule of high dis-
symmetry to be formed by the action of chance and
normal thermic agitation remains practically nil. Indeed, if
we suppose 500 trillion shakings per second (5 X 1 0 1 4 ) ,
which corresponds to the order of magnitude of light
frequencies (wave lengths comprised between 0*4 and 0-8
microns), we find that the time needed to form, on an
average, one such molecule (degree of dissymmetry 0-9) in
a material volume equal to that of our terrestrial globe is
about 1 0 2 4 2 billions of years (1 followed by 243 zeros).
15
 But we must not forget that the earth has only existed
for two billion years and that life appeared about one
billion years ago, as soon as the earth had cooled ( I X 10 9
years).
We thus find ourselves in the case of the player who
does not have at his disposal the time necessary to throw
his die often enough to have one single chance of obtaining
his series, but instead of a period three or four hundred
times too short, we are faced with an interval which is
more than 1 0 2 4 3 times too short. . . .
Life itself is not even in question b u t merely one of the
substances which constitute living beings. Now, one
molecule is of no use. Hundreds of millions of identical
ones are necessary. We would need much greater figures to
'explain' the appearance of a series of similar molecules,
the improbability increasing considerably as we have seen,
for each new molecule (compound probability), and for
each series of identical throws. If the probability of
appearance of a living cell could be expressed mathem-
atically the preceeding figures would seem negligible. . . .
Events which, even when we admit very numerous
experiments, reactions, or shakings per second, need an
infinitely longer time than the estimated duration of the
earth in order to have one chance, on an average, to
manifest themselves can, it would seem, be considered as
impossible in the human sense. (38)

From this du Noiiy draws his conclusion that we cannot

rationally account for the phenomenon of life without postu-
lating the anti-chance factor which he equates with God.
The fallacy within this argument has been well exposed by
Wallace I. Matson. (39) The improbability of which du Noiiy
speaks is that of a protein molecule suddenly being formed
by chance out of a random distribution of atoms. There is
indeed an astronomical order of probability against the
i n s t a n t a n e o u s collection and combination with the
appropriate internal arrangements of the upwards of two
thousand atoms constituting a protein molecule. But this is
not what natural science supposes to have happened. 'The
evolutionary concept is that just as man is the last stage
reached to date of an immensely slow and complicated
16
process of successive modifications in less complex creatures,
so also the protein molecule itself is the resultant of a very
large number of successive stages of synthesis, beginning with
quite simple compounds.' (40) Thus the probability of
protein coming into existence in nature is not the probability
(or improbability) defined b y du Nouy's calculation, but 'the
product of the probabilities of conditions permitting the
steps of the synthesis to be realised in succession'. (41) And,
as Matson further points out, we do not have (nor is it likely
that we ever shall have) sufficient comparative information
about the conditions at different stages in the history of our
own planet and of other comparable planets to be able to
conclude anything about the frequency, and hence the
probability, with which the various stages in the formation of
protein occur in the relevant situations.
Although du Noliy does not in his book refer to this, a
considerable amount of chemical and biochemical research
has been directed upon the question of the original formation
both of proteins and of the other basic constituent of
evolving life, nucleic acid. (42) The outcome to date of this
still continuing work seems to be that it is entirely credible
that in the 'primeval soup' polymers of amino-acids con-
stituting primitive proteins were formed automatically by
'the experimentally demonstrated capacity of molecules to
order themselves'. (43)
In short, then, du Nolly's argument is altogether lacking in
cogency.

17
2 Teleology and Probability

(a) Modem restatements of the design argument:

(2) F. R. Tennant

F. R. Tennant (1866-1957), who was active as a writer

throughout the first half of the century, teaching for many
years at Cambridge University, produced in his two-volume
'Philosophical Theology' (Cambridge University Press, 1928
and 1930, reprinted 1968) what many have regarded as the
ablest presentation of a natural theology from within our
science-oriented Western culture. Tennant regarded the
nineteenth-century approach to God through religious
experience as a mistake and continued the work of the seven-
teenth- and eighteenth-century deists, who believed that the
reality of God could be established by philosophical
reasoning from the evidences of nature.
However he rejected the rationalist ideal of a strict demon-
strative proof, believing that alike in science, in ordinary life
and in religion we live and can only live by venturing upon
probabilities. Accordingly he sought to establish the 'theistic
hypothesis' as the most probable explanation of the world,
man himself with his distinctive powers and experiences
being seen as part of the world that is to be explained.
But although Tennant's 'empirical approach to theism' is
thus in principle continuous with that of the early Boyle
Lecturers, or of Paley or the authors of the 'Bridgewater
Treatises', Tennant wrote after the delayed impact of Hume's
'Dialogues' had been felt and after Darwin's evolutionary
conception had become generally accepted. (1) Accordingly
his argument is more complex and more widely based than
those of the earlier design apologists. Tennant describes his
programme of 'cosmic teleology' as follows:

The forcibleness of Nature's suggestion that she is the out-

come of intelligent design lies not in particular cases of
adaptedness in the world, nor even in the multiplicity of
18
 them. It is conceivable that every such instance may
individually admit of explanation other than in terms of
cosmic or 'external' teleology. And if it also admits of
teleological interpretation, that fact will not of itself
constitute a rigorous certification of external design. The
forcibleness of the world's appeal consists rather in the
conspiration of innumerable causes to produce, by their
united and reciprocal action, and to maintain, a general
order of Nature. Narrower kinds of teleological argument
based on surveys of restricted spheres of fact, are much
more precarious than that for which the name of 'the
wider teleology' may be appropriated in that the compre-
hensive design-argument is the outcome of synopsis or
conspection of the knowable world. (2)

Tennant's detailed argument has five strands, and the

strength he claims for it is that of all these woven together.
First, 'the mutual adaptation of thought and things,
Nature and Knowledge'. (3) There are two aspects of this:
'the primary epistemological contribution to teleological
reasoning consists in the fact that the world is more or less
intelligible, in that it happens to be more or less of a cosmos,
when conceivably it might have been a self-subsistent and
determinate " c h a o s " in which similar events never occurred,
none recurred, universals had no place, relations no fixity,
things no nexus of determination and "real" categories no
foothold.' (4) And secondly, 'It is in that Nature evokes
thought of richer kind than is involved in scientific know-
ledge, and responds to thinking such as is neither logically
necessary nor biologically needful, thus suggesting a Beyond,
that considerations as to the relation between thought and
things assume their chief significance for the teleologist.' (5)
Second, the most significant aspect of the adaptation of
organisms to their environment is not the mechanism by
which this has come about — as to this Tennant is prepared
to accept the kind of description offered by Darwinism — but
'the progressiveness of the evolutionary p r o c e s s . . . as a
whole'. (6) For 'The discovery of organic evolution has
caused the teleologist to shift his ground from special design
in the products to directivity in the process, and plan in the
primary collocations.' (7) After all, 'The survival of the fittest
B H.A.E.G. I "
presupposes the arrival of the fit.' (8) Tennant is suggesting
that whilst the mechanism whereby the species are under
continual pressure to become better adapted to their environ-
ment can indeed be described without reference to a
Designer, nevertheless the existence of the whole process as a
successful going-concern remains unexplained and can best be
accounted for by the theistic hypothesis.
Third, there is the fitness of the physical world to produce
and sustain life. 'The fitness of our world to be the home of
living beings depends upon certain primary conditions,
astronomical, thermal, chemical, etc., and on the coincidence
of qualities apparently not causally connected with one
another, the number of which would doubtless surprise any-
one wholly unlearned in the sciences; and these primary
conditions, in their turn, involve many of secondary order.
Unique assemblages of unique properties on so vast a scale
being thus essential t o the maintenance of life, their
forthcomingness makes the inorganic world seem in some
respects comparable with an organism. It is suggestive of a
formative principle.' (9) It is no doubt logically possible that
the complex, life-producing physical order came about as a
result of the chance movements of molecules, as in Hume's
'Epicurean hypothesis'; but this is enormously improbable.
'Presumably the world is comparable with a single throw of
dice. And common sense is not foolish in suspecting the dice
to have been loaded.' (10)
Fourth, 'Besides possessing a structure that happens to
render it habitable by living creatures and intelligible to some
of them, the world is a bearer of values, thus evincing
affinity with beings such as can appreciate as well as under-
stand.' (11) Tennant is referring here especially to aesthetic
values. Nature is everywhere producing beauty. And
'Nature's beauty is of a piece with the world's intelligibility
and with its being a theatre for moral life; and thus far the
case for theism is strengthened b y aesthetic consider-
ations.' (12)
Fifth, there is the significance of man's moral nature, as an
aspect of the universe that is to be explained. For to insist
that man is part of nature is by implication to require that
any explanation of nature shall also be adequate to cover
man. The natural world is so structured as to have produced
20
rational and ethical life; and this is something that must be
accounted for in any reasonable explanation of the universe.
'Nature, then, has produced moral beings, is instrumental to
moral life and therefore amenable to "instrumental" moral
valuation, and is relatively modifiable by operative moral
ideas — or, rather, by moral agents pursuing ideals. Nature
and moral man are not at strife, but are organically one. The
whole process of Nature is capable of being regarded as
instrumental to the development of intelligent and moral
creatures.' (13)
Tennant's sixth point is that his previous five con-
siderations reinforce one another with cumulative effect.
Admittedly each aspect of the universe to which he has
drawn attention can, taken separately, be explained without
resort to the supernatural. But when we see them together as
aspects of a complex universe which is ordered to produce
the marvels of animal life, culminating in human intelligence
and personality, then in Tennant's view the theistic hypo-
thesis appears as highly plausible and reasonable. The
universe might have been a mere formless chaos; but it has
form and order, and not only this but an evolving order in
which one stage is built upon another to produce in man a
consciousness of the universe which also looks beyond it to a
transcendent purposive Mind. It is this total fact that
demands explanation. And a naturalistic explanation of the
details of the creative process merely begs the ultimate
question. Tennant insists that 'no explanation is contained in
the assertion that the world is an organic whole and conse-
quently involves adaptiveness. That is only a restatement of
the occult and wondrous fact that cries for explanation. The
world's "thusness" is explained, however, if it be attributable
to the design and creativeness of a Being whose purpose is, or
includes, the realisation of moral values. Further back than a
creative Spirit it is neither needful nor possible to go.' (14)
I think that Tennant's argument still stands as the most
comprehensive and serious modern restatement of the design
argument, and I want presently to discuss it as such, but it
will be convenient first to take account of the recent contri-
bution by Richard Taylor, who has presented in a new way
the epistemological aspect of the design argument, also
stressed by Tennant (see above, p . 19).
21
(b) Modern restatements: (3) Richard Taylor

As Richard Taylor formulates it, the argument 'rests upon

the consideration that our own faculties of sense and
cognition are not only remarkable in themselves but are in
fact relied upon by us for the discovery of truth': (15)

Suppose then [says Taylor] that you are riding in a

railway coach and glancing from the window at one of the
stops, you see numerous white stones scattered about on a
small hillside near the train in a pattern resembling these
letters: THE BRITISH RAILWAYS WELCOMES YOU TO
WALES. Now you could scarcely doubt that these stones
do not just accidentally happen to exhibit that pattern.
You would, in fact, feel quite certain that they were
purposefully arranged that way to convey an intelligible
message. At the same time, however, you could not prove,
just from a consideration of their arrangement alone, that
they were arranged by a purposeful being. It is possible —
at least logically so — that there was no guiding hand at all
in back of this pattern, that it is simply the result of the
operations of inanimate nature. It is possible that the
stones, one by one, rolled down the hill and, over the
course of centuries, finally ended up in that interesting
arrangement, or that they came in some other accidental
way to be so related to each other. For surely the mere
fact that something has an interesting or striking shape or
pattern, and thus seems purposefully arranged, is no proof
that it is. . . .
Here, however, is the important point which it is easy to
overlook; namely, that if, upon seeing from the train
window a group of stones arranged as described, you were
to conclude that you were entering Wales, and if your sole
reason for thinking this, whether it was in fact good
evidence or not, was that the stones were so arranged, then
you could not, consistently with that, suppose that the
arrangement of the stones was accidental. You would, in
fact, be presupposing that they were arranged that way by
an intelligent and purposeful being or beings, for the
purpose of conveying a certain message having nothing to
do with the stones themselves. Another way of expressing
22
 the same point is, that it would be irrational for you to
regard the arrangement of the stones as evidence that you
were entering Wales, and at the same time to suppose that
they might have come to that arrangement accidentally,
that is, as the result of the ordinary interactions of natural
or physical forces. (16)
So far Taylor's reasoning is clear. If we believe that marks
that can be read as words spelling out an intelligible message
have been formed solely by the accidental operation of
natural forces, we cannot consistently treat them as the
product of a mind which was seeking to convey a message to
the reader: we cannot have it both ways. Taylor now applies
this principle to the sense organs by which we are aware of
the world around us. He points out that 'just as it is possible
for a collection of stones to present a novel and interesting
arrangement on the side of a hill and [referring to another
example which Taylor has used] for marks to appear on a
stone in a manner closely resembling some human artefact,
and for these things still to be the results of natural, non-
purposeful forces, so also it is possible for such things as our
own organs of sense to be the accidental and unintended
results, over ages of time, of perfectly impersonal, non-
purposeful forces.' (17) In fact however we do not treat our
sense organs as accidental arrangements of matter, but rather
as normally reliable media of information, and just as to take
the stones as referring to something is to be taking them as
having been intelligently ordered for that purpose, so to take
our sense organs as informing us about the world is to be
taking them to have been deliberately designed to that end:

Just as we supposed that the stones on the hill told us that

we were entering Wales — a fact having nothing to do with
the stones themselves — so we also suppose that our senses
in some manner 'tell us' what is true, at least sometimes.
The stones on the hill could, t o be sure, have been an
accident, in which case we cannot suppose that they really
tell us anything at all. So also, our senses and all our
faculties could be accidental in their origins, and in that
case they do not really tell us anything either. But the fact
remains, that we do trust them, without the slightest
reflection on the matter. (18)
23
So Taylor concludes:

We saw that it would be irrational for anyone to say both

that the marks he found on a stone had a natural, non-
purposeful origin and also that they reveal some truth with
respect to something other than themselves, something
that is not merely inferred from them. One cannot
rationally believe b o t h of these things. So also, it is now
suggested, it would be irrational for one to say both that
his sensory and cognitive faculties had a natural, non-
purposeful origin and also that they reveal some truth with
respect to something other than themselves, something
that is not merely inferred from them. / / their origin can
be entirely accounted for in terms of chance variations,
natural selection, and so on, without supposing that they
somehow embody and express the purposes of some
creative being, then the most we can say of them is that
they exist, that they are complex and wondrous in their
construction, and are perhaps in other respects interesting
and remarkable. We cannot say that they are, entirely by
themselves, reliable guides to any truth whatever, save only
what can be inferred from their own structure and arrange-
ment. If, on the other hand, we do assume that they are
guides to some truths having nothing to do with them-
selves, then it is difficult to see how we can, consistently
with that supposition, believe them to have arisen by
accident, or by the ordinary workings of purposeless
forces, even over the ages of time. (19)

This is an interesting and thought-provoking but neverthe-

less, I believe, a fallacious piece of reasoning. Taylor says that
since we treat our sense organs as we treat the words on the
Welsh hillside, namely, as conveying information to us, we
ought to regard our sense organs as we regard 'The British
Railways Welcomes you to Wales', as the work of an intelli-
gent agent. This is the argument. And its weakness is that we
do not treat our sense organs or our sense experience as we
treat a set of words; and therefore consistency does not
require us to think of them, as we think of the words in a
sentence, as having been formed by deliberate intent. We
know that sentences are normally expressions of human
24
intelligence (even though words may occasionally fall into
the form of a sentence by pure chance) because we ourselves
so use them. But we have no parallel reason to believe that our
sense organs are products of an intelligent purpose. On the
contrary, the biologists tell us a detailed and convincing story
about the way in which the sense organs of the different
animal species have been developed through the natural
processes of evolution. According to this story the present
form, for example, of the human eye as a cognitive organ is a
result, not of conscious design on the part of a divine agent,
but of the operation over millions of years of the evolu-
tionary mechanism whereby natural selection favours the
development of organs that increase the viability of a species.
Taylor is, needless to say, perfectly well aware of the fact
of evolution, and of the way in which on the face of it this
accounts for the gradual development and improvement of
sense organs. He meets this as follows:

Again, it is sometimes said that the capacity to grasp truths

has a decided value to the survival of an organism, and that
our cognitive faculties have evolved, quite naturally,
through the operation of this principle. This appears far-
fetched, however, even if for no other reason than that
man's capacity to understand what is true, through
reliance upon his senses and cognitive faculties, far exceeds
what is needed for survival. One might as well say that the
sign on the hill welcoming tourists to Wales originated over
the course of ages purely by accident, and has been
preserved by the utility it was then found to possess. This
is of course possible, but also immensely implausible.
(20)

There are two points here. The first is that our human
cognitive powers exceed what we should need merely for
purposes of survival. Taylor is evidently now speaking, not of
our sense organs, but of the higher capacities of the mind for
reasoning and speculation, as in the formation of scientific
hypotheses and the creation of metaphysical systems. But his
design argument was based upon our reliance on our sense
organs as sources of information; and the counter-argument
that these have come about by natural evolution is not met
25
 by claiming that our higher intellectual powers have not
come about in this way. But further, it is by no means
implausible to suppose that our higher intellectual powers
have in fact developed naturally within the evolutionary
process. For there is a continuity on the scale of increasing
complexity between the purely pragmatic intelligence which
man shares with the apes and other higher vertebrates, and
^The freer, much more complexly associative and creative
intelligence of a scientist or a philosopher. Further, biologists
can suggest to us the sorts of factors that must have
stimulated the distinctively human development of intelli-
gence — for example, as our ape-like progenitors left the
forests for more open country and walked upright their
hands became available to use tools; and 'Once our ancestors
were using tools, there would have been greater need, and so
greater selection, for increased intelligence.' (21) Thus not
only the present size and structure of man's brain but also
the efforts of thought which it sustains can be seen as
growing naturally out of earlier and simpler states, and so as
not requiring the hypothesis of a special non-natural creation.
Taylor's second point is the unlikelihood of man's
cognitive apparatus having come into its present state 'purely
by accident', as the stones on the hillside could (however
implausibly) be imagined to have rolled by chance into their
present positions. But of course the evolutionary hypothesis
does not suggest that the development from a light-sensitive
patch on the skin to the complex organisation of the eye
happened by chance, b u t that it happened through a long
chain of causes and effects which have taken place in
accordance with the observed regularities which we call laws
of nature.
In short, just as the most plausible explanation of the
arrangement of the stones on the hillside, in the light of the
knowledge available to us, is that it is a result of conscious
planning, so the most plausible explanation of the structure
of our sense organs, in the light of the knowledge available to
us, is that they are an outcome of natural evolutionary
processes which do not require the postulate of a controlling
mind. This difference between the two cases entirely under-
mines Taylor's new form of design argument. (22)

26
(c) Theism and probability

We may now return to the central thrust of teleological

theistic reasoning as this has been comprehensively forrrn>
lated by F. R. Tehnant. He offers a broadly based probability
argument from the character and course of the world,
including the emergence within it of the human spirit, to a
divine purpose behind the occurrence of such a process.
Tennant's position is that theism constitutes our most
adequate available explanation of this evolving universe in its
totality to the present time. He seeks to show 'that there is a
theistic world-view commending itself as more reasonable
than other interpretations or than the refusal to interpret,
and congruent with the knowledge — i.e. the probability —
which is the guide of life and of science'. (23) W. R.
Matthews has defined the same enterprise in the following
terms: 'given the universe as disclosed in experience, to find
the most reasonable account of it. Several hypotheses present
themselves for consideration, among them Theism. The
question before the mind of the philosopher, therefore, is to
decide which of the possible hypotheses squares most
adequately with the whole experience of the universe which
is open to us. The Theist maintains that his hypothesis is the
most rational in this sense.' (24) And again, more cautiously,
H. H. Farmer has said of the idea of God that 'it ought, if we
bring it to the facts of the world, to help us to make sense of
them, better sense than, or at least as good sense as, any
other available interpretation.' (25) Such advocacy of theism
as the most reasonable interpretation of the universe is
usually based upon an expansion of the design argument to
include among its evidences not only the apparently
purposive development of the natural world but also those
levels of human experience — aesthetic, moral and religious —
which transcend the needs of our animal life. Taking account
of the entire range of phenomena, it is claimed, it is more
probable that there is a God than that there is not; or (corre-
ctively) it is more reasonable or rational to believe in the
reality of God than to disbelieve or to be agnostic. It is this
use of probability or likelihood or reasonableness of belief
that I want to question, on grounds pointed out by Hume
when (pp. 13-14 above) he argued that one cannot make a
" " ' ' 27""""
probability judgement about a unique object such as is the
universe in its entirety.
The basic reason for this restriction is that^grobability is a
relational concept, this being equally true whether we thirnT
oTthe probability of events or of the probabiUty^of the truth
of a proposition.
That a possible event will occur is probable in virtue of its
relation to other events. But if the event in question — in this
case the occurrence of the universe — is so defined that there
can be no other events, then the notion of probability cannot
be brought to bear upon it. More specifically, the concept of
probability used in the physical and behavioural sciences is
based on frequency within a plurality of instances^ so that if
there is only a single case the concept fails to apply. To claim
that the probability of the universe being God-produced is
represented by some particular mathematical ratio, 1/n,
would (according to the frequency theory) presuppose it to
be known (a) that there is a certain determinate number of
universes and (b) that a certain definite proportion of these,
namely 1/n, are God-produced. Not knowing whether our
own universe falls within the God-produced or the non-
God-produced fraction, we could nevertheless know that the
probability of its being God-produced is 1/n. However, any
such use of statistical probability is ruled out by the fact that
there is by definition only one universe. For by 'the universe'
in this context is meant 'the totality of all that is, apart from
any creator of everything other than himself. It follows from
this definition that there is but one universe; and there can
accordingly be no ground for a premise to the effect that a
certain proportion of universes exhibit a certain character-
istic. (26)
Thus Tennant's analogy of the single throw of dice
('Presumably the world is comparable with a single throw of
dice. And common sense is not foolish in suspecting the dice
to have been loaded.') (27) is merely confusing. For it pre-
supposes a set of prior circumstances, such as that the dice
has six faces, in relation to which there is an antecedently
calculable probability of a particular number turning up at a
single throw.
A similar point arises with regard to propositions. A
proposition has probability in virtue of its relation to other
28
evidence-stating propositions. But if a proposition's domain is
so wide that nothing remains outside it then, again, the
notion of probability cannot apply. But just this seems to be
the case with regard to propositions interpreting the universe
as a whole. In such a case there is no conceivable background
of information on which an estimate of probability could be
based. There can be no prior corpus of propositions in
relation to which a total interpretation could be judged to be
probable or improbable, since all our particular items of
information are included within the totality which is being
interpreted. There can, in other words, be no evidence in
favour of one total interpretation over against another.
What these considerations show is that any notion of
probability properly invoked by a comprehensive teleological
argument must be other than the usual statistical or logical
concept. Tennant is fully aware of this, and indeed argues to
the same effect. He considers the view 'that, if the world be
the sole instance of its kind, or be analogous to a single
throw, there can be no talk of chances or of antecedent
probability in connection with our question. Sound as this
caution is, [he says] it does not affect the teleologist; for,
when he calls coincidence on the vast scale improbable, he
has in mind not mathematical probability, or a logical
relation, but the alogical probability which is the guide of life
and which has been found to be the ultimate basis of all
scientific induction.' (28)
Tennant discusses this alogical probability which operates
within the theistic judgement (and on which he believes that
scientific induction also ultimately rests) and characterises it
as 'non-rational, yet reasonable, certitude determined psycho-
logically'. (29) His position is I think that when the human
mind surveys the universe in which it finds itself, its con-
viction that this indefinitely complex cosmos could not
have come into being in a completely random and unplanned
way is a reasonable even though not a logically compelling
conviction, reflecting a real implausibility or (alogical)
improbability in the chance hypothesis.
But the question still has to be raised whether even this
nonmathematical and 'alogical' concept of probability is
applicable to the theistic problem. It is of course a fact that
as men have looked at the world and have been especially
29
struck by this or that aspect of it they have concluded that
there is (or that there is not) a God, or have found in the
world confirmation of an already formed conviction as to the
existence (or non-existence) of God in terms varying in
degree from 'it seems on the whole more likely than not' to
'it is overwhelmingly more probable'. But the question
remains whether the notion of probability or likelihood is
being used in such judgements to express more than a purely
personal and imponderable 'hunch' or feeling.
The situation seems to be this. Of the immense number
and variety of apparently relevant considerations some, taken
by themselves, point in one direction and some in the other.
One group can fairly be said to count as at least prima facie
evidence for the existence of God. For not only do believers
urge these particular considerations as supporting their own
position but disbelievers concurringly treat them as points
requiring special explanation. And likewise there are other
considerations which taken by themselves constitute at least
prima facie antitheistic evidences. These are matters which
non-believers emphasise and in which the believer, on the
other hand, sees a challenge to his faith which he feels
obliged to try to meet.
As examples of prima facie theistic evidence, man's
distinctively religious experience and the reports of miracles
would never be pointed out by an atheist as tending
positively to support his own position; they are items for
which he feels the need for an explanation other than the one
which the facts themselves, when taken at their face value,
suggest. It is agreed for example that there is such a thing as
'religious experience', and this very name embodies a
religious interpretation of the experiences in question as
being in some way cognitive of the divine. Accordingly it is
incumbent upon the disbeliever to respond by offering a
naturalistic interpretation of these same experiences. Such
reinterpretations have been offered in abundance and have
usually followed the path marked out by Thomas Hobbes in
his paradigmatic remark that when a man says that God has
spoken to him in a dream this 'is no more than to say he
dreamed that God spake to him'. (30)
On the other hand, as examples of prima facie antitheistic
evidence, human wickedness and the suffering of all sentient
30
creature.including man are not facts which would be selected O
by the theist as favourable premises from which to launch his
own argument: they are rather difficulties which he must
endeavour to meet from the wider resources of theism, as has
been done by a succession of thinkers from Augustine to the
present day.
There are yet other factors which are not so manifestly
evidential as those already mentioned but which seem never-
theless to fit rather more readily into one conception than
the other. For example, moral experience finds readier
hospitality within a religious metaphysic, whilst on the other
hand the vastness of the physical universe and the insignifi-
cant place occupied in it by man can more immediately be
assimilated into a naturalistic world-view.
Now none of these factors or of the indefinitely many
others that could be added to them points so unequivocally
in a particular direction as to admit of only one possible
explanation. Although in isolation they each suggest a con-
clusion, nevertheless each is capable of being fitted into
either a religious or a naturalistic context. There is no item
offered as theistic or antitheistic evidence which cannot be
absorbed by a mind operating with different presuppositions
into the contrary view. The question then is whether one way
of interpreting them can be said to be more probable than
the other or (putting the same query in another way)
whether acceptance of one interpretation can be said to be
more reasonable or rational than acceptance of the other. For
the choice is never between explanation and blank lack of
explanation, but always between alternative explanations
employing radically different categories.
From the fact that there are particular considerations
which count as prima facie evidence both for and against
theism it follows that if we attend only to selected items we
may well receive the impression that the evidence as a whole
tends in one direction — which direction depending of course
upon which items of evidence are central in our thoughts.
However, since theism and naturalism can each alike lay
claim to prima facie evidences and must each admit the
existence of prima facie difficulties, any fruitful comparison
must treat the two alternative interpretations as comprehen-
sive wholes, with their distinctive strengths and weaknesses.
31
 In what sense, however, or on what basis can it be claimed
to be established that one such total interpretation is more
probable than another? Can we, for example, simply count
points for and against? Can we say that there are, say, ten
items of prima facie evidence in favour of theism and eight
against, so that theism wins by two points; or vice versa?
Clearly no such mechanical procedure will do, for the con-
flicting considerations do not form units of equal weight. Can
we perhaps however place each item in its position on an
evidential scale in which, without being assigned numerical
value, they are nevertheless listed in order of importance? To
some extent this is feasible as a separate operation on each
side of the debate. In many instances we can accord a greater
weight to one item of theistic (or of antitheistic) evidence
than to another, and can thus at least begin to construct two
parallel lists. But we still have no agreed way of weighing an
item on one list against its opposite number on the other list
nor, therefore, of evaluating one list as a whole in relation to
the other. There arc no common scales in which to weigh, for
example, human wickedness and folly against the fact of
man's moral experience, or the phenomenon of Christ against
the problem of human and animal suffering. Judgements on
such matters are intuitive and personal, and the notion of
probability, if it is applied, no longer has any objective
meaning.
What is sought to be done here is something which no one
has yet succeeded in doing, namely to show by arguments
acceptable to all parties that one comprehensive world-view
has superior probability to another. The criteria by which to
match metaphysical systems against each other which have
usually been suggested are those developed in connection
with the coherence theory of truth — the internal logical
consistency of each system of thought; their explanatory
comprehensiveness (so that if one covers data which the
other has to leave out of account the former is to that extent
superior); and the 'adequacy' with which they illuminate and
explain what they profess to explain. The first two of these
criteria will not help us at this point, since there are forms
b o t h of theism and of naturalism which are internally con-
sistent and which are equally comprehensive in the sense that
-there are no data that evade their explanatory scope. The
32
issue is, once again, not between explanation and no
explanation but between two radically different kinds of
explanation. The crucial question is thus whether one way of
accounting for the data can be said to be inherently more
adequate than the other. This is in effect our original
problem as to whether theism or naturalism can meaningfully
be said to possess a superior antecedent probability. And it
now seems that there is no objective sense in which one
consistent and comprehensive word-view can be described as
inherently more probable than another. It is of course a
truism, if not a tautology, that to the theist theism seems
more adequate and that to the naturalist naturalism seems
more adequate. But this is because they are judging from
importantly different standpoints and with different criteria
and presuppositions. And it appears that the issue between
them is not one that can be settled by appeal to any agreed
procedure or by reference to any objectively ascertainable
probabilities.

(d) 'Design' as question rather than answer

I think it is clear, then, that the design argument neither

proves the existence of a creative mind behind the physical
universe nor in any objective sense shows this to be probable.
And yet, as many have pointed out, the response to the
complex order of the world expressed in the design argument
continues, relatively untroubled by the logical insufficiency
of the argument itself. As Kant (who described this proof as
'the oldest, the clearest, and the most accordant with the
common reason of mankind') (31) wrote:

It would therefore not only be uncomforting but utterly

vain to attempt to diminish in any way the authority of
this argument. Reason, constantly upheld by this ever-
increasing evidence, which, though empirical, is yet so
powerful, cannot be so depressed through doubts
suggested by subtle and abstruse speculation, that it is not
at once aroused from the indecision of all melancholy
reflection, as from a dream, by one glance at the wonders of
nature and the majesty of the universe — ascending from
33
 height to height up to the all-highest, from the conditioned
to its conditions, up to the supreme and unconditional
Author (of all conditioned being). (32)

The reason, I think, is that the argument focuses our

attention upon aspects of the world that evoke a sense of
wonder and an awareness of mystery independently of the
ratiocinative activity of the mind._Cosmic evolution consti-
tutes a transcendence-suggesting mystery to which religion
is a natural response. For all explanation and understanding
functions within a given framework of the ultimate facts that
there is anything at all; that what exists is a universe instead
of a mere chaos; and that it is the particular universe that it is
and includes our own questioning minds. Our thoughts
normally move within the boundaries set by these ultimate
circumstances, exploring the internal structure of our
universe and directing our actions in relation to it. But as well
as analysing the world we sometimes simply contemplate it
and wonder at it, allowing ourselves to be grasped by its
mystcriousness. Our thought then moves beyond its
customary limits to face an ultimate inexplicable 'given', and
to marvel at its unlimiteidly" "mfricate and "yet coHerent
character. The ultimate 'given' is an evolving order which has
not only produced all the facets of the physical world but
also all the qualitative marvels of human experience. And the
mystery is that this infinitely complex universe just is — it
hangs, as it were, unsupported, except by a great question
mark. And so the possibility inevitably presents itself that
this development of galactic gases to produce the spirit of
man is not just a sheer inexplicable given fact, but stems from
the only source that we can as minds see as independently
self-explanatory, namely purposive mind.
Recent developments in astronomical and cosmological
theory have emphasised the mysteriousness of the universe
revealed to us by the natural sciences. According, for
example, to the 'big bang' theory of the origin of the present
state of the universe, the matter composing it was, a large b u t
finite number of years ago, bunched together in a state of
maximum density from which it 'exploded' into the still
expanding universe in which we find ourselves. If that initial
condition was an absolute beginning, we have the mystery of
34
 a universe coming into being out of nothing. If on the other
hand the present expansion was preceded by a contraction,
and so on back in an infinite series of expansions and con-
tractions, then the ultimate character of the universe as a
pulsating system does not at all resemble a random state out
of which order has randomly emerged. Or if on the other
hand the rival 'steady state' theory should be correct, with its
corollary of the continuous coming into existence of millions
of hydrogen atoms every second to compensate for the
attenuation of the universe by expansion, we are again faced
with an ultimate system, and a system subject to a
mysterious process of repair, rather than with randomness.
Further, the religious hypothesis that the universe is a
system designed to produce intelligent personal life now
seems more plausible in the light of the widely held view that
there is life throughout the universe. It used to be said that it
is absurd to see the function of the universe in terms of the
emergence of man, who is only a fleeting organic excrescence
on the surface of a single planet of a minor star out on the
periphery of a medium-sized galaxy. Can we think of the
whole universe, whose immensity staggers our imagination, as
existing for the sake of this little pin-point of consciousness?
But it now seems likely that a very large number, perhaps the
majority, of stars have planets, and (on the evidence of our
own solar system) that something like one planet in ten is
capable of sustaining some form of life. In this case there will
be life at varying levels of development on thousands of
millions of planets circling stars in uncounted numbers of
galaxies. Intelligence will exist, often on much higher levels
than our own, all over the universe. It is not, in that case, so
implausible to consider the universe as a system designed to
produce personal Jife. For apparently it is a system, rather
than a random collocation of matter, and apparently it is
producing intelligent personal life at innumerable points
throughout its vast extent.
If this complex evolving cosmos does indeed embody a
purpose, then its mystery is illumined by that fact; otherwise
it presents a problem that has no solution. Thus the
ultimately inconclusive reasoning of the 'physico-theological
proof' leaves us with a query to which the answer may be —
God. The question may on the other hand have no answer;
35
but until this is proved it will go on being asked. Thus what,
the teleological argument (like the other theistic arguments)
does is not to establish divine existence, but to pose a
question which either has no answer or has God as its
answer. It establishes the possibility of God by drawing
attention to a mysteriousness which only theism could finally
resolve.

36
3 The Cosmological Argument

(a) The Thomist arguments

In the widest sense of the term, any theistic argument that

proceeds from the world to God can be described as cosmo-
logical. In this sense all the a posteriori arguments are
cosmological, including the design argument; and that they
do indeed have something important in common is attested
by Aquinas, who saw his five Ways, the last of which is
teleological, as variations on a single basic theme. However,
for expository purposes the term 'cosmological' has
commonly been restricted either more narrowly to the
argument from the contingency of the world (Aquinas's third
Way), or — more usefully and less narrowly — to the family
of arguments which proceed by means of the principle of
sufficient reason from the non-self-explanatory character of
the universe to a being whose existence is self-explanatory.
The central thought of the cosmological argument in this
latter sense has been well expressed by Leibniz:

Let us suppose the book of the elements of geometry to

have been eternal, one copy always having been written
down from an earlier one; it is evident that, even though a
reason can be given for the present book out of a past one,
nevertheless out of any number of books taken in order
going backwards we shall never come upon a full reason;
though we might well always wonder why there should
have been such books from all time — why there were
books at all, and why they were written in this manner.
What is true of the books is true also of the different states
of the world; for what follows is in some way copied from
what precedes (even though there are certain laws of
change). And so, however far you go back to earlier states,
you will never find in those states a full reason why there
should be any world rather than none, and why it should
be such as it is. (1)
37
 We have a classic statement of three forms of cosmological
argument in the first three of Aquinas's five Ways; and if we
now look at these, and try with the aid of contemporary
neo-Thomist philosophers to restate them in their strongest
form, it will become clear that they all hinge upon the
'principle of sufficient reason', which we shall then have to
examine more closely.
The first Way, derived directly from Aristotle (2), is from
the fact of motion or, more generally, from the fact of
change, to a prime mover. (Accordingly it is sometimes called
the kinetological argument.) Aquinas's exposition is as
follows:

Some things in the world are certainly in process of

change: this we plainly see. Now anything in process of
change is being changed by something else. This is so
because it is characteristic of things in process of change
that they do not yet have the perfection towards which
they move, though able to have it; whereas it is
characteristic of something causing change to have that
perfection already. For to cause change is to bring into
being what was previously only able to be, and this can
only be done by something that already is: thus fire, which
is actually hot, causes wood, which is able to be hot, to
become actually hot, and in this way causes change in the
wood. Now the same thing cannot at the same time be
both actually x and potentially x, though it can be actually
x and potentially y: the actually hot cannot at the same
time be potentially hot, though it can be potentially cold.
Consequently, a thing in process of change cannot itself
cause that same change; it cannot change itself. Of
necessity therefore anything in process of change is beirg
changed by something else. Moreover, this something eise,
if in process of change, is itself being changed by yet
another thing; and this last by another. Now we must stop
somewhere, otherwise there will be no first cause of the
change, and, as a result, no subsequent causes. For it is
only when acted upon by the first cause that the inter-
mediate causes will produce the change: if the hand does
not move the stick, the stick will not move anything else.
Hence one is bound to arrive at some first cause of change
38
 not itself being changed by anything, and this is what
everybody understands by God. (3)

Motus, 'motion' or 'change', includes not only movement

from one place to another (local motion) but also change of
size, as when a tree grows bigger or a candle smaller, and
change of state (alteration), as when a green leaf becomes
brown or a piece of iron becomes red-hot. (4) The latter is
for Aquinas the more fundamental mode of change, and is
understood in Aristotelian terms as the actualising of a
potentiality. Iron is potentially hot, and for a piece of iron to
become hot is for that potentiality to be realised. But
potentialities are not self-actualising. Iron does not become
hot by itself: its capacity to become hot is only realised when
fire or some other source of heat is applied to it. Actual heat,
in the fire, can actualise the iron's potentiality for heat; but if
the world did not already contain actual heat, potentiality
for heat could never become actualised. In other words, to
change is to be changed: change presupposes an operating
cause. But that prior causative activity likewise presupposes a
cause; and that, one prior to it; and so on. The regress must
either be infinite or must terminate in an 'unmoved mover',
an original source of change, whose activity does not pre-
suppose a yet prior mover but who (or which) possesses
intrinsically the power to produce change. Aquinas rules out
the idea of an infinite regress and is thus able to conclude
that there must be 'some first cause of change not itself being
changed by anything, and this is what everybody understands
by God'.
The main steps of the argument are thus:
1. Whatever changes from being potentially x (for
example, potentially hot, or potentially at a certain place, or
potentially twice its present size) to being actually x is
changed by something else.
2. That 'something else' must be actually x.
3. There cannot be an infinite regress of such 'something
else's', and therefore there must be an ultimate source of
change which does not, in order to operate, require to be
moved by something else.
Each of these steps is highly debatable. The last has
frequently been questioned, and the questioning of it has led
39
(as we shall see presently) to a neo-Thomist tendency to
assimilate the first Way to the third, in which the appeal to
the principle of sufficient reason is most explicit. The first
two steps have less often been attacked. However in his
recent book on 'The Five Ways' Anthony Kenny effectively
challenges them both. (5)
In the case of the first, Kenny points out that there is a
gap in the reasoning. Aquinas was anxious to exclude the
possibility that things are self-moving, and he produced a
series of Aristotelian arguments to show that if they are
moved they must be moved by something external to them
('Summa contra Gentiles', I 13). But he did not consider the
possibility that some things are in movement without being
or having been moved, whether by something else or by
themselves. Kenny points out that movetur covers both of
the meanings which are distinguished in English by 'is
moving' and 'is being moved'. And he suggests that this
ambiguity of the Latin may possibly account for Aquinas's
failure to state, and to argue for, the additional premise
required by his argument that 'whatever is moving is being
moved'. (6) For unless this is established it remains possible
that some things are just naturally — and conceivably
eternally — in motion (in accordance with Newton's first law)
without being caused to move at all. Thus the possibility that
the physical universe has had no initial state and consists
eternally of matter in motion has to be excluded before the
first Way can lead anywhere.
As regards the second step — that only something that is
actually x can cause something else to move from being
potentially to being actually x — Kenny points out that this
holds in some cases but not in others. It is true enough, for
example, that a wet towel will not dry you. And until the
discovery of electricity it was reasonably believed that only
something hot can make another thing hot; but now we are
familiar with temperatureless currents of electricity making
wires hot. Again, 'a kingmaker need not himself be king, and
it is not dead men who commit murders.' (7) Further, it does
not require something at the south pole to move things to the
south pole. (Otherwise how did the first explorer ever get
there?) And 'Applied to change in size, the principle seems
even more inapplicable. A man who fattens oxen need not
40
himself be fat.' (8) In short, Aquinas's second premise is not
generally enough true to sustain his argument.
When we turn to the third step — the exclusion of an
infinite regress — we have to consider whether this is a regress
of causes stretching backwards in time or a regress of
simultaneous causal conditions. Aquinas's own example of
the stick moving, say, a stone and being itself moved by the
hand, which is in turn moved by . . . suggests the latter type
of regress. So do his Aristotelian arguments against an infinite
regress in 'Summa contra Gentiles', I 13. In any case a first
Way argument based on the idea of a temporal regress to God
would lack cogency for the modern mind, which sees no a
priori impossibility in the idea of a temporally as well as
spatially unbounded universe. And indeed it was one of
Aquinas's more controversial contributions to theological
discussion in his own day to maintain that the non-eternity
of the world cannot be established by demonstrative
argument but is known by faith on the basis of revelation. (9)
It will therefore be proper to interpret the argument here in
non-temporal terms, as is done by a number of modern neo-
Thomists. According to this interpretation the infinite regress
that is excluded is a regress of simultaneous causal
conditions. (10) For an infinite chain of necessary causal
conditions would still not explain the original phenomenon.
In an analogy which some contemporary Thomist writers
have used, you do not explain the movement of a railway
truck along the track by saying that it is being pulled by
another truck in front of it, which is in turn being pulled by
one in front of it, and so one, unless you add that the whole
line of trucks is being pulled by something which moves with-
out itself having to be moved by anything else — the self-
moving mover being in this case the engine. Or again, we can
explain the movement of one of the wheels in a watch by
showing how it gears in with another wheel by which it is
being moved. But however numerous may be the wheels,
each moved by another, no multiplicity of them will explain
how this arrangement of wheels comes to be in motion. To
account for this we have to refer to something else, namely
the spring, which imparts movement to the whole system.
Thus the infinite regress which Aquinas excludes when he
says that 'we must stop somewhere' is not a chain of effects
41
and causes stretching backwards in time but a regress of
causally explanatory circumstances. The railway trucks and
the cogs in a watch all move at the same time in moving one
another; and to extend their number, even to infinity, would
still leave their motion ultimately unexplained. T o account
for this we have to postulate an independent source of
motion which originates the series of movements, regardless
of whether the series is finite or infinite. And likewise in the
case of the universe as a complex interrelated process. We do
not explain the fact of change merely by tracing out its
moving patterns either in the present or into the past. If the
fact of universal process is to be ultimately intelligible it must
be seen to depend upon a spontaneous source of motion
outside the system itself.
Aquinas's second Way, the first cause (or aetiological)
argument, is as follows:

The second way is based on the nature of causation. In the

observable world causes are found to be ordered in series;
we never observe, nor ever could, something causing itself,
for this would mean it preceded itself, and this is not
possible. Such a series of causes must however stop some-
where; for in it an earlier member causes an intermediate
and the intermediate a last (whether the intermediate be
one or many). Now if you eliminate a cause you also
eliminate its effects, so that you cannot have a last cause,
nor an intermediate one, unless you have a first. Given
therefore no stop in the series of causes, and hence no
effect, and this would be an open mistake. One is therefore
forced to suppose some first cause, to which everyone
gives the name 'God'. (11)

At first sight it is rather harder, as a matter of exegesis, to

interpret this in the modern neo-Thomist manner, because
Aquinas uses the word 'cause' in a sense which suggests that a
cause precedes its effect in time: an event could not cause
itself, he says, 'for this would mean it preceded itself. And so
he argues that a present effect must have had a prior cause,
which must in turn have had a prior cause, and so backwards
either in an infinite regress or to the point at which the
temporal series was launched b y an uncaused cause. However
42
if we understand the argument in this way it has very little
plausibility. For it is not evident that the causal sequence, or
rather complex of interlocking sequences, may not extend
indefinitely into the past without ever reaching an initial
state.
However prius does not necessarily mean temporally prior;
it may mean logically prior. (12) And so as in the case of the
first Way, the argument can be restated in non-temporal
terms. It is then a variation on the theme of the first Way.
'We are not', says Eric Mascall, 'arguing about a chain of
causes stretching back into the past, but about a chain of
causes existing in the present and each depending on the one
beyond.' (13) And F. C. Copleston offers in illustration the
fact that a person's present activity is causally dependent
upon (or has as a necessary condition) the existence of the air
which he is breathing; and this in turn is causally dependent
upon other wider physical conditions, and these upon others.
We thus have 'a hierarchy of causes, in which a subordinate
member is here and now dependent on the causal activity of
a higher member'. (14)
This series of necessary conditions must either be infinite
or depend upon a first cause which is not itself dependent
upon a yet prior cause. In this interpretation the motive
power of the argument is the need to explain the universe. As
Copleston says of the series of movers or causes in the first
two Ways, 'unless there is a "first" member, a mover which is
not itself moved or a cause which does not itself depend on
the causal activity of a higher cause, it is not possible to
explain the " m o t i o n " or the causal activity of the lowest
member.' (15) The essence of the argument is thus that if
reality is not to be ultimately inexplicable it must include a
being whose existence is self-explanatory, in relation to
which the existence of everything else can be understood.
The third Way rests upon the same basic principle of
explanatoriness or sufficient reason. Aquinas's text is as
follows:

Some of the things we come across can be but need not be,
for we find them springing up and dying away, thus some-
times in being and sometimes not. Now everything cannot
be like this, for a thing that need not be, once was not; and
43
 if everything need not be, once upon a time there was
nothing. But if that were true there would be nothing even
now, because something that does not exist can only be
brought into being by something already existing. So that
if nothing was in being nothing could be brought into
being, and nothing would be in being now, which
contradicts observation. Not everything therefore is the
sort of thing that need not be; there has got to be some-
thing that must be. Now a thing that must be, may or may
not owe this necessity to something else. But just as we
must stop somewhere in a series of causes, so also in the
series of things which must be and owe this to other
things. One is forced therefore to suppose something
which must be, and owes this to no other thing than itself;
indeed it itself is the cause that other things must be. (16)

Once again, the argument can be strengthened by removing

the reference to time. The idea that if each existing thing has
a beginning then there must have been a time in the past
when nothing existed is clearly fallacious. (17) For the world
might consist of a beginningless stream of events (or 'things'),
each individually having a beginning and an end but over-
lapping in such a way that there is no time when none of
them exists. However if, with the modern neo-Thomists, we
see the third Way as being concerned with the intelligibility
of the world, then the reference to a 'time when nothing
existed' can be dispensed with. The argument begins by
pointing to the fact of contingency — that is, the fact that
there are (at least) some things which 'can be but need not
be'. The existence of such a thing does not constitute a self-
explanatory fact. In order to find it intelligible under the
principle of sufficient reason we have to look beyond it to
other circumstances by reference to which its existence is
explained. For example, the existence of the tennis ball on
the road is explained by the facts that there is a tennis court
on the other side of the fence and that the ball has been hit
too hard and has landed on the road. But then the existence
of the court and of the game being played on it are likewise
contingent facts, explicable only by reference beyond them-
selves. To explain the existence of the tennis court we have
to refer to the people who made it, whose existence was in
44
turn dependent on that of their parents, and so on back
down the generations and down the evolutionary stream; and
to explain the existence of the site and the materials we have
to refer to the structure of the earth, and then t o its
formation, and then to the formation of the solar system.
These explanatory regresses, taken far enough, meet in the
complex fact of the spatio-temporal universe as a whole. But
this also is not a self-explanatory phenomenon. It is not self-
evident that matter, with the properties that it has, must
exist or that it must be ordered as it is. The existence of the
physical universe with its particular structure, although so
enormously vaster and more complex a fact, is as much in
need of explanation as was the existence of the tennis ball on
the road. And if we suppose, as we well may, that the
universe is both spatially and temporally unlimited and has
had no beginning, its existence with the concrete character
that it has is still not explained. It is merely pointed to as a
uniquely comprehensive fact. There is no evident reason why
the universe should not consist of empty space; or of half, or
twice, as much matter as it does; or why such laws as the
conservation of energy or such basic characteristics as the
speed of light should obtain. The existence of the universe —
that is, the existence of space-time structured as it is, and of
matter with the properties that it has — is a sheer given
'brute' fact. And to apply to it the principle of sufficient
reason is to see it as a contingent fact, pointing beyond itself
for the ground of its intelligibility.
If then — the cosmological argument claims — the
existence of the universe is an ultimately intelligible fact, it
must be so by reference to a reality whose existence and
character is self-explanatory and whose relation to the space-
time universe provides a sufficient reason for the latter's
existence. And this, as Aquinas says at the end of several of
his 'Ways', is what we mean by God.
In the third Way passage itself there is another turn t o the
argument before this conclusion is reached. Having shown
that there must be a necessary being, Aquinas adds, 'Now a
thing that must be, may or may not owe this necessity to
something else. But just as we must stop somewhere in a
series of causes, so also in the series of things which must be
and owe this to other things.' Thus he thinks it possible for
45
there to be a number of necessary beings. What does he have
in mind? The answer is that angels, human souls and the
heavenly bodies were for Aquinas necessary beings in virtue
of the fact that they were by nature immortal. But whilst
Aquinas calls these things necessary, nevertheless they clearly
do not possess the unqualified necessity of the divine nature.
Although Aquinas does not put the distinction in this way,
one respect in which they differ from God is that whereas he
exists eternally, they only exist sempitemally — i.e. they have
a beginning but (apart from divine annihilation) no end.
However the introduction of these sempiternals does not
affect the outcome of the third Way argument; for they
depend for their existence upon a creator, and if this creator
is himself sempiternal he too must depend upon a creator
. . . and the regress ends in a being w h o is necessary in the
absolute sense of existing without beginning or end. It is this
that 'all men call God'.

(b) The self-explanatory and the non-self-explanatory

Understood in this way the argument hinges upon the claim
that the space-time continuum, as a contingent, non-self-
explanatory phenomenon only becomes intelligible when
seen in relation to an eternal self-existent being who has
established it. Let us call this claim, which is the nerve of this
family of arguments, the cosmological principle. More briefly
it is the premise that the existence of God would be self-
explanatory, whereas the existence of the physical universe is
not. But this principle can be and has been challenged. Would
not an eternal self-existent deity constitute as sheerly and
starkly 'given' a fact as an eternal realm of matter or energy?
Antony Flew insists that 'Facts about God, however
important, do not thereby cease to be, simply, facts. . . . No
reason whatever has yet been given for considering that God
would be an inherently more intelligible ultimate than — say
— the most fundamental laws of energy and stuff.' (18) I
think that Flew is right in complaining that cosmological
arguers have often been content to treat as intuitively evident
the principle that the existence of an eternal creative Mind
would be self-explanatory in a way in which the existence of
the physical universe, exhibiting the fundamental laws which
46
it does exhibit, would not. He is right to want to have the
grounds of this principle produced. And indeed the
exploration of these grounds is perhaps the most interesting
part of an examination of the cosmological argument today,
when its inner principle has been explicitly questioned. I
think it will appear, on the one hand, that the cosmological
principle is not capable of demonstrative proof but, on the
other hand, that it has a certain inevitability as an expression
of the fact that to us as conscious beings the fact of
conscious mental existence is not a candidate for explanation
as is the fact of the physical universe. It is only with an effort
that we can temporally suspend this very natural prejudice
built into our nature. It has a status comparable with that of
the 'natural beliefs' — such as our acceptance of the indepen-
dent existence of the perceived world — to which Hume drew
attention. And accordingly a defence of the cosmological
principle will consist in displaying its deep roots in this
ground of natural or commonsense judgement.
But first it will be well to formulate more precisely the
issue to which this principle is relevant. It is not a response to
the very peculiar problem (perhaps a product of language
when it is idling), Why is there something rather than
nothing? For that there is something is a circumstance than
which there could be none more ultimate by which to
explain it. We can only accept as a given starting-point for
thought the fact that 'something is'. But whilst to seek an
explanation of the fact that anything at all exists would be to
seek for what cannot be, to seek an explanation of the fact
that what exists is as it is rather than otherwise is a logically
permissible project. Indeed the assumption that there is
always a reason (whether or not we can know it) why the
universe or any bit of it is at a given time thus rather than
otherwise has been dignified with the name of the Principle
of Sufficient Reason. (19) This principle cannot be demon-
strated; b u t it is presupposed by so many of the processes of
thought which we call rational as itself to count as a funda-
mental principle of rationality. To reject it seriously and in
practice would probably lead to one being certified as insane,
and certainly to one's opting out of all forms of rational
inquiry. To operate the principle of sufficient reason, then, is
to look for explanations; and we must now notice how this
47
process of explaining proceeds. To explain a puzzling
phenomenon is to set it in a wider context, or in connection
with some further circumstance, in relation to which it is no
longer puzzling. But one may of course then be puzzled by
this explanation-providing situation, and consider how it in
turn is explained. However in such a process something must
in the end be left unexplained, namely that ultimate state of
affairs which is not related to anything yet more ultimate, by
which it might itself be rendered intelligible. If there is some-
thing (such as the universe) so comprehensive that it can have
no wider context, or something (such as a creator of every-
thing that exists other than himself) of such a nature that
there is nothing that could stand in an explanatory relation
to it, then this thing will de facto terminate the explanatory
process.
Now as de facto ultimates, God and the physical universe
enjoy an equal status; and observing this fact, many critics of
the cosmological argument have concluded that it lacks all
force. But we are only now approaching a view of the special
feature of our situation to which the cosmological argument
points. For it alleges that the idea of God provides a de jure
as well as a de facto terminus to the explanatory process.
That is to say the idea of God is the idea of a self-explanatory
reality, or better, of an entity which is ultimate within a kind
(namely consciousness) which we, being ourselves of this
kind, accept as not requiring explanation in terms of any-
thing of a different kind.
Thus the two contentions involved in the cosmological
principle are (a) that the existence of the physical universe
itself, with its fundamental laws of energy, is not something
which anyone can on reflection regard as self-explanatory,
and (b) that the idea of God can very naturally and
reasonably be regarded as exhibiting self-explanatoriness.
Both of these contentions must now be given supporting
commentary.
As regards the first, we can conceive that the universe
might have been very different. The fundamental laws of
matter might have been otherwise. The law of gravitation, for
instance, which has contributed to making possible the
evolution of matter to produce solar systems and life, might
not have obtained. Indeed there might have been no universe
48
at all in the sense of an orderly arrangement of matter. And
because we can conceive of matter-energy displaying a funda-
mentally different character, we are left with the question —
whether or not we can answer it — as to why matter is in fact
disposed as it is rather than otherwise.
Now in the application of the principle of sufficient reason
it is quite acceptable, and indeed by no means unusual, to
cross the category boundary between the mental and the
physical. We frequently accept as valid a reference to minds
and their purposes and activities in explanation of the
dispositions of matter. For example, when we see a moving
object — say a car travelling along a road, — if we find that
we are able to explain its movement by reference to human
intelligence and volition we do not look any further. We take
it that we have arrived at a real explanation of the fact that
the car is moving along the road. (Not of how it is moving —
by what system of combustion in cylinders connected with
rods and wheels — but of why all this is going on, rather than
not going on, in that place at that time.) In this case we treat
mind as a reality of a higher order in the explanatory
hierarchy than matter. And indeed, as minds, we inevitably
generalise this principle. It is true that we know of plenty of
instances in which the prior behaviour of matter determines
the dispositions of mind — as when a rock hits someone on
the head and renders him confused or unconscious. But this
does not vitiate the general priority of mind over matter,
from the point of view of mind, in the explanatory hierarchy.
In the case of our car driver we may of course pursue the
explanation indefinitely further within the realm of con-
sciousness and intention. We can ask why he was driving
along that road at that time; and if it was in order to keep a
business appointment we can ask why he wanted to do this;
and so on. But we shall not thereby repeat the kind of move
that we made when we turned from matter to mind for the
explanation of the car's being where it was and moving as it
did. We shall not be appealing to a reality of a yet higher
order than mind in the explanatory hierarchy. Indeed we
cannot conceive of any further order of reality in terms of
which mind might itself be explained. We can readily
conceive of superior minds to ourselves; b u t not of kinds of
reality superior to mind. Thus there is for us an explanatory
49
ultimacy about mind which we do not find in the existence
or the laws of matter. As minds we do not ask why there
should be any such thing as mind, although we do ask why
there should be any such thing as matter obeying the
particular laws which we find matter to obey.
This explanatory ultimacy of mind for minds, or inevitable
prejudice of mind in its own favour as an intrinsically
intelligible kind of entity, may be said to place the cosmo-
logical principle among the 'natural beliefs' to which
humanity tends spontaneously to assent. As conscious minds
we can accept the existence of purposive intelligence as an
ultimate fact, neither requiring nor permitting explanation in
terms of anything more ultimate than itself. We can
accordingly conceive that the structuring of matter as it is
and the consequent course that its evolution has taken are
explicable by reference to a creative divine Mind; and no
further question then arises as to why that divine Mind
should exist. As minds, we can rest in the thought of an
eternal and infinite self-existent Mind behind the contingent
phenomena of a physical universe within which our own
finite minds have emerged.
This then is the kind of support — not demonstration — of
which the cosmological principle is capable. Although no one
is logically obliged to accept it, yet the principle is so entirely
natural an expression of man's own self-awareness that to
adopt it cannot be regarded as in any way irrational. There is
thus, for many people at least, force in the dilemma proposed
by the cosmological argument: either the existence of the
universe, ordered as it is, is explicable by reference to God, or
it is not explicable at all but remains a sheer brute fact fruit-
lessly provoking the question, Why should it be? Hence it is
both a logically permissible and a very natural view that if the
existence of the universe, as an ordered cosmos, is ultimately
explicable or intelligible it must be so in virtue of its
dependence upon an eternal self-existent reality which is of
the same order as conscious mind.
It seems to me — differing here from Flew — that an
atheist could perfectly well accept this cosmological
principle. That is to say, he could agree that if there is an
eternal, self-existent creative Mind responsible for the
existence of the physical universe, with the concrete
50
character that it has, then the question, Why does the
universe exist? would have an answer, and an answer in terms
of something which we can accept as having being in its own
right and not requiring araison d'etre referring beyond itself.
But having agreed that the universe is either unexplained or is
to be explained theistically, he would add that there is no
reason here to adopt the latter alternative. There is no
adequate reason to d o other than accept the universe as
simply an ultimate inexplicable datum. For whilst the cosmo-
logical argument presents us with the options: universe as
brute fact or as divine creation, it does not provide any
ground for preferring one to the other.
And of the two the more economical option is the
atheistic one. Thus in a celebrated debate between Bertrand
Russell and F. C. Copleston, in which a cosmological
argument and an atheistic response to it are clearly displayed,
Copleston urges that if there is no God the universe is (as the
atheist existentialist, Sartre, said) 'gratuitous'; and Russell,
rejecting the term 'gratuitous', replies T should say that the
universe is just there, and that's all.' (20) And with this
acceptance by Russell of the world itself as an ultimate
datum their dialogue reaches an impasse. In another
discussion of the argument from contingency, Copleston
acknowledges in effect that Russell's position is unassailable.
He says:
One can hardly admit that the existence of a finite being at
all constitutes a serious problem and at the same time
maintain that the solution can be found anywhere else
than in affirming the existence of the transfinite. If one
does not wish to embark on the path which leads to the
affirmation of transcendent being, however the latter may
be described (if it is described at all), one has to deny the
reality of the problem, assert that things 'just are' and that
the existential problem in question is a pseudo-problem.
And if one refuses even to sit down at the chess-board and
make a move, one cannot, of course, be check-mated. (21)
And this, it seems to me, is where the cosmological argument
leaves us. It points very clearly to the possibility of God as
the ground of the ultimate intelligibility of the universe in
which we find ourselves, and of ourselves as part of it. But in
C H.A.E.C. 51
doing this it does not constitute a demonstration of God's
existence. It leaves us with the alternatives that the universe
is an inexplicable brute fact, or that its existence with the
structure that it has is intelligible in the only way in which it
could ever finally be intelligible to us, namely through its
dependence upon a reality that is ultimate in the order of
mind.

52
4 Moral Arguments

(a) God as a postulate of practical reason

The modern attempt to show that God is known through or

is implied by man's moral experience begins with Immanuel
Kant, who in his 'Critique of Practical Reason' argued that
the existence of God is a postulate of the practical or moral
reason. In his 'Critique of Pure Reason' Kant had attacked
the prevailing rationalist view, as represented b y such
thinkers as Leibniz and Wolff, that the reality of God b o t h
requires and is capable of logical demonstration. Kant
examined the traditional theistic arguments and maintained
that the cosmological and ideological presuppose the onto-
logical proof and cannot stand if it falls; and he subjected the
ontological proof to powerful and penetrating criticisms
which have ever since been widely regarded as fatal to it (see
pp. 81-2). Thus for Kant the existence of God is not some-
thing that in the strict sense of the term we can know. This
does not however mean that we are not entitled to believe it.
For as well as being a theorising intellect man is a moral
agent, in touch with reality as it impinges upon him as an
ethical being. When the practical reason finds it necessary to
believe something Kant does not speak of knowledge but of
faith (Glaube); and one outcome of the critical inquiry by
which he defined the limits of theoretical reason was to 'deny
knowledge, in order to make room for faith.' (1) This Glaube
takes the form of the postulation by practical reason of
certain metaphysical facts which are presupposed in its own
operation: namely human freedom, human immortality and
the existence of God.
How does Kant arrive at this conclusion in the case of the
existence of God?
It is a fundamental principle of his ethical thinking that 'It
is impossible to conceive anything at all in the world, or even
out of it, which can be taken as good without qualification,
except a good will'. (2) But whilst the good will, or virtue, is
53
the only thing that is intrinsically good, in that it is good in
all possible circumstances, and is thus the highest good in one
sense of that term (summum in the sense of supremum), it is
nevertheless not the highest good in the sense of the perfect
or most complete good (summum in the sense of con-
summatum). This latter, identified by Christianity with the
Kingdom of God, (3) would be the best possible state of
affairs, and must as such consist in more than moral good-
ness. For if all men were virtuous but were also in pain and
misery, their virtue would still be intrinsically valuable but
nevertheless the total situation would not be the best
possible. The summum bonum, in its most comprehensive
sense, would still remain unrealised. That requires not only
moral goodness but also the crowning of it with happiness.
'For to be in need of happiness and also worthy of it and yet
not to partake of it could not be in accordance with the
complete volition of an omnipotent rational being. . . .' (4) In
other words, reason approves as the only rationally satisfying
state of affairs one in which moral goodness exists and is
appropriately correlated with happiness.
Kant does not argue directly from these premises to the
existence of God, but claims that divine existence is a
postulate, or presupposition, of the claim of morality upon
us. His reasoning can be set out as follows.
First, the ultimate object of a good will — that which a
good will seeks to bring about as far as it is able — is the
summum bonum, or perfect state of affairs: 'the highest good
is the necessary highest end of a morally determined will and
a true object thereof. (5) In short, it is 'our duty to promote
the highest good'. (6)
Second, since 'ought' implies 'can', it follows from the fact
that it is the good will's duty to try to bring about the
highest good that it is possible for that highest good to come
about. Thus it is 'a necessity connected with duty as a
requisite to presuppose the possibility of this highest good'.
(7)
Third, although the realisation of the highest good must
thus be possible, it is not within our power as finite beings to
bring it about. We are free to achieve virtue in ourselves, but
not to ensure that happiness is added to virtue, thus realising
the perfect state of affairs.
54
 Fourth, there must consequently be a rational and moral
being who as creator and ruler of the world has the power to
bring moral desert and happiness into harmony with one
another. Therefore 'the existence is postulated of a cause of
the whole of nature, itself distinct from nature, which
contains the ground of the exact coincidence of happiness
with morality'. (8)
Further, since the apportioning of happiness to virtue does
not take place in this life it must be attained in eternity, (9)
and so the postulate of immortality (for which Kant has
already argued on the ground that the obligation to attain
perfect goodness can only be met by an infinite progress
towards perfection, which in turn requires infinite time for
its occurrence) is closely connected with the postulate of
divine existence.
The exact status of these postulates (together with the
third postulate of human freedom) in Kant's thought is not
easy to determine. They do not constitute knowledge, i.e.
they are not established conclusions of theoretical reason;
although on the other hand the theoretical intellect has no
reason to resist or reject them. (10) What, I think, Kant
means is that to take our ethical nature fully seriously,
accepting moral obligations as having objective and binding
validity, is to presuppose that the universe in which we exist
has a particular character, the basic features of which are
indicated by the metaphysical postulates of God, freedom
and immortality.
However Kant's reasoning, as regards the postulate of
divine existence, is open to criticism. It is a turning-point of
his argument that the summum bonum (consisting of virtue
rewarded with happiness) is possible, and that in order for it
to be possible God must exist as a moral and omnipotent
Being. But what does 'possible' mean in this context? It may
merely mean 'logically possible'. But in order for the
summum bonum to be possible in this sense it is not
necessary either that the ideal state of affairs should ever
actually come about or therefore that a Being should exist
with the power to bring it about. All that is required is that
the concept of the summum bonum be not self-
contradictory. Thus the mere logical possibility of the
summum bonum cannot require us to postulate divine
55
existence. Accordingly Kant's argument demands that when
we affirm the summum bonum to be possible we are
affirming it to be factually (and not merely logically)
possible.
Now one ground on which, according to Kant, a state of
affairs can be known to be factually possible is that someone
is under a moral obligation to bring it about. For 'ought'
implies 'can'; so that if I ought to create a certain state of
affairs it follows that I can create it, and therefore that it can
exist. The question, then, is, Who is under a moral obligation
to realise the summum bonum? Clearly we cannot at this
point suggest God as the answer; for it is God whose
existence we are seeking grounds for affirming. It must then
be man. But it is an essential part of Kant's argument that
man himself does not have the power to bring about the
summum bonum — for which reason we have to postulate
God. Man's obligation is to do all that he can towards the
realisation of the summum bonum. But the summum bonum
contains two distinct elements — the existence of good wills,
and the proportioning of happiness to desert — such that
either could be realised without the other. It could be the.
case that there is moral goodness, but no correlation between
the good will and good fortune: and conversely it could be
the case that good fortune is distributed according to moral
desert but that no good wills exist. And our obligation to do
all we can to realise the summum bonum is an obligation to
become good wills; from which it follows that it is possible
for us to do so. We are not however under obligation to bring
about the second part of the summum bonum, for this is not
within our power; and therefore there is no implication, from
any obligation lying upon us, concerning the factual
possibility of this second part or, accordingly, of the
summum bonum as a whole. Nor therefore is there any
proper ground in our moral duty for postulating the
existence of God as the agent necessary to bring about the
summum bonum.

(b) The post-Kantian type of moral argument

In Idealist philosophy and theology since Kant the moral

56
argument for the existence of God has been deployed in
various ways b y a large number of thinkers. We may take as
representative the version offered by Hastings Rashdall in his
'The Theory of Good and Evil' (1907). Rashdall freely
recognised that a man might be moved by a moral ideal, and
might live by its behests, whilst forming no beliefs, or only
negative beliefs, concerning the ultimate source and status of
that ideal. But, Rashdall says, 'the question arises whether,
when the attempt to harmonize and so to justify our beliefs
is honestly made, the man who wishes to defend and
rationalize his practical recognition of moral obligation may
not be forced into the alternative of giving up his ethical
creed or of giving up certain views of the Universe which
reflection has shown to be inconsistent with that creed.' (11)
The starting-point of Rashdall's argument is thus the
'absoluteness' or 'objectivity' of moral obligation. This, he
assumes, is something which the agnostic will agree in
affirming. 'The truth', he says, 'that the moral ideal is what it
is whether we like it or not is the most essential element in
what the popular consciousness understands by "moral
obligation". Moral obligation means moral objectivity. That
at least seems to be implied in any legitimate use of the term:
at least it implies the existence of an absolute, objective
moral ideal.' (12) The problem, then, is how to conceive the
nature of the universe consistently with the reality of
objective moral values and laws. At this point Rashdall
launches his main argument:

We say that the Moral Law has a real existence, that there
is such a thing as an absolute Morality, that there is some-
thing absolutely true or false in ethical judgements,
whether we or any number of human beings at any given
time actually think so or not. . . . We must therefore face
the question where such an ideal exists, and what manner
of existence we are to attribute to it. Certainly it is to be
found, wholly and completely, in no individual human
consciousness. . . . Only if we believe in the existence of a
Mind for which the true moral ideal is already in some
sense real, a Mind which is the source of whatever is true in
our own moral judgements, can we rationally think of the
moral ideal as no less real than the world itself. Only so
57
 can we believe in an absolute standard of right and wrong,
which is as independent of this or that man's actual ideas
and actual desires as the facts of material nature. The
belief in G o d . . . is the logical presupposition of an
'objective' or absolute Morality. A moral ideal can exist
nowhere and nohow but in a mind; an absolute moral ideal
can exist only in a Mind from which all Reality is derived
(13) Our moral ideal can only claim objective validity in so
far as it can rationally be regarded as the revelation of a
moral ideal eternally existing in the mind of God. (14)

Clearly the argument hinges upon the 'objectivity' of

moral values and laws, that is to say, upon their indepen-
dence of the human minds which become conscious of them.
If moral values do indeed have such a status then Rashdall's
argument is well on the way to its conclusion, which in effect
gives the name 'God' to the independent source and ground
of ethics! But the 'objectivity' of morality is not the un-
controversial premise that Rashdall seems to have assumed.
On the contrary, the agnostic (or humanist, to use the term
more generally preferred today) denies it and offers instead a
purely naturalistic account of the origin and nature of moral
values. We shall look at a particular naturalistic ethical theory
— that of Bertrand Russell — presently. But nearly all such
theories follow a common basic pattern. Man, it is pointed
out, is a gregarious animal; and human beings cannot live
together successfully in societies without obeying rules of
behaviour which harmonise their interests. Such rules,
gradually developed through the experience of group
existence, are internalised by social conditioning in the form
of moral ideals and obligations. They owe their felt
'absoluteness' and 'objectivity' to the tremendous power and
authority of the clan over the individual members whose
moral outlook and habits it has moulded from birth. This
accounts for the feeling that conscience represents a divine
'voice' whose commands and prohibitions we must obey — a
feeling which various nineteenth-century thinkers, in
particular, developed into a form of moral argument for the
existence of God. Cardinal J . H. Newman, for example,
wrote: 'If, as is the case, we feel responsibility, are ashamed,
are frightened, at transgressing the voice of conscience, this
58
implies that there is One to whom we are responsible, before
whom we are ashamed, whose claim upon us we fear.' (15)
But from the naturalistic point of view Newman was mis-
reading the internalised requirements of our social nature,
generated entirely by the circumstances of the life of the
human animal, as the commands of a supernatural Being.

(c) A reformulated moral argument

We have not thus far found any proof moving from man's
moral experience to the existence of a supreme being. Nor, I
believe, is it possible to find one. But nevertheless the fact of
moral obligation, even on a naturalistic meta-ethical analysis,
can, I think, present a fatal challenge to a humanist
philosophy; and I should like now to formulate this
challenge. I shall suggest that a humanist who performs an act
of extreme self-sacrifice for the good of humanity is thereby
presupposing the falsity of his professed beliefs. From the
point of view of his creed — I shall argue — his action is
irrational; whilst in relation to his action his creed is
inadequate.
Let us consider the case — the admittedly extreme case —
of the humanist who knowingly sacrifices his life for the sake
of humanity as a whole. We are thus thinking of self-sacrifice
on moral principle rather than of the rather different case of
the impulsive self-giving of one who loses his life whilst, for
example, trying to save a child from drowning; or the
different case again of the soldier who risks his life in the
belief that he is protecting himself and his family and
community from some dire threat to their liberty and values.
And the question that I wish to pose is how such con-
scientious self-sacrifice — not for family, nation or any
in-group, but for mankind at large — can be defended in
humanist terms as a rational or reasonable act. I am not
asking whether humanists do in fact sacrifice their lives for
such reasons of conscience more or less often than religious
believers; or what the nature of their motivation is; but
simply how their action is rationally defensible within the
framework of humanistic belief. For the humanist would not,
presumably, claim that such self-sacrifice is merely an
59
irrational aberration; he would wish it to be seen as a
consistent outcome of a distinctively humanist policy for
living.
We need at this point to have before us a statement of the
humanist ethic, and for this purpose I turn to Bertrand
Russell's 'Human Society in Ethics and Polities'. (16) The
major points of his theory are:
1. The notions of good and bad, or good and evil, arise
from the fact that human beings are conscious of desires and
aversions. If all experiences were equally welcome to us we
should have no occasion to distinguish between them from
the point of view of the desirability of their occurrence. But
our nature is such that we seek some things (for example,
pleasure) and shun others (for example, pain). 'I suggest',
says Russell, 'that an occurrence is "good" when it satisfies
desire, or, more precisely, that we may define "good" as
"satisfaction of desire". One occurrence is " b e t t e r " than
another if it satisfies more desires or a more intense desire'
(p. 55).
2. The ideas of right and wrong are dependent upon those
of good and bad: "right" conduct is that which, on the
evidence, is likely to produce the greatest balance of good
over evil or the smallest balance of evil over good. . . . [The]
sum-total of moral obligation is contained in the precept that
a man ought to do right in the above sense' (p. 50).
3. Given that good is to be defined as satisfaction of desire,
'The general good will be the total satisfaction of desire, no
matter by whom enjoyed. The good of a section of mankind
will be the satisfaction of the desires of that section, and the
good of an individual will be the satisfaction of the desires of
that individual' (p. 60).
4. Our desires are not necessarily all self-centred. 'Most
people desire the happiness of their children, many that of
their friends, some that of their country, and a few that of all
mankind' (p. 56).
5. Morality, central to which is the idea of an obligation to
act rightly, has come about because we do not spontaneously
seek the good of our group or of mankind as a whole. Ethics,
in fact, is 'part of an attempt to make man more gregarious
than nature made him' (p. 129). Thus, 'One may lay it down
broadly that the whole subject of ethics arises from the
60
pressure of the community on the individual. Man is very
imperfectly gregarious, and does not always instinctively feel
the desires which are useful to his herd. The herd, being
anxious that the individual should act in its interests, has
invented various devices for causing the individual's interest
to be in harmony with that of the herd. One of these is
government, one is law and custom, and one is morality' (p.
124).
6. Russell recommends the ethical principle that the good
which everyone ought to seek is the general good. 'When A
says to B, "You ought to do X", I shall define the word
" o u g h t " as meaning that, of all acts that are possible for B, X
is the one most likely to further the interests of mankind, or
of all sentient beings' (p. 125). Consequently he can speak of
'the stock of all those mental goods which distinguish man
from the ape and civilized man from the savage', and can say
of them, 'It is these things that make the unique importance
of man, and it is of these things that each generation in turn
is the trustee. To hand on the treasure, not diminished, but
increased, is our supreme duty to posterity' (p. 137).
To set this ethic within the context of the humanist
conception of man's place in the universe I shall again quote
Bertrand Russell. This humanist picture, which Russell
eloquently articulated, depicts man as simply an intelligent
and gregarious animal, destined to perish like the sheep and
the grass, so that the moral and spiritual values which move
him, and which form the basis of 'our supreme duty to
posterity', have no status except as modifications of our con-
sciousnesses, and are doomed to extinction with the
extinction of these consciousnesses. As Russell wrote in a
famous essay:

That Man is the product of causes which had no prevision

of the end they were achieving; that his origin, his growth,
his hopes and fears, are but the outcome of accidental
collocations of atoms; that no fire, no heroism, no
intensity of thought and feeling, can preserve an individual
life beyond the grave; that all the labours of the ages, all
the devotion, all the inspiration, all the noonday brightness
of human genius, are destined to extinction in the vast
death of the solar system, and that the whole temple of
61
 Man's achievement must inevitably be buried beneath the
debris of a universe in ruins — all these things, if not quite
beyond dispute, are yet so nearly certain, that no
philosophy which rejects them can hope to stand. Only
within the scaffolding of these truths, only on the firm
foundation of unyielding despair, can the soul's habitation
henceforth be safely built. (17)

Russell's view then is that each individual automatically

seeks his own good, which consists in the satisfaction of his
desires; and that aspect of our nature which underlies the
family and the propagation of the species leads many to
include in the scope of their own good that of their spouse
and of their offspring, at least when the latter are young and
dependent. But the problem I am raising only comes into
view when we consider the further extension of one's good
far beyond this natural biological unit to the human species
as a whole. For nothing less than this is required in the kind
of case I have postulated, in which an individual voluntarily
sacrifices his life for the sake of mankind. Let us now specify
this case more fully by adding that his sacrifice is for the sake
of future generations and that he has no offspring of his own
through whom he might feel a more or less direct stake in the
welfare of later generations. Such self-sacrifice for the good
of mankind can only be the product of morality — of that
device whereby the herd induces the individual to sub-
ordinate his own interests to those of the group. ('The herd,
being anxious that the individual should act in its interests,
has invented various devices for causing the individual's
interest to be in harmony with that of the herd. One of
these . . . is morality': 'Human Society in Ethics and Polities',
p. 124.) The definition which Russell offers of the key moral
notion of 'ought' or obligation, in terms of action in the
interests of mankind as a whole, is thus itself simply a part of
this device: in producing such a definition Russell is acting,
whether consciously or not, as an instrument of the herd in
disciplining its members.
Now let us grant that morality, as the psychological
mechanism by which individuals are caused to serve their
group, has its roots in the social side of man's nature. As
Margaret Knight reminds us in a recent exposition of
62
humanist ethics, 'Animal psychology in recent years has
undergone a minor revolution through the growth of
ethology — the study of animal species in their natural state.
One result of this development has been a growing realization
of the extent to which gregarious animals exhibit co-
operative, altruistic behaviour, n o t only towards sexual
partners and offspring, but towards other members of what
can reasonably be called the community.' (18) In the more
gregarious of the lower forms of life, such as the ant, instinct
plays the role played by morality in human life. The
individual ant, we are told, fulfils his function automatically
in the life of the ant-hill, even to the extent, if circumstances
require it, of destroying himself for the preservation of the
group. However, whereas the ant is moved to self-sacrifice by
instinct, and so exercises no choice in the matter, man is able
rationally to criticise the mechanism of conscience and to
accept or reject its promptings. Imagine for a moment an ant
suddenly endowed with the knowledge contained in Russell's
book and with the freedom to make his own personal
decisions. Suppose him to be called upon to immolate
himself for the sake of the ant-hill. He feels the powerful
pressure of instinct pushing him towards this self-destruction.
But he asks himself why he should voluntarily embrace this
fate. There is a cause that may lead him to do so, namely the
force of instinct. But is there any reason why, having the
freedom to choose, he should deliberately carry out the
suicidal programme to which instinct prompts him? Why
should he regard the future existence of a million million
other ants as more important to him than his own continued
existence? After all, they are all ants, all but fleeting
moments of animation produced and then annihilated by
mindless and meaningless forces — for that an ant 'is the
product of causes which had no prevision of the end that
they were achieving; that his origin, his growth, his hopes and
fears, are but the outcome of accidental collocations of
atoms . . .' is so nearly certain that it would be irrational to
act on any other assumption. Since all that he is and has or
ever can have is his own present existence, surely in so far as
he is free from the domination of the blind force of instinct
he will opt for life — his own life.
Now why should a humanist choose differently?
63
 He experiences a tension between on the one hand the
natural valuation of life by a living creature, so that he desires
to continue to live, and on the other hand the claim of moral
obligation, which presents itself to him as an absolute
demand summoning him to sacrifice his life. For conscience
characteristically takes the form of an unconditional claim
upon us of such a kind that nothing else whatever, not even
life itself, can legitimately outweigh it. Morality confronts
our wills as, in Kant's phrase, a categorical imperative. And
only such an absolute claim, against which nothing else can
properly be set in the balance, could rationally lead one
voluntarily to sacrifice one's life. Thus, in giving his life for
purely ethical reasons anyone, whether humanist or religious
believer, is implicitly acknowledging a moral claim upon him
that is entitled to override even the basic natural desire for
self-preservation. But in the case of a humanist this ack-
nowledgement involves a profound self-contradiction. For he
believes that morality is simply a device whereby the herd
subordinates to its interests those of the individual. Thus
from the point of view of his practice his official belief is
false; and from the point of view of his belief his practice is
irrational.
The irrationality arises from the fact that within the terms
of the humanist philosophy it is unreasonable for anything to
be of more value to a man than his own existence. For
according to humanism we are simply complex animal
organisms sustaining for some seven or eight decades an
intermittent thread of consciousness. We have and are
nothing but our own existence as consciousnesses subject to a
series of agreeable and disagreeable modifications through
time. To give this up is to give up everything, to be nothing,
simply not to be. But we cannot rationally desire not to be so
long as it is true for us that it is good to be alive. (19) On
humanist principles no possible object of desire could, on a
rational calculation, be worth to me the price of my own
existence. There is nothing that can substitute for my own
existence so as to give me either equal or greater satisfaction;
for without my own existence nothing has the power to give
me any satisfaction at all.
Let me put this point again slightly differently.
According to humanism there are simply individual
64
consciousnesses, who experience desires, and to whom any-
thing that they desire but do not have, or have but would
desire if they did not have it, is by definition good. That is to
say, goodness is a correlate of either unfulfilled or fulfilled
desire, and the attaining or retaining of the good, together
with the avoidance of its opposite, comprises the motivation
of human action. From the point of view of such a con-
sciousness, his own existence is a necessary condition for
anything being good to him and thus producing a motive for
him to act. Now suppose that as a result of moral
conditioning he finds that he desires there to be a state of
affairs (the well-being of mankind in the future) which he can
only help to bring about by prematurely ceasing to exist. He
now has an object of desire the existence of which is
incompatible with his own existence and thus with his having
any desires. It is a good which he can desire to have but
which in the nature of the case he cannot have. This paradox
must provoke him to question whether he can rationally
nurse such a desire. He can only conclude in the negative. For
he cannot rationally take as an object of desire a state of
affairs that would be incompatible not only with the grati-
fication of any other of his desires but even with the
gratification of this paradoxical desire itself. None of his
desires can be satisfied if he no longer exists; and it cannot be
the satisfaction of a reasonable desire of his that he should
have no desires or their satisfactions. Therefore if he is able
by the power of reason to discipline his emotions so as to
reduce or eliminate this desire, he will certainly do so. It
follows that the act of self-destruction for the sake of man-
kind is not an act that can be rationally chosen or defended
within the limits of the humanist conception of man.
But might not our hypothetical humanist calculate that if
he chose to preserve his own individual life at the expense of
the future welfare of mankind he 'would be unable to live
with himself — he would be so tortured by remorse that life
would not be worth having? And would not such a calcu-
lation provide a rational ground for his self-sacrifice? The
answer, I think, is that it would not. For we are supposing
the humanist to have 'seen through' the delusion of
conscience; he knows that it is merely a device whereby the
herd induces its individual members to subordinate their own
65
interests to it's. And having exposed this deception in his
own mind he will no longer be subject to it. He will not feel
any pangs of conscience after deciding to live out his own life
rather than give it up for the sake of the herd.
But there is still another possible reasonable ground on
which the humanist might make this sacrifice. (20) He might
consider that the pleasure to be gained from contemplating
the future good he is bringing about outweighs that to be
derived from continued life. For it can be rational t o prefer a
briefer but more intense pleasure to a longer but less intense
one. And if the humanist receives tremendous pleasure from
the thought of the happiness of generations as yet unborn, he
may feel amply compensated for the loss of another forty
years of his own life by the enjoyment of this benevolent
pleasure during the relatively short period prior to his self-
sacrificial death.
There are however counters to this reasoning. One is that
the period of time during which this pleasure is t o be enjoyed
might be very brief. It might be a matter of minutes or even
only seconds. Is it then realistic to think of chosing a few
seconds of pleasure, however intense, in preference to the
prospect of some forty years of ordinarily enjoyable life?
This seems a difficult question t o determine: can the
humanist be content to rely on a confident answer to it? We
must also consider, as a limiting case, the act of mortal self-
sacrifice for others which is an instantaneous response to
some emergency, allowing no time at all between decision
and act for the imaginative enjoyment of a future state of
affairs, or indeed for the making of a hedonic calculus at all.
Such a response would stem from the agent's existing basic
moral stance or life-policy, which it would spontaneously
embody in action in some moment of crisis. A humanist
would, I think, want to be able without contradiction to
admire and praise such a person. But if the argument of this
chapter has been well founded it does not seem that he can.
Let me at this point emphasise again that I have not been
discussing the psychological question whether a humanist
would be likely to give his life for an ethical reason relating
to the welfare of humanity as a whole; nor the comparative
question whether a humanist is more or less likely to do this
than a religious believer. I have been discussing the question
66
whether such self-sacrifice on a humanist's part could be
rationally defended on humanist principles. And the point
that I have emphasised is that whilst we share with some
lower animals social instincts which could, and indeed some-
times do, lead us to mortal self-sacrifice in the interests of the
species, we also have a strongly developed individual self-
consciousness and power of rational choice. And when we
exercise this power within the framework of the humanist
philosophy in a situation in which the social instinct
summons us to such self-destruction, our reason can only see
this urge as a deadly threat, to be resisted so far as possible.
Thus in so far as the power of reason reigns in the lives of
humanists they will never be guilty of the folly of ultimate
self-sacrifice for the sake of the human community. And in
so far as humanists think of such extreme self-sacrifice as
praiseworthy, and as representing a pinnacle of human
excellence, they are involved in a profound inconsistency.
The conclusion that I draw is of the same modest kind as
in the cases of the design and cosmological arguments. We do
not have here a proof of divine existence. But to follow out
the implications of our own moral insights and convictions is
to raise a question to which the answer may be — God. Is the
mortal self-sacrifice for the good of humanity, which our
moral nature prompts us to salute, rational or irrational? If
the naturalistic picture of the universe is correct, such action
is irrational and can occur only because men are not in the
last resort able to assert their reason against the power of
nurture, internalised as conscience. But if, on the other hand,
we trust our conscience, believing it to be rational so to do,
then we ought to disavow the naturalistic picture and move
in a direction which might in the end lead to belief in God.

67
5 The Ontological Argument:
First Form

(a) Introductory

The ontological argument is, philosophically, perhaps the

most interesting of the traditional 'theistic proofs', involving
as it does such fundamental concepts as perfection, deity,
existence and necessity — both logical necessity and the idea
of necessary being. It differs from all the other proofs in
being a priori, proceeding from the idea of God as infinite
perfection to his existence, instead of from some feature of
the perceived universe to God as the ground either of its
being or of its intelligibility. As an a priori argument it has
the form of a logical demonstration; and as such it either
totally succeeds or totally fails. It differs from the other
arguments also in the scope of the conclusion which, if
successful, it authorises. Whereas the other arguments profess
to show that there is a divine designer behind the orderliness
of nature, or a prime mover, first cause or necessary being
behind finite existence — leaving open the question to what
extent such a being possesses the moral attributes of deity —
the ontological argument professes to demonstrate the reality
of an unsurpassably perfect being. And it is more plausible, at
least, to claim that such ideas as goodness, wisdom and love
are contained within the notion of infinite perfection than
that they are contained within the notion of a first cause or a
necessary being. Indeed this was the ground of Kant's
contention that the cosmological and teleological arguments
both presuppose the ontological and cannot succeed if it
fails: only the latter professes to link the idea of necessary
and unconditioned reality with that of perfection. For given
that the cosmological argument has shown that there is a
necessary being, we can still ask what sort of a being this is.
What reason have we for thinking it to be God, the infinite
sum of perfections? Kant supplies a reason: 'The necessary
being can be determined in one way only, that is, by one out
of each possible pair of opposed predicates. It must therefore
68
be completely determined through its own concept. Now
there is only one possible concept which determines a thing
completely a priori, namely, the concept of the ens
realissimum. The concept of the ens realissimum is therefore
the only concept through which a necessary being can be
thought.' (1) But this reason uses the conclusion of the onto-
logical argument: that the idea of a supreme being, or ens
realissimum, necessarily entails the existence of such a being.
Only if this conclusion is true can the cosmological argument
amount to a proof of God's existence; but if that conclusion
is true God's existence is already proved and the cosmological
argument is entirely unnecessary. Thus the cosmological pre-
supposes the ontological argument, and is rendered otiose by
it. (2)
But whilst the ontological proof is thus the one which, if it
succeeded, would succeed most definitively, and the one
whose conclusion, if established, would be most worth
establishing, it is also the one which, in the opinion of most
philosophers, most definitively fails. Yet even in its failure it
is still, from a philosophical point of view, in many ways the
most rewarding of the proofs to study; and its fascination
shows no sign of failing even after nearly a thousand years of
intermittent discussion.
It will become clear as we proceed that the rather curious
term 'the ontological argument' (3) names a family — though
still a fairly closely knit familty — of arguments. And the
founder of this famous family, despite certain previous
stirrings of thought in a similar direction, was St Anselm.
Some writers have seen Anselm's reasoning prefigured in
Augustine's 'De Libero Arbitrio', II, and have even regarded
Augustine as the true originator of the ontological proof.
Augustine's argument for divine existence does indeed
approach Anselm's sufficiently closely for it to be altogether
likely that it was the latter's reading of the 'De Libero
Arbitrio' — in which, incidentally, the Fool who said, 'There
is no God' is used to introduce the task of proving divine
existence (II ii 5) — that first suggested to him the line of
thought which he was to pursue so much further and more
penetratingly than his master. (4)
A particular phrase that may well have passed directly into
Anselm's mind, ultimately expanding into his new theistic
69
argument, is Augustine's definition of God as that 'above
which there is no superior'. (5) It is likely that this formula is
a direct ancestor of Anselm's definition: 'that than which no
greater can be conceived'. But nevertheless, Anselm's
argument is not Augustine's: it is genuinely new, justifying
the great excitement with which Anselm greeted it when it
came to him in a flash of intellectual illumination comparable
with the moment when the law of gravitation dawned upon
Newton. (6) Augustine had argued in Platonic fashion that
out intelligence must recognise something superior to itself,
namely wisdom or truth; and either this is God, or, if there
be something superior to it, then this is God. (7) Anselm's
reasoning, however, as we must now proceed to see, was very
different.

(b) Anselm's Proslogion 2 argument

Anselm (1033-1109), Abbot of Bee in Normandy and later

Archbishop of Canterbury, was one of the most original and
one of the most logically rigorous of Christian philosophical
theologians. It was at Bee that in 1077-8 he wrote his
'Proslogion', a short work composed in the form of a prayer,
in which he seeks in a single movement of thought to
demonstrate both the existence and the nature of God; and it
is here, and in Anselm's Reply to his first critic, Gaunilo, that
his a priori argument for the existence of God is presented. It
has recently been stressed that Anselm offers what can either
be regarded as two arguments or two forms or stages of the
same argument, the one in 'Proslogion' 2 and the other in
'Proslogion' 3 and the Reply. I do not think that Anselm
himself regarded these as two different arguments; but never-
theless they have had widely different histories since his day
and have in that sense become different arguments. The first
argument (that in 'Proslogion' 2) was taken up by Descartes
and criticised by Kant and has frequently been discussed in
the histories of philosophy as 'the ontological argument',
whilst the second has come into prominence only within the
last decade or so, particularly through its advocacy,
independently, by Charles Hartshorne and Norman Malcolm.
I shall therefore discuss the two arguments separately before
70
offering some thoughts about the relation between them.
Anselm formulates the concept of God in his famous
definition: aliquid quo nihil mains cogitari possit, 'something
than which nothing greater can be thought'. (8)
Even an atheist (represented by the biblical Fool who
'saith in his heart, There is no God') (9) can have this concept
in his mind. Thus something-than-which-no-greater-can-be-
thought exists in intellectu — as concept, or in the mind. The
question is whether it exists not only in the mind b u t also in
external reality (in re). The proof that it does is as follows.
To exist in reality, as well as in the mind, is greater than to
exist only in the mind. Therefore if something exists only in
the mind it cannot be that than which no greater can be
thought. For we can conceive of something greater than a
'greatest conceivable being' which exists only in the mind,
namely that same thing existing also in reality. Therefore that
than which no greater can be thought cannot exist only in
the mind b u t must exist in external reality as well.
There are several points here which require commentary.
1. It is clear that by 'greater' Anselm does not mean
spatially larger (and indeed he does not think of God as
occupying space at all) (10) but 'more perfect'. Indeed he
sometimes uses melius, 'better', in place of maius, 'greater'.
(11)
2. The formula 'that than which no more perfect can be
conceived' is open-ended and not circumscribing. That is to
say, it does not tell us what God, as that than which no more
perfect can be conceived, is like. Thus Anselm insists later in
the 'Proslogion' that 'not only are You that than which a
greater cannot be thought, but you are also something greater
than can be thought.' (12) Accordingly the formula operates
negatively, forbidding us to identify as God any being such
that it is conceivable for this being to be surpassed in value.
(13) The idea of God, according to Anselm, is the idea of the
ne plus ultra in the dimensions (combined by neoplatonism)
of reality and value.
3. Although the formula 'that than which no greater can
be conceived' is in itself formal, and does not tell us what
particular qualities must belong to such a being, yet Anselm
believed that the theologian can in fact to some extent
describe the divine nature by listing those attributes which it
71
is better to have than to be without. 'What goodness, then,
could be wanting to the supreme good, through which every
good exists? Thus You are just, truthful, happy, and what-
ever it is better to be than not to be — for it is better to be
just rather than unjust, and happy rather than unhappy.' (14)
Indeed the unfolding of the nature of God from the aliquid
quo nihil maius formula by means of an intuitive knowledge
of those qualities which it is better to have than to lack
represents the programme of the 'Proslogion' as a whole.
Anselm does not deal with the question, which others have
since raised, whether the resulting conception is one that
could be instantiated. The possible existence of the unlimited
sum of perfections has been queried at two points. First, does
not existence necessarily involve limitation; and would not
unlimited being thus be indistinguishable from non-being? Is
there not, then, a contradiction in the notion of an infinite
existent? The answer usually given to this question is that
God does not 'exist' in the sense in which finite realities
exist, i.e. by being one entity among others. Rather, he is
'being itself; or he has actuality in distinction from
existence; or he is, b u t does not exist as an item within the
universe. Second, are t h e various perfections attributed to
God compatible with one another when all are multiplied to
infinity? (15) Justice and mercy, for example, seem to be
mutually limiting attributes. A certain degree of justice can
coexist with a certain degree of mercy; but is unlimited
justice compatible with unlimited mercy? There are other
aspects of the Christian concept of God which raise similar
problems, independently of those generated by multi-
plication to infinity: e.g. can God be perfect and self-
sufficient and yet a creator; can he be eternal and yet an
agent in time; can he be immutable and yet love his creatures
and sympathise with them in their sufferings? This is not the
place to treat the large subject of the concept of God and its
philosophical coherence. (On this see the volume in this series
on 'Concepts of Deity'.) We must be content here to con-
clude that Anselm's argument operates against what M . J .
Charlesworth (16) calls the 'factual atheist', who accepts the
issue of divine existence as a meaningful question, which he
answers in the negative, in distinction from the 'logical
atheist', who denies that the concept of God is a viable
72
concept in relation to which the question of existence or
instantiation can properly be discussed at all. (17)
4. Anselm assumes the principle that it is better to exist (in
re) than not to do so. 'For', he says, 'if it [i.e. that than
which a greater cannot be conceived] stands at least in
relation to the understanding, it can be conceived to be also
in reality, and this is something greater.'1 (18) But how, it
might be asked, do we know that it is greater or more perfect
to exist in reality than not to do so? Is not this an un-
supported assumption of Anselm's argument? And is not a
contrary assumption also possible — as for example in the
Buddhist rejection of empirical existence as evil?
There is indeed at this point a presupposed background to
Anselm's reasoning, namely the background of neoplatonism,
which reached Anselm both directly from Augustine and also
from an intellectual mileu much of which had continued
unchanged through the six hundred years that separated
Augustine and Anselm, and within which neoplatonism was
an important element. The neoplatonic picture of the
universe embodied the long-lived and influential equation of
being with goodness. The divine One is both absolute being
and absolute goodness, so that goodness and being are in
their ultimate source identical. (This pregnant thought doubt-
less goes back to Plato's teaching that it is the Good itself
which is the source both of being and of goodness in every-
thing else). (19) And they are likewise identical in our finite
world, which is an emanation of the One, an 'overflowing' of
the divine nature. The ultimate One radiates outwards to
produce the universe, which is thus an extension by
attenuation of what at its source is both Being itself and the
Good itself. Thus the descending levels of being embody
descending forms of goodness, less and less adequately
expressing the fullness of the creative Good, until they vanish
together into the non-being which is equated with evil. (20)
Christian thought, rejecting the idea of divine self-emanation
in favour of that of creation ex nihilo, nevertheless (and
perhaps inconsistently) retained the platonic idea of the
convertibility of being and goodness. Thus 'more perfect'
meant for Anselm, as for Augustine, 'more being-ful' or
'more real'. And within this framework of thought it is self-
evident that it is good to have being, since to have being is
73
ipso facto to have goodness.
Neoplatonism apart, the premise that it is greater or more
perfect to exist in reality than to exist only in thought would
seem to represent a necessary presupposition or prejudice of
consciously existing beings. For we should presumably not
remain voluntarily in existence if we did not in practice
accept this premise. Thus whether or not there is any further
sense in which this proposition may said to be true, it is one
to which existing free beings will always subscribe. It there-
fore seems permissible for Anselm to employ it in his
argument.
5. It is to be noted that Anselm's formula does not refer to
the most perfect being that there is, but to the most perfect
possible being. That 'the most perfect being that there is,
exists' is tautologically true. But 'the most perfect being that
there is' might or might not be God, i.e. worthy of man's
worship. The formula would leave open the possibility that
the most perfect existing being is, for example, a human
being and that there is nothing superior in value to man.
However Anselm's formula is significantly different: that
than which no more perfect can be thought, or is con-
ceivable.
6. In an interesting article in the special Supplement
devoted to the ontological argument in 'Religious Studies',
1968, M . J . A. O'Connor raises the question what force, if
any, the word 'thought' has in the formula 'that than which
nothing greater can be thought' or 'conceived' (cogitari).
Would the meaning of the phrase be different without it:
'that than which nothing greater can be'? O'Connor argues
that the cogitari adds nothing, or at least nothing that
Anselm wanted to add; for it if adds anything, 'it builds into
the definition relation to human thought as essential to God's
nature. (21) But in that case Anselm's definition would be
unacceptable. 'It would not define a transcendent necessary
being, but a contingent being whose existence would be
contingent on the existence of human thought, or whose
qualities could not transcend human conceptions — a being
who could not be greater than the greatest we can conceive.
The definition is only acceptable therefore, if we regard the
final word " t h o u g h t " as merely rhetorical, if we treat the
formula as meaning simply "something-than-which-nothing-
74
greater-can-foe" and the final word " t h o u g h t " as strictly
redundant.' (22)
The only effect of the final cogitari O'Connor suggests, has
been to obscure the all-important distinction between (a) the
existence in the mind of that-than-which-nothing-greater-
can-be and (b) the existence in the mind of the-zefea-of-that-
than-which-nothing-greater-can-be. The failure to focus upon
this distinction, according to O'Connor, has led b o t h Anselm
and many of his critics to assume that the Fool acknowledges
the ontological statement that that-than-which-nothing-
greater-can-be exists in the mind when in fact he only
acknowledges the psychological statement that the-idea-oi-
that-than-which-nothing-greater-can-be exists in the mind.
And whereas from the ontological statement we might be
able to proceed to the ontological conclusion that God exists
in reality as well as in the mind, from the psychological state-
ment we can only proceed to the conceptual conclusion that
the idea of God is the idea of an existing being.
This is an ingenious suggestion; but nevertheless I am not
convinced that it is correct.
As regards the alleged misunderstanding concerning what
the Fool acknowledges, Gaunilo — Anselm's first critic — did
indeed make this mistake, and Anselm promptly corrected it.
Gaunilo drew O'Connor's distinction in paragraph 2 of his
'On Behalf of the Fool', and assumed that Anselm's argument
requires that God somehow exists in the mind and not
merely that the idea of God exists in the mind. However
Anselm ('Reply', 6) corrects him, pointing out that his own
argument only requires as its starting-point that the idea of
'that than which nothing greater is conceivable' exists in the
mind. It therefore seems that the mistake which O'Connor
sees as pervading the history of the ontological argument was
scotched very early on, and certainly did not affect Anselm's
own reasoning.
Nor, I think, is the cogitari in Anselm's definition
dispensable. For in the shortened formula 'that than which
nothing greater can be' there is an ambiguity in the term 'can'
[posse). Is this a factual or a logical 'can'? Are we speaking of
a being than which there cannot in fact be anything greater,
because of surrounding circumstances of some kind in virtue
of which there is no 'room' for anything to exist beyond a
75
certain degree of 'greatness'; or are we speaking of a being
than which a greater is not conceivable, because not logically
possible? I think it is clear that Anselm intends the latter.
'That than which no greater can be thought' means that than
which no greater is conceivable, i.e. logically possible. And
Anselm differentiates this idea from that of the greatest that
is possible within a given context or structure of fact by his
use of cogitari. The cogitari is thus essential to convey the
full force of his definition of God.
7. We are concerned here with Anselm's argument from
the point of view of philosophy rather than of the history of
ideas, and are therefore interested in it as an attempted proof
of the existence of the greatest conceivable being. It should
however be said that there are many important historical
questions concerning Anselm's arguments, and many
different attempted answers. Did Anselm intend to offer a
proof of God's existence that ought to convince any rational
inquirer; or was he explicating the inner logic of faith; or was
he concerned to produce a mystical experience of the divine
presence? Did the neoplatonic background of his thought
mean that for him there was no problem of moving from an
idea to reality, because ideas were for him already aspects of
reality? There is no space to pursue such questions here; but
a considerable amount of attention has been devoted to
them, and the reader can find a brilliant conspectus of these
investigations, together with valuable threads with which to
find his own way through the extensive international litera-
ture, in A. C. McGill's 'Recent Discussions of Anselm's
Argument'. (23)

(c) Gaunilo 's criticism

The first critique of the 'Proslogion' proof came in Anselm's

own lifetime in a short book entitled 'On Behalf of the Fool',
presumed to be by Gaunilo, a monk of Marmoutiers; and
Anselm's 'Reply' to this is quite as important as his original
exposition in the 'Proslogion'.
Gaunilo's most important criticism is an attempted reduc-
tio ad absurdum by applying Anselm's reasoning to a most
perfect island. The passage embodying 'Gaunilo's Island' is
quite short and is worth quoting in full:
76
 Consider this example: Certain people say that some-
where in the ocean there is an island, which they call the
'Lost Island' because of the difficulty or, rather, the
impossibility of finding what does not exist. They say that
it is more abundantly filled with inestimable riches and
delights than the Isles of the Blessed, and that although it
has no owner or inhabitants, it excels all the lands that
men inhabit taken together in the unceasing abundance of
its fertility.
When someone tells me that there is such an island, I
easily understand what is being said, for there is nothin 6
difficult here. Suppose, however, as a consequence of this,
that he then goes on to say: You cannot doubt that this
island, more excellent than all lands, actually exists some-
where in reality, because it undoubtedly stands in relation
to your understanding. Since it is more excellent, not
simply to stand in relation to the understanding, b u t to be
in reality as well, therefore this island must necessarily be
in reality. Otherwise, any other land that exists in reality
would be more excellent than this island, and this island,
which you understand to be the most excellent of all
lands, would then not be the most excellent.
If, I repeat, someone should wish by this argument to
demonstrate to me that this island truly exists and is no
longer to be doubted, I would think he were joking; or, if I
accepted the argument, I do not know whom I would
regard as the greater fool, me for accepting it or him for
supposing that he had proved the existence of this island
with any kind of certainty. (24)

It will be noted that Gaunilo's definition of his island is

not an accurate transposition of Anselm's definition of God.
Instead of speaking of 'that island than which no more
perfect can be conceived', Gaunilo speaks of an island 'more
excellent than all lands' (insulam Mam terris omnibus
praestantiorem). This is an unfortunate formula. For it refers
to that land — it does not matter whether it be an island or
not — which is more excellent than all others. It merely
speaks, in other words, of the most excellent land, in the
sense of that land which is not excelled by any other.
Gaunilo seems to think it impossible to prove that such a
77
land exists. But in fact nothing could be easier than to prove
this, for it is a tautology that the most excellent land there is,
exists. However (as we saw above, p. 74) this makes no
connection with Anselm's own argument, which is n o t t o the
effect that the most excellent thing that there is, exists, but
to the quite different effect that that than which no more
excellent can be conceived exists.
However Gaunilo's formula has been tacitly rewritten by a
succession of commentators so as to parallel Anselm's: an
island than which no more perfect can be conceived. If one
could use Anselm's own form of reasoning to prove the
existence of such an island this would indeed be the reductio
ad absurdum that Gaunilo was attempting. However Anselm
denies that any such argument is capable of being formu-
lated. 'I can confidently say,' he maintains, 'that if anyone
discovers for me something existing in fact or at least in
thought, other than "that than which a greater cannot be
conceived", and is able to apply the logic of my argument to
it, I shall find that "Lost Island" for him and shall give it to
him as something which he will never lose again. (25) The
point — developed in Anselm's 'Reply' as a whole and
applied, though not very fully or explicitly, in paragraph 3 —
is that 'that than which no greater can be conceived' exists
necessarily, whilst an island (or any other finite object that
Gaunilo might have chosen instead) can exist only contin-
gently. Anselm's argument in its second form — and it is this
second form that is developed in the 'Reply' to Gaunilo — is
that it is more perfect to exist necessarily than to exist
contingently, and that therefore that than which no more
perfect can be conceived exists necessarily, and therefore
exists. But this argument cannot be applied to 'that island
than which no more perfect can be conceived.' An island,
being by definition a part of the physical world and thus
dependent for its formation and character upon other aspects
of the world, as well as sharing the contingent nature of the
world as a whole, cannot be the subject of the Anselmian
argument. That reasoning only applies to the unique case of
the being than which no more perfect can be conceived.

78
(d) Cartesian reformulation and Kantian criticism

No discussions of Anselm's argument are known during more

than a century following the appearance of the 'Proslogion'.
But after this strange delay it received considerable attention
in the thirteenth and fourteenth centuries, being attacked or
defended (though not always in precisely Anselm's form) by
Bonaventura, Thomas Aquinas, Duns Scotus and a number of
others. But the second main period in the history of the
ontological argument begins with Rene Descartes, the 'father
of modern philosophy'. In his 'Meditations' (1641) he
formulated a proof that is recognisably akin to Anselm's
despite certain variations. (It is not clear whether Descartes
had in fact read the 'Proslogion'; but if not he must surely
have met the argument in some other writer's references to
Anselm or in one of the forms in which the argument
appeared in later medieval literature.) Descartes' definition of
God differs from Anselm's. Instead of 'that than which no
greater can be conceived' Descartes speaks of 'a supremely
perfect being' (un etre souverainement parfaii). But his main
contribution to the discussion of the argument consisted in
highlighting the previously concealed assumption that
existence is a quality or attribute or predicate such that it is
possible to ask of a given x whether or not it has this
attribute. Descartes asserts explicitly that 'existence is a
perfection', (26) i.e. a desirable attribute, which it is more
excellent to have than to lack; and he summarises his
argument as that 'it is in truth necessary for me to assert that
God exists after having presupposed that He possesses every
sort of perfection, since existence is one of these.' (27)
Given, then, that existence is a perfection we can,
Descartes claims, see an analytical connection between the
other attributes of a supremely perfect being and the
attribute of existing. He says:

It is certain that I no less find the idea of God, that is to

say, the idea of a supremely perfect Being, in me, than that
of any figure or number whatever it is; and I do not know
any less clearly and distinctly that an [actual and] eternal
existence pertains to this nature than I know that all that
which I am able to demonstrate of some figure or number
79
 truly pertains to the nature of this figure or number . . .
(28)

'Existence', he therefore concludes, 'can no more be

separated from the essence [or definition] of God than can
its having its three angles equal to two right angles be
separated from the essence of a [rectilinear] triangle, or the
idea of a mountain from the idea of a valley; and so there is
not any less repugnance to our conceiving a God (that is, a
being supremely perfect) to whom existing is lacking (that is
to say, to whom a certain perfection is lacking), than to
conceive of a mountain which has no valley.' (29)
The centrality, in Descartes' argument, of the concept of
existence as a predicable characteristic is indicated again by
his reply to an objection which he considers, namely that
'from the fact that I conceive of a mountain with a valley, it
does not follow that there is such a mountain in the world;
similarly although I conceive of God as possessing existence,
it would seem that it does not follow that there is a God
which exists.' (30) His reply is: 'from the fact that I cannot
conceive a mountain without a valley, it does not follow that
there is any mountain or any valley in existence . . . . While
from the fact that I cannot conceive God without existence,
it follows that existence is inseparable from Him, and hence
that He really exists.' (31)
Ever since the logical structure of the argument was thus
made clear by Descartes' work criticism has centred upon the
premise that 'existence is a predicate', i.e. an attribute which
something may have or lack. This assumption was challenged
by one of Descartes' contemporaries, Pierre Gassendi, who
declared that 'existence is a perfection neither in God nor in
anything else; it is rather that in the absence of which there is
no perfection.' (32) He continued:

This must be so if, indeed, that which does not exist has
neither perfection nor imperfection, and that which exists
and has various perfections, does not have its existence as a
particular perfection and as one of the number of its
perfections, but as that by means of which the thing itself
equally with its perfections is in existence, and without
which neither can it be said to possess perfections, nor can
80
 perfections be said to be possessed by it. Hence neither is
existence held to exist in a thing in the way that per-
fections do, nor if the thing lacks existence is it said to be
imperfect (or deprived of a perfection), so much as to be
nothing. (33)

Descartes was unable to offer any reply beyond that of

simply reaffirming the premise that had been challenged. (34)
Essentially the same criticism was later made, more
influentially, by Immanuel Kant in his 'Critique of Pure
Reason' (first edition 1 7 8 1 , second edition 1787). Kant's
discussion is in two phases. (35) In the first he grants,
hypothetically, the claim that the thought of the sum of
perfection includes within itself the thought of existence, so
that it would be self-contradictory to posit an all-perfect
being (i.e. to suppose such to be) and yet to posit it as
lacking existence. However, he insists, it b y no means follows
from this that there is an all-perfect being. Given the idea of
an existing all-perfect being, the question remains completely
open whether this idea answers to anything in reality:

If, in an identical proposition, I reject the predicate while

retaining the subject, contradiction results; and I therefore
say that the former belongs necessarily to the latter. But if
we reject subject and predicate alike, there is no contra-
diction; for nothing is then left that can be contradicted.
To posit a triangle, and yet to reject its three angles, is
self-contradictory; but there is no contradiction in
rejecting the triangle together with its three angles. The
same holds true of the concept of an absolutely necessary
being. If its existence is rejected, we reject the thing itself
with all its predicates; and no question of contradiction
can then arise. (36)

In the second phase of his discussion Kant challenges the

concealed premise of the ontological argument that existence
is a predicate which, like other predicates, can be included in
the definition of a kind of object. 'Being unmarried' and
'having four feet' are examples of ordinary predicates which
can figure in definitions; so that it is possible to demonstrate
a priori that 'all bachelors are unmarried' and 'all quadrupeds
81
are four-footed'. If existence were a predicate of the same
logical type as these it would be possible to demonstrate a
priori, as the ontological argument professes to do, that an
all-perfect being has the attribute of existing. But against any
such procedure Kant insists that '"Being" is obviously not a
real predicate' and that 'all existential propositions [i.e.
propositions of the form 'x exists'] are s y n t h e t i c ' (37) That
is to say, the logical function of 'exists' is not to add a
further predicate to a definition, but to assert that the
definition applies to something in the world. In other words,
'By whatever and by however many predicates we may think
a thing — even if we completely determine it — we do not
make the least addition to the thing when we further declare
that this thing is.' (38) Thus, 'The proposition, "God is
omnipotent", contains two concepts, each of which has its
object — God and omnipotence. The small word " i s " adds no
new predicate, but only serves to posit the predicate in its
relation to the subject. If, now, we take the subject (God)
with all its predicates (among which is omnipotence), and say
"God is", or "There is a God", we attach no new predicate to
the concept of God, but only posit the subject in itself with
all its predicates, and indeed posit it as being an object that
stands in relation to my concept.' (39)
The point has been made even more luminously clear in
the twentieth century by the analysis of the function of
'exists' which is part of Bertrand Russell's theory of
descriptions. (40) The upshot of Russell's discussions is that
when we say for example, that 'cows exist' but that 'unicorns
do not exist', we are not speaking about cows and saying of
them that they have the attribute of existence, or of unicorns
and affirming that they lack this same attribute. If that were
the case we should have on our hands the problem of the
status of those objects of discourse, such as unicorns, which
are evidently in some sense 'there' to be talked about but of
which we say that they do not exist. This problem has given
rise to metaphysical theories which — if Russell is right —
have been responses to a false problem.
We are talking instead about the concepts 'cow' and
'unicorn' and are saying of them that the one is instantiated
and the other not. That cows exist means that the concept of
a cow has instances; or putting it slightly differently, that the
82
propositional function 'x is a cow' is sometimes true. And
that unicorns do not exist means that the concept of a
unicorn has no instances; or that the propositional function
'x is a unicorn' is never true.
This account of the logical function of 'exists' is one of the
few fairly generally agreed steps forward in philosophy; and
its effect upon the ontological argument (in its first form) is
to rule out its indispensable premise that existence is a
predicable attribute. The issue of course is not whether
'exists' functions as a predicate in any sense, but whether it
functions as the kind of predicate that the ontological
argument requires it to be. Clearly 'exists' is grammatically a
predicate. And beyond this it is possible to think of sentences
in which it is natural to regard 'exists' as fulfilling the
function of a real predicate — for example, 'This (apparent)
table exists (i.e. is real rather than hallucinatory)', or 'This x
exists (i.e. it is logically possible that it might not have
existed)' (41); or again, 'President Nixon exists (i.e. is now
alive), but President Kennedy does not'. (42) However these
cases do not affect the conclusion that existence is not a
predicate in any sense that would validate the ontological
argument; for that argument does not seek to prove that God
exists as a real in distinction from a hallucinatory material
object, or as something presented that might not have been
presented, or as a living in distinction from a deceased human
being. The essence of the argument, in its 'Proslogion' 2 and
Cartesian form, is that existence is a predicate such that it
makes sense to say that a being which has it is superior to one
which does not, and therefore that an unsurpassably perfect
being must have it, and is thereby proved to exist. But
existence is not a predicate in that sense, and any argument
which presupposes that it is must be an invalid argument.

D H.A.E.G. 83
6 The Ontological Argument:
Second Form

(a) 'Necessary being'

'Proslogion' 2 seems to be offered as an argument, complete

in itself, for its conclusion that 'something than which a
greater cannot be conceived undoubtedly both stands in
relation to the understanding and exists in reality.' However
'Proslogion' 3 repeats the same form of argument, in terms
now not of existence but of necessary existence — meaning
by necessary existence (a phrase which Anselm himself does
not use) the existence of something which is such that it
cannot be thought not to exist. Anselm's argument in this
second form is that it is greater to have necessary existence
than not to have it; and that that than which no greater can
be conceived accordingly has necessary existence and there-
fore necessarily exists.
'Necessary existence' or 'necessary being', as the existence
of something which cannot be thought not to exist, can have
at least three different meanings, and it is important to see
what meaning it has in Anselm's argument.
1. It could be intended to mean logically necessary
existence. For 'necessary' has come in philosophical
discussion today to mean, primarily, 'logically necessary'.
And there is prima facie support for the idea of logically
necessary existence in the converse fact of logically necessary
non-existence. For self-contradictory concepts are not merely
contingently but necessarily not instantiated. Square circles
and things that are both green and blue all over necessarily
fail to exist: it is a logical truth that there are no such things.
Why may there not, then, be a concept which is necessarily
instantiated; an entity such that it is a logical truth that there
is such a thing? And why may not the idea of the infinitely
perfect being meet this specification?
The answer to this prima facie consideration is that the
notions of necessary non-existence and necessary existence
are not logically symmetrical. It is non-controversial that a
84
concept can guarantee its own non-instantiation by being
self-contradictory. But the converse idea that a concept can
guarantee its own instantiation, the idea in other words of a _
being whose existence is logically necessary, is excluded by
the modern empiricist understanding of the nature of logical
necessity.
In the first place, logical necessity is not a property of
things but of propositions; so that 'logically necessary being'
is, strictly, a meaningless conjunction of words. The proper
question is not whether there is a logically necessary being,
but whether the proposition 'God exists' or 'An un-
surpassably perfect being exists' is a logically necessary truth.
But if, as Hume, Kant, Russell and many others have argued,
all existential propositions (i.e. propositions of the form 'x
exists') are synthetic, then 'God exists' cannot, as an
existential proposition, be logically necessary. Thus the
notion of a being whose existence is logically necessary, or of
whom it is a logical truth that he exists, is a malformed
notion. A number of philosophers have assumed that the
theologians' concept of God as necessary being is precisely
this mistaken notion of God as logically necessary being, and
have accordingly rejected the idea as meaningless. (1) But the
concept which they are rejecting is not that employed by
Anselm — or indeed by other major Christian theologians,
such as Thomas Aquinas, w h o have spoken of God as
necessary being.
2. It could mean that God's non-existence is unthinkable
within the framework of theistic faith. This is necessity
relative to a certain context, and is explained as follows by
J . J . C. Smart, who introduced the notion into contemporary
discussion:

It is not a logical necessity that the velocity of light in a

vacuum should be constant. It would, however, upset
physical theory considerably if we denied it. Similarly it is
not a logical necessity that God exists. But it would clearly
upset the structure of our religious attitudes in the most
violent way if we denied it or even entertained the
possibility of its falsehood. So if we say that it is aphysicdl
necessity that the velocity of light in vacuo should be
constant — (deny it and prevailing physical theory would
85
 have to be scrapped or at any rate drastically modified) —
similarly we can say that it is a religious necessity that God
exists. That is, we believe in the necessity of God's
existence because we are Christians; we are not Christians
because we believe in the necessity of God's existence. (2)

This is not however a concept of necessary being, but of

necessary belief. Within the setting of Christian thought it is
necessary to believe that God exists; and it is necessary in the
sense that one who does not so believe is, by definition, not
operating within the Christian thought-world. This necessity
is thus relative to the acceptance of a systematic body of
ideas which includes belief in the reality of God. This is not
however at all what Anselm had in mind when he spoke of
God as a being who cannot be thought not to exist.
3. Anselm himself states clearly what he means by a being
which exists in such a way that it cannot be conceived not to
exist. He says: 'that alone cannot be conceived not to be in
which conceiving discovers neither beginning nor end nor
combination of parts, and which it finds existing always and
everywhere in its totality.' (3) He is here feeling towards
what later became known as the notion of aseity (from a se
esse) usually Englished as 'self-existence'. In its full develop-
ment this is the idea of the existence of something that
simply and unqualifiedly is, without beginning or end and
without dependence for its existence or for its characteristics
upon anything other than itself. If, following traditional
usage, we describe such a being as 'necessary', this does not
imply either a logically necessary being, or a being belief in
whose existence is necessary within a certain conceptual
framework. It describes instead what has sometimes been
called a factually or ontologically necessary being.
From God's aseity his eternity, indestructibility and
incorruptibility can be seen to follow. A self-existent being
must be eternal, i.e. without temporal limitation. For if he
had begun to exist or should cease to exist, he must have
been caused to exist or to go out of existence by some power
other than himself; and this would be inconsistent with his
aseity. By the same token, he must be indestructible, for to
say that he exists in total independence is to say that there is
and could be no reality able to constitute or to destroy him;
86
and likewise he must be incorruptible, for otherwise his
aseity would be qualified as regards its duration. Again, it is
meaningless to say of the self-existent being that he might
not have existed or that he merely happens to exist. For what
would it mean to say of the eternal, uncreated Creator of
everything other than himself that he merely happens to
exist? When we assert of a dependent and temporally finite
being, such as myself, that I only happen to exist, we mean
that if such-and-such an event had occurred in the past, or if
such-and-such another event had failed to occur, I should not
now exist. But no such meaning can be given to the state-
ment that a self-existent being only happens to exist, or
might not have existed. There is no conceivable event such
that if it had occured, or failed to occur, a self-existent being
would not have existed; for the concept of aseity is precisely
the exclusion of such dependence. There is and could be
nothing that would have prevented a self-existent being from
coming to exist, for it is meaningless even to speak of a
self-existent being as coming to exist. (4)
We now have before us the two senses of 'necessary' which
enter into discussions of the second form of the ontological
argument — namely the logical necessity of analytic proposi-
tions and the factual necessity of eternal and independent
beings — and can proceed to look at the argument itself.

(b) The argument in Anselm

Anselm's own discussion is concerned with ontological or

factual necessity. He states his argument in slightly different
ways in two successive paragraphs of 'Reply', 1. The first is as
follows:

If this [i.e. 'that than which a greater cannot be con-

ceived'] can at least be conceived to be, it necessarily
follows that it exists. For 'that than which a greater cannot
be conceived' cannot be conceived t o be, except as
without a beginning. However, whatever can be conceived
to be and actually is not can be conceived to be through a
beginning. Therefore, it is not the case that 'that than
which a greater cannot be conceived' can be conceived to
87
 exist and yet does not exist. Therefore, if it can be con-
ceived to be, it necessarily is. (McGill, p. 22.)

In examining this argument let us first identify the three

concepts involved in it. They are:

A. The (one-member) class of that than which a greater

cannot be conceived.
B. The class of things which can be conceived to be
through a beginning.
C. The class of things which can be conceived to be and
actually are not (i.e. which do not exist, but could possibly
exist).
Anselm's argument, stated in a valid form, is:
Every A is a non-B
Every C is a B
•'• Every A is a non-C

The proper conclusion of the argument is thus that 'that than

which a greater cannot be conceived' does not fall in the class
of things which do not exist b u t could possibly exist. In
other words, it is not the case that (a) the idea of God is the
idea of a non-eternal or contingent being, and that (b) God,
so defined, does not exist. It is not however hereby proved
that he is a non-contingent being who does exist (or for that
matter, though Anselm was not interested in proving this,
that he is a contingent being who does exist). In short, it is
not proved that God exists.
It may be useful to set forth this same argument of
Anselm's again in a slightly different form, which I have used
elsewhere: (5)

(i) To be unsurpassably perfect is to be incapable-of-

having-a-beginning;
(ii) to be non-existent-but-capable-of-existing is not to be
incapable-of-having-a-beginning; and
(iii) therefore to be unsurpassably perfect is not to be
non-existent-but-capable-of-existing.

What this argument proves is that God is not

non-existent-but-capable-of-existing, that is, that he is not
88
contingently non-existent. But it does not prove that he
exists.
Anselm's second formulation is as follows:

Further, if it can be conceived in any way at all, it is

necessarily the case that it exists. For while someone may
deny or doubt the existence of something than which a
greater cannot be conceived, he will not deny or doubt
that, if it does exist, then in fact and for the understanding
it is impossible for it not to be. Otherwise, it would not be
that than which a greater cannot be conceived. As for
things which can be conceived and yet do not exist, even if
such things were to exist, in fact and for the understanding
it is possible for them not to be. Therefore, if 'that than
which a greater cannot be conceived' can be conceived at
all, it cannot not be. (McGill, pp. 25-6.)

Here the three concepts involved are:

A. The (one-member) class of that than which a greater

cannot be conceived.
B. The class of things which can be conceived and yet do
not exist (i.e. which do not exist, but could possibly exist).
C. The class of things such that it is impossible for them
not to be.

Once again Anselm's own conclusion is a non sequitur, and

the valid argument from his premises is as follows:
All A's are C's
All B's are non-C's
.'. All A's are non-B's
The conclusion is that 'that than which a greater cannot be
conceived' does not fall in the class of things which do not
exist but could possibly exist. That is to say, divine existence
is not a contingent possibility which happens not to be
realised. But in proving that 'that than which a greater cannot
be conceived' is not a contingent which does not exist, it is
not proved that it is a non-contingent which does exist.
To set out the valid part of Anselm's argument in another
way:(6)
89
 (i) Every non-existent-which-might-exist is a contingent;
(ii) no unsurpassably-perfect is a contingent;
(iii) t h e r e f o r e no unsurpassably-perfect is a
non-existent-which-might-exist; and
(iv) therefore every unsurpassably-perfect is other than a
non-existent-which-might-exist (i.e. is other than contin-
gently non-existent).

Once again, what is proved is that God is not a contingent

being, or more precisely that he does not contingently
not-exist. In being other t h a n a non-existent-
which-might-exist he either exists or is a non-existent which
could not exist (i.e. whose existence is impossible). But what
is n o t proved is that he exists.
Having now examined, in the previous chapter, Anselm's
'Proslogion' 2 argument, and in this chapter his 'Proslogion' 3
and 'Reply' argument, we may next ask what the relation was
in Anselm's own thinking between these two pieces of
reasoning. And we may begin by noting that the distinction
between them has been accentuated by the now traditional
division of the 'Proslogion' into separate chapters, each with
its own title — Chapter 2, 'That God Truly Is', and Chapter 3,
'That It Is Impossible to Conceive That God Is Not'. The
.original text formed a single continuous piece of prose with,
for ease of reference, paragraph numbers in the margin and
identificatory phrases attached to these numbers in a table of
contents at the beginning. As A. C. McGill points out, 'It was
not Anselm b u t his later editors who inserted the chapter
titles into the text and so broke up its continuity into what
look like self-contained and definitively entitled units. . . .
For that reason, there are no grounds for presuming
that Chapter II is a self-contained unit.' (7) Hence the
question of the relation between the two forms or phases of
Anselm's argument must be answered by reference to the
internal logic of his reasoning rather than the external
divisions of the text.
It appears to me that Karl Barth's view of the matter is
essentially correct. (8) That is to say, Anselm is offering a
single argument which divides into two phases. In the first
phase he seeks to prove that God exists in the sense in which
other things exist — that God is one of the items in a
90
complete inventory of the universe. This is what Barth calls
God's 'general' existence. But Anselm's overall concern in the
'Proslogion' is not simply to establish God's existence, but
to establish in a single argumentum both his existence and his
unique nature. And so in the second phase he seeks to show
that the reasoning which proves God's existence also, at a
deeper level, proves his unique nature as self-existent reality,
as that which not merely exists b u t has necessary or ultimate
existence. This is what Barth calls God's 'special' existence.
However the discussion of Anselm's intention is necessarily
conjectural, and the philosophical consideration of the
argument or arguments attributed to him does not depend
upon the answer to such historical questions. Whatever
Anselm's intention, it is still legitimate to emphasise the
differences between the two phases of his reasoning and to
argue, as Norman Malcolm and Charles Hartshorne have
recently done, that whereas the argument in 'Proslogion' 2 is
vulnerable to the Kantian criticism, that in 'Proslogion' 3 and
the 'Reply' to Gaunilo is not.

(c) Norman Malcolm

In his important article 'Anselm's Ontological Arguments' (9)

Norman Malcolm delineates with great clarity Anselm's con-
cept of God's necessary being as eternal and independent
existence. (10) He then quotes the first of the two paragraphs
of Anselm's which I have discussed above (pp. 87-89), and
comments:

What Anselm has proved is that the notion of contingent

existence or of contingent non-existence cannot have any
application to God. His existence must either be logically
necessary or logically impossible. The only intelligible way
of rejecting Anselm's claim that God's existence is
necessary is to maintain that the concept of God, as a
being a greater than which cannot be conceived, is self-
contradictory or nonsensical. Supposing that this is false,
Anselm is right to deduce God's necessary existence from
his characterisation of Him as a being greater than which
cannot be conceived. (11)
91
 Thus Malcolm's interpretation of the argument is as
follows: An eternal being (12) either necessarily exists (in
that by definition if he exists he cannot cease to exist) or
necessarily does not exist (in that by definition if there is no
such being, none can come to exist). Hence — to quote
Malcolm's own words — 'God's existence is either impossible
or necessary. It can be the former only if the concept of such
a being is self-contradictory or in some way logically absurd.
Assuming that this is not so, it follows that He necessarily
exists.' (13)
This argument proceeds by ignoring the circumstances that
the logical necessity and logical impossibility of God's
existence are both, to use an anomalous phrase, hypothetical
necessities: if God exists eternally, it is logically impossible
for him, as an eternal being, to cease to exist; and if he does
not exist, it is logically impossible for such a being to come
into existence. In other words, it is logically impossible for
the existence or non-existence, as the case may be, of an
eternal being to be reversed. But one cannot deduce from this
that providing the concept of an eternal being is not self-
contradictory it is logically necessary that there is an eternal
being. Even if the concept of an eternal being is entirely free
from contradiction, it in no way follows that there must be
such a being. For the 'logical necessity of God's existence' is
simply the logical impossibility that, if he exists, he should
cease existing. It is not logically impossible for there to be no
eternal being, although it is logically impossible, if it is true
that there is one, that it should become true that there is not.
However, later in his essay Malcolm discusses and seeks to
rebut this view that the logical necessity of God's existence is
a conditional necessity. He rejects the idea, first propounded
by Caterus, a contemporary critic of Descartes, and then by
Kant and subsequently by many others, that God's necessary
existence means that if he exists, he exists necessarily.
Malcolm's counter-argument is as follows:

I think that Caterus, Kant, and numerous other philoso-

phers have been mistaken in supposing that the proposi-
tion 'God is a necessary being' (or 'God necessarily exists')
is equivalent to the conditional proposition 'If God exists
then He necessarily exists'. For how do they want the
92
 antecedent clause 'If God exists' to be understood? Clearly
they want it to imply that it is possible that God does not
e x i s t . . . . Let us make this implication explicit in the con-
ditional proposition, so that it reads: 'If God exists (and it
is possible that He does not) then He necessarily
exists' . . . . But so far as from it being the case that the
proposition 'God necessarily exists' entails the proposition
'It is possible that God does not exist', it is rather the. case
that they are incompatible with one another! Can anything
be clearer than [that] the conjunction 'God necessarily
exists but it is possible that He does not exist' is self-
contradictory? Is it not just as plainly self-contradictory as
the conjuction 'A square necessarily has four sides but it is
possible for a square not to have four sides'? In short, this
familiar criticism of the ontological argument is self-
contradictory, because it accepts both of two incompatible
propositions. (14)

But Malcolm is ignoring the circumstance that necessary

being, in the case of God, is equivalent t o eternal being (or,
more fully, to aseity) and that the logical impossibility of an
eternal being ceasing to exist is conditional upon an eternal
being existing. We must therefore translate the absurd 'If God
exists (and it is possible that He does not) then He necessarily
exists' into 'If God exists (and it is possible that He does
not) then He exists eternally'. (The corresponding propos-
ition about a square would be 'If there is a square — and it is
possible that there is not — then it necessarily has four
sides'.)

(d) Charles Hartshorne

Charles Hartshorne has been maintaining the validity of the

second ('Proslogion' 3 and 'Responsio') form of the onto-
logical argument for a quarter of a century or more; and to
him, more than to any other individual, must be given the
credit for having evoked the current wave of renewed interest
in the argument in the English-speaking world. Hartshorne's
collection of writings on the subject, consisting of a book,
several chapters in other books, and numerous articles, notes
93
and rejoinders to reviews, (15) is'so considerable that it has in
the end diffused the impact of his contribution an&Tnade it
harder, instead of easier, to concentrate upon his central
contentions; and the highly polemical and almost obsessive
tone of some of his writings on this subject has likewise
tended to obscure their logical content. However in what is
perhaps his most important essay on the ontological
argument, 'Ten Ontological or Modal Proofs for God's
Existence' in 'The Logic of Perfection', Hartshorne formalises
his argument; and here we have a purely logical treatment
which can be examined as such:

The logical structure of the Anselmian argument, in its

mature or 'Second' form, may be partially formalized as
follows:
q for (3x)Px, There is a perfect being, or perfection exists
N for 'it is necessary (logically true) that'
~ for 'it is not true that'
v for 'or'
p ->• q for 'p strictly implies q' or N~[p & ~q)

1. q -* Nq 'Anselm's Principle': perfection could not exist

[hence, the assertion that it exists could not be
contingently but only necessarily true] (16)
2. Nq v~Nq Excluded Middle
3. ~Nq -> N~Nq Form of Becker's Postulate: modal
status is always necessary
4. Nq v N~Nq Inference from (2,3)
5. N~Nq -> N~q Inference from (1): the necessary
falsity of the consequent implies that of the antecedent
(Modal form of modus tollens)
6. Nq v N~q Inference from (4,5)
7.~N~q Intuitive postulate (or conclusion from other
theistic arguments): perfection is not impossible
8. Nq Inference from (6,7)
9. Nq—> q Modal axiom
10. q Inference from (8,9) (17)

I believe that this argument is fallacious. The fallacy is

basically that already encountered in Malcolm and consists in
an equivocation in the use of the term 'necessary'. Because
94
Hartshorne so conveniently formalises the argument one can
point in his case to the exact place where this equivocation
occurs, namely at prop. 6. The two senses of 'necessary'
which are switched at that point are those discussed above:
the logical necessity of analytic propositions, and the factual
or ontological necessity of a being who is defined as having
eternal and independent existence. Hartshorne himself stales
that his argument is to be understood exclusively in terms of
logical necessity, which he defines in accordance with the
modern empiricist doctrine that such necessity is ultimately
tautological. Hartshorne says, 'In general it [JV] means
analytic or L-true, true by necessity of the meanings of the
terms employed. This is the sense intended in the present
essay.'(18)
However, if we interpret JV in this way Hartshorne's
argument fails decisively at the outset. For the first
proposition reads:

'Perfection exists' strictly implies ' "Perfection exists" is

logically true'.

In other words, Hartshorne's initial premise is that

'Perfection exists' is an analytic truth. But it is basic to the
modern .enigiricist.JJnderstanding of_7V, to which Hartshorne
expTicitily "appeals,""'" tKaT""existen'fiaT" propositions are not
analytic and therefore not L—true. It is accordingly
impossible to make sense of Hartshorne's first proposition in
the terms in which he says it is to be understood. If his first
proposition were acceptable, the argument could proceed by
valid steps to its conclusion. But the initial proposition is
ruled out by the very logic to which Hartshorne appeals when
he says that JV is to be interpreted as 'true by necessity of the
meanings of the terms employed'.
To add to the difficulty, Hartshorne labels his first
proposition 'Anselm's Principle', namely that perfection
could not exist contingently. We have already seen that
Anselm's principle is not that 'God exists' is an analytic
proposition but that divine existence is eternal and
independent existence ('that alone cannot be conceived not
to be in which conceiving discovers neither beginning nor end
nor combination of parts, and which it finds existing always
95
and everywhere in its totality'). (19) It is clear from the texts
that the notion of logical necessity in the modern sense was
not in Anselm's mind; and it is therefore surprising that
Hartshorne, who frequently chides the scholarly world for
not bothering to read Anselm, (20) should so entirely have
misstated Anselm's basic principle.
We are thus offered two conflicting clues. On the one hand
Hartshorne's prop. 1 is to be interpreted in terms of the
modern empiricist understanding of logical necessity as
analytical; and on the other hand it is to be interpreted in
terms of Anselm's notion of a factual necessity which is
equivalent to aseity. The first interpretation, we have seen, is
completely abortive. But the second, if we were to adopt it,
would enable the argument to begin and to proceed as far as
prop. 6:

1. That God exists means that he exists eternally.

2. Either God exists eternally or it is not the case that he
exists eternally.
3. That God does not exist eternally means that it is
eternally the case that he does not exist (eternally). (21)
4. Either God exists eternally or it is eternally the case that
he does not exist (eternally).
5. That it is eternally the case that God does not exist
(eternally) means that eternally he does not exist (eternally).
6. Either God exists eternally or it is eternally the case that
he does not exist (eternally).

However if we thus interpret N consistently in terms of

factual or ontological necessity, in accordance with Anselm's
principle, the argument is unable to progress beyond this
point. For from the disjunction in prop. 6 we cannot infer
either that there is an eternal being or that there is not. We
have proved that since the divine nature is defined as eternal,
God's existence is either factually necessary (i.e. he eternally
exists) or factually impossible (i.e. eternally he does not
exist). But the argument does not supply us with any grounds
for preferring one of these possibilities to the other.
In order that Hartshorne's argument shall proceed from
prop. 6 to its conclusion in prop. 10, the meaning of
'necessary' has to change at this point from factual to logical
96
necessity. Propositions 1-6 have to be interpreted in terms of
factual necessity and 6-10 in terms of logical necessity, with
prop. 6 itself being established as a conclusion in terms of the
former and then used as a premise in terms of the latter.
With prop. 6 now understood as ' "God exists" is either
logically true or logically false' (or 'God's existence is either
logically necessary or logically impossible'), one can argue
that 'God exists' would only be logically false if the concept
of deity is internally incoherent, but that since it has not
been shown to be so, we must assume the contrary. This
assumption is Hartshorne's 'intuitive postulate' in prop. 7. (It
will be remembered that Malcolm makes the same move:
'God's existence is either impossible or necessary. It can be
the former only if the concept of such a being is self-
contradictory or in some way logically absurd.') (22) The
argument can then proceed:

6. Either 'God exists' is logically necessary or 'God does

not exist' is logically necessary.
7. 'God does not exist' is not logically necessary.
8. 'God exists' is logically necessary.
9. ' "God exists" is logically necessary' implies 'God exists'.
10. God exists.

But the continuity of the argument from props. 1 to 10

depends upon the change of interpretation of N at prop. 6;
and the argument as a whole is therefore irremediably invalid.
The conclusion of this whole discussion must therefore be
that the second form of the ontological argument is after all
subject to the same basic criticism as the first — it is unable
to deduce real existence from a concept.

(e) A new Scotist modal argument

An American philosopher, James F. Ross, has recently

published an interesting full-scale treatment of what can be
described as a first cousin to Hartshorne's modal argument.
Like Hartshorne, Ross argues from possibility to necessity;
but whereas Hartshorne traces his own proof back to Anselm,
Ross traces his to Duns Scotus. Ross's discussion is both
97
lengthy and elaborate; b u t it leads eventually to two
arguments which, he claims, 'adequately establish the
existence of a being which is properly called " G o d " '. (23)
The first argument is as follows:

1. It is possible that whatever is not the case should have

either a self-explanation in terms of the inherent absurdity
of its having been the case, or a hetero-explanation in
terms of causes, agents, or producers whose actions
prevent its being the case or whose own failure to act or to
obtain is causally sufficient for its not being the case.
2. 'God does not exist' cannot be the case, because;
(a) There is no absurdity or contradiction in 'God does
exist', and hence 'God does not exist' is not self-
explanatory.
(b) 'God' cannot denote anything unless that thing be
both uncaused and unprevented by any other thing and,
furthermore, be both uncausable and unpreventable by
any entity whatever. Therefore, 'God' cannot denote
anything which is such that 'God does not exist' is hetero-
explicable.
3. Then 'God exists' must be the case. (24)

This argument specifies two conditions under which it may

be the case that x does not exist — (i) that the notion of x as
existing is absurd or self-contradictory, and (ii) that some-
thing has causally prevented x from existing. But these are
not claimed to exhaust the set of conditions under which it
can be the case that x does not exist. ('It is possible [my
italics] that whatever is not the case should have either . . . . ' )
However the argument would only be formally valid if this
list of two conditions were exhaustive. For the argument is
that since neither of these two conditions applies to the non-
existence of God, it cannot be the case that God does not
exist, and therefore God does exist. But with the possibility
left open that there may be other conditions under which it
is the case that x does not exist, nothing at all follows con-
cerning the existence of God — or indeed concerning the
existence of anything.
If Ross were to seek to amend the argument by
incorporating an assertion that his two conditions are jointly
98
exhaustive, we should be obliged to resist such a move. For
Ross would thereby be begging the question. The whole issue
is whether or not there could fail to be a God without the
conception of God being defective and without there being
any causes, agents or producers preventing there being a God.
Let us agree that the concept of God is not self-
contradictory. It does not follow that the existence or
otherwise of God is a contingent matter, in the sense of being
dependent upon causal factors. On the contrary (as Ross
makes clear in 2(b)) the concept of God itself excludes this.
If God exists he exists eternally and independently, and not
as a result of causal conditions or permissions; and if there is
no God, there is eternally no God, this situation likewise not
being a result of any causal conditions. Only if God were
defined as a finite object within the universe would his
existence or non-existence depend upon causal circum-
stances. Thus Ross's first argument fails. His second argument
is as follows:

1. That there is at least one Uncausable Producer [or

uncreated creator] is logically possible.
2. Whatever is logically possible is either actual or
potential.
3. Whatever is potential is causable.
4. No Uncausable Producer is causable.
5. Hence, no Uncausable Producer is potential.
6. Therefore, at least one Uncausable Producer is actual,
that is, exists. (25)

A similar flaw appears in this argument. Is the meaning of

'logically possible' exhausted by the dichotomy 'either actual
or potential' when this is glossed as meaning 'either actual or
causable'? Clearly not; such a dichotomy only applies to
contingent things. But the concept of God is the concept of
an eternal and uncaused being. And whether or not there is
such a being cannot be determined by a process of reasoning
which applies only to contingents. Thus this argument also
fails.
Having thus concluded that the ontological proof, in all its
known forms, fails, we can nevertheless still ask whether, like
the teleological, cosmological and moral arguments, it can
99
still be seen as pointing to the possibility of God, b y posing a
question to which God might be the answer. But it does not
seem even to do this. It does not point, as do the other
arguments, to a problematic aspect of human experience
which might be illuminated by postulating the existence of
God. It is thus in a different category from the other
arguments which we have examined — as indeed we might
expect from the fact that it is the only purely a priori
argument.

100
7 Rational Theistic Belief
Without Proofs

(a) The religious rejection of the theistic arguments

We have seen that the major theistic arguments are all open
to serious philosophical objections. Indeed we have in each
case concluded, in agreement with the majority of contem-
porary philosophers, that these arguments fail to do what
they profess to do. Neither those which undertake strictly to
demonstrate the existence of an absoluteBeing, nor those which
profess to show divine existence to be probable, are able to
fulfil their promise. We have seen that it is impossible to
demonstrate the reality of God by a priori reasoning, since
such reasoning is confined to the realm of concepts;
impossible to demonstrate it by a posteriori reasoning, since
this would have to include a premise begging the very
question at issue; and impossible to establish it as in a greater
or lesser degree probable, since the notion of probability
lacks any clear meaning in this context. A philosopher
unacquainted with modern developments in theology might
well assume that theologians would, ex officio, be supporters
of the theistic proofs and would regard as a fatal blow this
conclusion that there can be neither a strict demonstration of
God's existence nor a valid probability argument for it. In
fact however such an assumption would be true only of
certain theological schools. It is true of the more traditional
Roman Catholic theology, (1) of sections of conservative
Protestantism, (2) and of most of those Protestant apologists
who continue to work within the tradition of nineteenth-
century idealism. (3) It has never been true, on the other
hand, of Jewish religious thought; (4) and it is not true of
that central stream of contemporary Protestant theology
which has been influenced by the 'neo-orthodox' movement,
the revival of Reformation studies and the 'existentialism' of
Kierkegaard and his successors; or of the most significant
contemporary Roman Catholic thinkers, who are on this
issue (as on so many others) in advance of the official
101
teaching of the magisterium. Accordingly we have now to
take note of this theological rejection of the theistic proofs,
ranging from a complete lack of concern for them to a
positive repudiation of them as being religiously irrelevant or
even harmful. There are several different considerations to be
evaluated.
1. It has often been pointed out that for the man of faith,
as he is depicted in the Bible, no theistic proofs are necessary.
(5) Philosophers in the rationalist tradition, holding that to
know means to be able to prove, have been shocked to find
that in the Bible, which is supposed to be the basis of
Western religion, no attempt whatever is made to demon-
strate the existence of God. Instead of professing to establish
the divine reality by philosophical reasoning the Bible
throughout takes this for granted. Indeed to the biblical
writers it would have seemed absurd to try to establish by
logical argumentation that God exists. For they were
convinced that they were already having to do with him and
he with them in all the affairs of their lives. They did not
think of God as an inferred entity but as an experienced
reality. Many of the biblical writers were (sometimes, though
doubtless not at all times) as vividly conscious of being in
God's presence as they were of living in a material world. It is
impossible to read their pages without realising that to them
God was not a proposition completing a syllogism, or an idea
adopted by the mind, but the supreme experiential reality. It
would be as sensible for a husband to desire a philosophical
proof of the existence of the wife and family who contribute
so much of the meaning and value of his life as for the man
of faith to seek for a proof of the existence of the God
within whose purpose he believes that he lives and moves and
has his being.
As Cook Wilson wrote:

If we think of the existence of our friends; it is the 'direct

knowledge' which we want: merely inferential knowledge
seems a poor affair. To most men it would be as surprising
as unwelcome to hear it could not be directly known
whether there were such existences as their friends, and
that it was only a matter of (probable) empirical argument
and inference from facts which are directly known. And
102
 even if we convince ourselves on reflection that this is
really the case, our actions prove that we have a con-
fidence in the existence of our friends which can't be
derived from an empirical argument (which can never be
certain) for a man will risk his life for his friend. We d o n ' t
want merely inferred friends. Could we possibly be
satisfied with an inferred God? (6)

In other words the man of faith has no need of theistic

proofs; for he has something which for him is much better.
However it does not follow from this that there may not be
others who do need a theistic proof, nor does it follow that
there are in fact no such proofs. All that has been said about
the irrelevance of proofs to the life of faith may well be true,
and yet it might still be the case that there are valid
arguments capable of establishing the existence of God to
those who stand outside the life of faith.
2. It has also often been pointed out that the God whose
existence each of the traditional theistic proofs professes to
establish is only an abstraction from and a pale shadow of the
living God who is the putative object of biblical faith. A First
Cause of the Universe might or might not be a deity to whom
an unqualified devotion, love and trust would be appropriate;
Aquinas's Et hoc omnes intelligunt Deum ('and this all under-
stand to be God') is n o t the last step in a logical argument
but merely an exercise of the custom of overlooking a gap in
the argument at this point. A Necessary Being, and indeed a
being who is metaphysically absolute in every respect —
omnipotent, omniscient, eternal, uncreated — might be
morally good or evil. As H. D. Aitken has remarked,
'Logically, there is no reason why an almighty and
omniscient being might not be a perfect stinker. (7) A divine
Designer of the world whose nature is read off from the
appearances of nature might, as Hume showed, be finite or
infinite, perfect or imperfect, omniscient or fallible, and
might indeed be not one being but a veritable pantheon. (8)
It is only by going beyond what is proved, or claimed to have
been proved, and identifying the First Cause, Necessary
Being, or Mind behind Nature with the God of biblical faith
that these proofs could ever properly impel to worship. By
themselves and without supplementation of content and
103
infusion of emotional life from religious traditions and
experiences transcending the proofs themselves they would
never lead to the life of faith.
The ontological argument on the other hand is in this
respect in a different category. If it succeeds it establishes the
reality of a being so perfect in every way that no more
perfect can be conceived. Clearly if such a being is not
worthy of worship none ever could be. It would therefore
seem that, unlike the other proofs, the ontological argument,
if it were logically sound, would present the relatively few
persons who are capable of appreciating such abstract
reasoning with a rational ground for worship. On the other
hand, however, whilst this is the argument that would
accomplish most if it succeeded it is also the argument which
is most absolutely incapable of succeeding; for it is, as we
have seen, inextricably involved in the fallacy of professing to
deduce existence from a concept.
3. It is argued by some religious writers that a logical
demonstration of the existence of God would be a form of
coercion and would as such be incompatible with God's
evident intention to treat his human creatures as free and
responsible persons. A great deal of twentieth-century
theology emphasises that God as the infinite personal reality,
having made man as person in his own image, always treats
men as persons, respecting their relative freedom and
autonomy. He does not override the human mind by
revealing himself in overwhelming majesty and power, but
always approaches us in ways that leave room for an un-
compelled response of human faith. Even God's own entry
into our earthly history, it is said, was in an 'incognito' that
could be penetrated only by the eyes of faith. As Pascal put
it, 'willing to appear openly to those who seek him with all
their heart, and to be hidden from those who flee from him
with all their heart, he so regulates the knowledge of himself
that he has given indications of himself which are visible to
those who seek him and not to those who do not seek him.
There is enough light for those to see who only desire to see,
and enough obscurity for those who have a contrary
disposition.' (9) God's self-revealing actions are accordingly
always so mediated through the events of our temporal
experience that men only become aware of the divine
104
presence by interpreting and responding to these events in
the way which we call religious faith. For if God were to
disclose himself to us in the coercive manner in which our
physical environment obtrudes itself we should be dwarfed to
nothingness by the infinite power thus irresistibly breaking
open the privacy of our souls. Further, we should be
spiritually blinded by God's perfect holiness and paralysed by
his infinite energy; 'for human kind cannot bear very much
reality.' (10) Such a direct, unmediated confrontation break-
ing in upon us and shattering the frail autonomy of our finite
nature would leave no ground for a free human response of
trust, self-commitment and obedience. There could be no call
for a man to venture upon a dawning consciousness of God's
reality and thus to receive this consciousness as an authentic
part of his own personal existence precisely because it has
not been injected into him or clamped upon him by
magisterial exercise of divine omnipotence.
The basic principle invoked here is that for the sake of
creating a personal relationship of love and trust with his
human creatures God does not force an awareness of himself
upon them. And (according to the view which we are
considering) it is only a further application of the same
principle to add that a logically compelling demonstration of
God's existence would likewise frustrate this purpose. For
men — or at least those of them who are capable of following
the proof — could then be forced to know that God is real.
Thus Alasdair Maclntyre, when a Christian apologist, wrote:
'For if we could produce logically cogent arguments we
should produce the kind of certitude that leaves no room for
decision; where proof is in place, decision is not. We do not
decide to accept Euclid's conclusions; we merely look to the
rigour of his arguments. If the existence of God were
demonstrable we should be as bereft of the possibility of
making a free decision to love God as we should be if every
utterance of doubt or unbelief was answered by thunder-
bolts from heaven.' (11) This is the 'religious coercion'
objection to the theistic proofs.
To what extent is it a sound objection? We may accept the
theological doctrine that for God to force men to know him
by the coercion of logic would be incompatible with his
purpose of winning the voluntary response and worship of
105
free moral beings. But the question still remains whether the
theistic proofs could ever do this. Could a verbal proof of
divine existence compel a consciousness of God comparable
in coerciveness with a direct manifestation of his divine
majesty and power? Could anyone be moved and shaken in
their whole being by the demonstration of a proposition, as
men have been by a numinous experience of overpowering
impressiveness? Would the things that have just been said
about an overwhelming display of divine glory really apply to
verbal demonstrations — that infinite power would be
irresistibly breaking in upon the privacy of our souls and that
we should be blinded by God's perfect holiness and paralysed
by his infinite energy? Indeed could a form of words,
culminating in the proposition that 'God exists', ever have
power by itself to produce more than what Newman calls a
notional assent in our minds? (12)
It is of course true that the effect of purely rational
considerations such as those which are brought to bear in the
theistic proofs are much greater in some minds than in
others. The more rational the mind the more considerable is
the effect to be expected. In many persons — indeed taking
mankind as a whole, in the great majority — the effect
of a theistic proof, even when no logical flaw is found in it,
would be virtually nil! But in more sophisticated minds the
effect must be greater, and it is at least theoretically possible
that there are minds so rational that purely logical con-
siderations can move them as effectively as the evidence of
their senses. It is therefore conceivable that someone who is
initially agnostic might be presented with a philosophical
proof of divine existence — say the ontological argument,
with its definition of God as that than which no more perfect
can be conceived — and might as a result be led to worship
the being whose reality has thus been demonstrated to him.
This seems to be possible; but I believe that even in suchj^
case there must, in addition to an intelligent appreciation of
tFe argument, be a distinctively religious response to the idea
of God which the argument presents. Some propensity to
respond to unlimited perfection as holy and as rightly
claiming a response of unqualified worship and devotion
must operate, over and above the purely intellectual capacity
for logical calculation. (13) For we can conceive of a purely
"106
or merely logical mind, a kind of human calculating machine,
which is at the same time devoid of the capacity for
numinous feeling and worshipping response. Such a being
might infer that God exists but be no more existentially
interested in this conclusion than many people are in, say,
the fact that the Shasta Dam is 602 feet high. It therefore
seems that when the acceptance of a theistic proof leads to
worship, a religious reaction occurs which turns what would
otherwise be a purely abstract conclusion into an immensely
significant and moving fact. In Newman's terminology, when
a notional assent to the proposition that God exists becomes
a real assent, equivalent to an actual living belief and faith in
God, there has been a free human response to an idea which
could instead have been rejected b y being held at the
notional level. In other words, a verbal proof of God's
existence cannot by itself break down our human freedom; it
can only lead to a notional assent which has little or no
positive religious value or substance.
I conclude, then, that the theological objections to the
theistic proofs are considerably less strong than the philo-
sophical ones; and that theologians who reject natural
theology would therefore do well to do so primarily on
philosophical rather than on theological grounds. These
philosophical reasons are, as we have seen, very strong; and
we therefore now have to consider whether, in the absence of
any theistic proofs, it can nevertheless be rational to believe
in the existence of God.

(b) Can there be rational theistic belief without proofs?

During the period dominated by the traditional theistic
arguments the existence of God was often treated by
philosophers as something to be discovered through
reasoning. It was seen as the conclusion of an inference; and
the question of the rationality of the belief was equated with
that of the soundness of the inference. But from a religious
point of view, as we have already seen, there has always been
something very odd about this approach. The situation which
it envisages is that of people standing outside the realm of
faith, for whom the apologist is trying to build a bridge of
rational inference to carry them over the frontier into that
107
realm. But of course this is not the way in which religious
faith has originally or typically or normally come about.
When the cosmological, ontological, teleological and moral
arguments were developed, theistic belief was already a
functioning part of an immemorially established and
developing form of human life. The claims of religion are
claims made by individuals and communities on the basis of
their experience — and experience which is none the less their
own for occuring within an inherited framework of ideas. We
are not dealing with a merely conceivable metaphysical
hypothesis which someone has speculatively invented but
which hardly anyone seriously believes. We are concerned,
rather, with convictions born out of experience and
reflection and living within actual communities of faith and
practice. Historically, then, the philosophical 'proofs' of God
have normally entered in to support and confirm but not to
create belief. Accordingly the proper philosophical approach
would seem to be a probing of the actual'foundations and
structure of a living and operative belier rather than of
theoretical and non-operative arguments subsequently formu-
lated for holding those beliefs. The question is not whether it
is possible to prove, starting from zero, that God exists; the
question is whether the religious man, given tl^jdjjjin^tively
religious form of human existence in which he participates, is
properly entitled aTa'rationaTperson to believe what he does
believe?
At this point we must consider what we mean by a rational
belief. If by a belief we mean a proposition believed, then
what we are to be concerned with here are not rational
beliefs but rational believings. Propositions can be well-
formed or ill-formed, and they can be true or false, b u t they
cannot be rational or irrational. It is people who are rational
or irrational, and derivately their states and their actions,
including their acts and states of believing. Further, apart
from the believing of analytic propositions, which are true by
definition and are therefore rationally believed by anyone
who understands them, the rationality of acts (or states) of
believing has to be assessed separately in each case. For it is a
function of the relation between the proposition believed and
the evidence on the basis of which the believer believes it. It
might conceivably be rational for Mr X to believe p but n o t
108
rational for Mr Y to believe p, because in relation to the data
available to Mr X p is worthy of belief but not in relation to
the date available to Mr Y. Thus the question of the
rationality of belief in the reality of God is the question of
the rationality of a particular person's believing, given the
data that he is using; or that of the believing of a class of
people who share the same body of data. Or putting the same
point the other way round, any assessing of the belief-
worthiness of the proposition that God exists must be an
assessing of it in relation to particular ranges of data.
Now there is one area of data or evidence which is
normally available to those who believe in God, and that
provides a very important part of the ground of their
believing, but which is normally not available to and there-
fore not taken into account by those who do not so believe;
and this is religious experience. It seems that the religious
man is in part basing his believing upon certain data of
religious experience which the non-religious man is not using
because he does not have them. Thus our question resolves
itself into one about the theist's right, given his distinctively
religious experience, to be certain that God exists. It is the
question of the rationality or irrationality, the well-
groundedness or ill-groundedness, of the religious man's, claim
to know God. The theist cannot hope to prove that God
exists; but despite this it may nevertheless be possible for him
to show it to be wholly reasonable for him to believe that
God exists.
What is at issue here is not whether it is rational for some-
one else, who does not participate in the distinctively
religious mode of experience, to believe in God on the basis
of the religious man's reports. I am not proposing any kind of
'argument from religious experience' by which God is
inferred as the cause of the special experiences described by
mystics and other religious persons. It is not the non-religious
man's theoretical use of someone else's reported religious
experience that i* to be considered, but the religious man's
own practical use of it. The question is whether he is acting
rationally in trusting his own experience and in proceeding to
live on the basis of it.
In order to investigate this question we must consider what
counts as rational belief in an analogous case. The analogy
109
that I propose is that between the religious person's claim to
be conscious of God and any man's claim to be conscious of
the physical world as an environment, existing independently
of himself, of which he must take account.
In each instance a realm of putatively cognitive experience
is taken to be veridical and is acted upon as such, even
though its veridical character cannot be logically demon-
strated. So far as sense experience is concerned this has
emerged b o t h from the failure of Descartes' attempt to
provide a theoretical guarantee that our senses relate us to a
real material environment, and from the success of Hume's
attempt to show that our normal non-solipsist belief in an
objective world of enduring objects around us in space is
neither a product of, nor justifiable by, philosophical
reasoning but is what has been called in some expositions of
Hume's thought (though the term does not seem to have
been used by Hume himself) a natural belief. It is a belief
which naturally and indeed inevitably arises in the normal
human mind in response to normal human perceptual
experience. It is a belief on the basis of which we live and the
rejection of which, in favour of a serious adoption of the
solipsist alternative, would so disorient our relationship to
other persons within a common material environment that
we should be accounted insane. Our insanity would consist in
the fact that we should no longer regard other people as
independent centres of consciousness, with their own
purposes and wills, with whom interpersonal relationships are
possible. We should instead be living in a one-person world.
It is thus a basic truth in, or a presupposition of, our
language that it is rational or sane to believe in the reality of
the external world that we inhabit in common with other
people, and irrational or insane not to do so.
What are the features of our sense experience in virtue of
which we all take this view? They would seem to be twofold:
the givenness or the involuntary character of this form of
cognitive experience, and the fact that we can and do act
successfully in terms of our belief in an external world. That
is to say, being built and circumstanced as we are we cannot
help initially believing as we do, and our belief is not
contradicted, but on the contrary continuously confirmed,
by our continuing experience. These characteristics jointly
110
constitute a sufficient reason to trust and live on the basis of
our perceptual experience in the absence of any positive
reason to distrust it; and our inability to exclude the
theoretical possibility of our experience as a whole being
purely subjective does not constitute such a reason. This
seems to be the principle on which, implicitly, we proceed.
And it is, by definition, rational to proceed in this way. That
is to say, this is the way in which all human beings do
proceed and have proceeded, apart from a very small
minority who have for that very reason been labelled by the
majority as insane. This habitual acceptance of our per-
ceptual experience is thus, we may say, part of our operative
concept of human rationality.
We can therefore now ask whether a like principle may be
invoked on behalf of a parallel response to religious
experience. 'Religious experience' is of course a highly elastic
concept. Let us restrict attention, for our present purpose, to
the theistic 'sense of the presence of God', the putative
awareness ot a transcendent divine Mind within whose field
of consciousness we exist and with whom therefore we
stand in a relationship of mutual awareness. This sense of
'living in the divine presence' does not take the form of a
direct vision of God, b u t of experiencing events in history
and in our own personal life as the medium of God's dealings
with us. Thus religious differs from non-religious experience,
not as the awareness of a different world, b u t as a different
way of experiencing the same world. Events which can be
experienced as having a purely natural significance are
experienced by the religious mind as having also and at the
same time religious significance and as mediating the presence
and activity of God. (14)
It is possible to study this type of religious experience
either in its strongest instances, in the primary and seminal
religious figures, or in its much weaker instances in ordinary
adherents of the traditions originated by the great exemplars
of faith. Since we are interested in the question of the claims
which religious experience justifies it is appropriate to look at
that experience in its strongest and purest forms. A
description of this will accordingly apply only very partially
to the ordinary rank-and-file believer either of today or in the
past.
Ill
 If then we consider the sense of living in the divine
presence as this was expressed by, for example, Jesus of
Nazareth, or by St Paul, St Francis, St Anselm or the great
prophets of the Old Testament, we find that their 'awareness
of God' was so vivid that he was as indubitable a factor in
their experience as was their physical environment. They
could no more help believing in the reality of God than in the
reality of the material world and of their human neighbours.
Many of the pages of the Bible resound with the sense of
God's presence as a building might reverberate from the tread
of some gigantic being walking through it. God was known to
the prophets and apostles as a dynamic will interacting with
their own wills; a sheerly given personal reality, as in-
escapably to be reckoned with as destructive storm and
life-giving sunshine, the fixed contours of the land, or the
hatred of their enemies and the friendship of their neigh-
bours.
Our question concerns, then, one whose 'experience of
God' has this compelling quality, so that he is no more
inclined to doubt its veridical character than to doubt the
evidence of his senses. Is it rational for him to take the
former, as it is certainly rational for him to take the latter, as
reliably cognitive of an aspect of his total environment and
thus as knowledge in terms of which to act? Are the two
(\.) **————-
features noted above in our seyise experience — its givenhess,
or involuntary character,andHUe fact thaFwe can successfully
act in terms of iF — also fo"und here? It seems that they are.
The sense of the presence of God reported by the great
religious figures has a similar involuntary and compelling
quality; and as they proceed to live on the basis of it they are
sustained and confirmed by their further experiences in the
conviction that they are living in relation, not to illusion, but
to reality. It therefore seems prima facie, that the religious
man is entitled to trust his religious experience and to
proceed to conduct his life in terms of it.
The analogy operating within this argument is between our
normal acceptance of our sense experience as perception of
an objective external world, and a corresponding acceptance
of the religious experience of 'living in God's presence' as the
awareness of a divine reality external to our own minds. In
each case there is a solipsist alternative in which one can
112
affirm solus ipse to the exclusion of the transcendent — in
the one case denying a physical environment transcending
our own private consciousness and in the other case denying
a divine Mind transcending our own private consciousness. It
should be noted that this analogy is not grounded in the
perception of particular material objects and does not turn
upon the contrast between veridical and illusory sense
perceptions, but is grounded in our awareness of an objective
external world as such and turns upon the contrast between
this and a theoretically possible solipsist interpretation of the
same stream of conscious experience.

(c) Religious and perceptual belief

Having thus set forth the analogy fairly boldly and starkly I
now want to qualify it by exploring various differences
between religious and sensory experience. The resulting
picture will be more complex than the first rough outline
presented so far; and yet its force as supporting the
rationality of theistic faith will not, I think, in the end have
been undermined.
The most obvious difference is that everyone has and
cannot help having sense experiences, whereas not everyone
has religious experiences, at any rate of the very vivid and
distinct kind to which we have been referring. As bodily
beings existing in a material environment, we cannot help
interacting consciously with that environment. That is to say,
we cannot help 'having' a stream of sense experiences; and we
cannot help accepting this as the perception of a material
world around us in space. When we open our eyes in daylight
we cannot but receive the visual experiences that come to us;
and likewise with the other senses. And the world which we
thus perceive is not plastic to our wishes but presents itself to
us as it is, whether we like it or not. Needless to say, our
senses do not coerce us in any sense of the word 'coerce' that
implies unwillingness on our part, as when a policeman
coerces an unwilling suspect to accompany him to the police
station. Sense experience is coercive in the sense that we
cannot when sane believe that our material environment is
not broadly as we perceive it to be, and that if we did
113
 momentarily persuade ourselves that what we experience is
not there we should quickly be penalised by the environment
cA.>and indeed, if we persisted, destroy.by it. (15)
In contrast to this we are not obliged to interact
consciously with a spiritual environment. Indeed it is a
commonplace of much contemporary theology that God
does not force an awareness of himself upon mankind but
leaves us free to know him by an uncompelled response of
faith. And yet once a man has allowed himself freely to
become conscious of God — it is important to note — that
experience is, at its top levels of intensity, coercive. It creates
the situation of the person who cannot help believing in thei
reality of God. The apostle, prophet or saint may be soi
vividly aware of God that he can no more doubt the veracity
of his religious awareness than of his sense experience. Duringl
the periods when he is living consciously in the presence ofl
God, when God is to him the divine Thou, the question
whether God exists simply does not arise. Our cognitive!
freedom in relation to God is not to be found at this poinll
but at the prior stage of our coming t o be aware of him. Thel
individual's own free receptivity and responsiveness plays an
essential part in his dawning consciousness of God; b u t oncq
he has become conscious of God that consciousness can
possess a coercive and indubitable quality. (16)
It is a consequence of this situation that whereas everyone
perceives and cannot help perceiving the physical world, by
no means everyone experiences the presence of God. Indeed
only rather few people experience religiously in the vivid and
coercive way reported by the great biblical figures. And this
fact immediately suggests a sceptical question. Since those
who enjoy a compelling religious experience form such a
small minority of mankind, ought we not to suspect that
they are suffering from a delusion comparable with that of
the paranoiac who hears threatening voices from the walls or
the alcoholic who sees green snakes?
This is of course a possible judgement to make. But this
judgement should not be made a priori, in the absence of
specific grounds such as we have in the other cases
mentioned. And it would in fact be difficult to point to
adequate evidence to support this hypothesis. On the
contrary the general intelligence and exceptionally high
114
moral quality of the great religious figures clashes with any
analysis of their experience in terms of abnormal psychology.
Such analyses are not indicated, as is the parallel view of
paranoiacs and alcoholics, by evidence of general dis-
orientation to reality or of incapacity to live a productive and
satisfying life. On the contrary, Jesus of Nazareth, for
example, has been regarded by hundreds of millions of
people as the fulfilment of the ideal possibilities of human
nature. A more reasonable negative position would therefore
seem to be the agnostic one that whilst it is proper for the
religious man himself, given his distinctive mode of
experience, to believe firmly in the reality of God, one does
not oneself share that experience and therefore has no
ground upon which to hold that belief. Theism is then not
positively denied, but is on the other hand consciously and
deliberately not affirmed. This agnostic position must be
accepted by the theist as a proper one. For if it is reasonable
for one man, on the basis of his distinctively religious
experience, to affirm the reality of God it must also be
reasonable for another man, in the absence of any such
experience, not to affirm the reality of God.
The next question that must be raised is the closely
connected one of the relation between rational belief and
truth. I suggested earlier that, strictly, one should speak of
rational believings rather than of rational beliefs. But never-
theless it is sometimes convenient to use the latter phrase,
which we may then understand as follows. By a rational
belief we shall mean a belief which it is rational for the one
who holds it to hold, given he data available to him. Clearly
such beliefs are not necessarily or always true. It is sometimes
rational for an individual to have, on the basis of incomplete
data, a belief which is in fact false. For example, it was once
rational for people to believe that the sun revolves round the
earth; for it was apparently perceived to do so, and the
additional theoretical and observational data were not yet
available from which it has since been inferred that it is the
earth which revolves round the sun. If, then, a belief may be
rational and yet false, may not the religious man's belief be
of this kind? May it not be that when the data of religious
experience are supplemented in the believer's mind by
further data provided by the sciences of psychology or
F. II.A.F.C;. 115
sociology, it ceases to be rational for him to believe in God?
Might it not then be rational for him instead to believe that
his 'experience of the presence of God' is to be understood as
an effect of a buried infancy memory of his father as a
benevolent higher power; or of the pressure upon him of the
human social organism of which he is a cell; or in accordance
with some other naturalistic theory of the nature of religion?
Certainly this is possible. Indeed we must say, more
generally, that all our beliefs, other than our acceptance of
logically self-certifying propositions, are in principle open to
revision or retraction in the light of new data. It is always
conceivable that something which it is now rational for us to
believe, it may one day not be rational for us to believe. But
the difference which this general principle properly makes to
our present believing varies from a maximum in relation to
beliefs involving a considerable theoretical element, such as
the higher-level hypotheses of the sciences, to a minimum in
relation to perceptual beliefs, such as the belief that I now
see a sheet of paper before me. And I have argued that so far
as the great primary religious figures are concerned, belief in
the reality of God is closer to the latter in that it is analogous
to belief in the reality of the perceived material world. It is
not an explanatory hypothesis, logically comparable with
those developed in the sciences, but a perceptual belief. God
was not, for Amos or Jeremiah or Jesus of Nazareth, an
inferred entity but an experienced personal presence. If this
is so, it is appropriate that the religious man's belief in the
reality of God should be no more provisional than his belief
in the reality of the physical world. The situation is in each
case that, given the experience which he has and which is part
of him, he cannot help accepting as 'there' such aspects of his
environment as he experiences. He cannot help believing
either in the reality of the material world which he is
conscious of inhabiting, or of the personal divine presence
which is overwhelmingly evident to him and to which his
mode of living is a free response. And I have been suggesting
that it is as reasonable for him to hold and to act upon the
one belief as the other.

116
(d) The problem of conflicting religious beliefs

We must now take note of another circumstance which

qualifies and threatens to erode our analogy. What are we to
make of the immense variety of the forms of religious
experience, giving rise as they do to apparently incompatible
beliefs? In contrast to this, human sense experience reveals a
world which is public in that normally the perceptions of any
two individuals can readily be correlated in terms of the
hypothesis of a common world which they jointly inhabit.
The variety commonly brought under the name of religion
is indeed as wide as the range of man's cultural and psycho-
logical diversities. By no means all religious experience is
theistic; ultimate reality is apprehended as non-personal and
as multi-personal as well as unipersonal. And if we choose to
extend the notion of religious experience, as Abraham
Maslow has recently done by his concept of peak-
experiences, (17) the variety is multiplied again. But even
apart from this last expansion of the field it is clearly true
that religious experience is bewilderingly varied in content
and that the different reports to which it gives rise cannot
easily be correlated as alternative accounts of the same
reality. And therefore since one could restate the argument
of the earlier part of this chapter from the point of view of
many different religions, with their different forms of
religious experience and belief, the question arises whether
the argument does not prove too much. In establishing the
rationality of the Judaic-Christian theist's belief in the reality
of God, must it not also and equally establish the rationaltiy
of the Buddhist's belief, arising out of his own coercive
religious experience, and likewise of Hindu belief and of
Islamic belief, and so on?
We need, I think, have no hesitation in accepting this
implication. The principle which I have used to justify as
rational the faith of a Christian who on the basis of his own
religious experience cannot help believing in the reality of
'the God and Father of our Lord Jesus Christ', also operates
to justify as rational the faith of a Muslim who on the basis
of his religious experience cannot help believing in the reality
of Allah and his providence; and the faith of the Buddhist
who on the basis of his religious experience cannot help
Yl H.A.F.G. 117
accepting the Buddhist picture of the universe; and so on.
But this is not the end of the matter. Various possibilities
now open before us. I can only in conclusion attempt a small-
scale map of the different paths that may be taken, showing
in what direction they each lead and forecasting to some
extent the kind of difficulties that are to be expected if one
chooses to travel along them.
The first fork in the road is constituted by the alternative
possibilities that the truth concerning the nature of the
universe will, and that it will not, ultimately be a matter of
public knowledge. The question is whether there will
eventually be a situation in which all rational persons will
find themselves obliged to agree, on the basis of a common
body of experience, that the universe has this or that specific
character. The issue, in other words, is that of the ultimate
public verifiability and falsifiability of religious faiths.
On the one hand, in one conceivable picture of the
universe it is possible for adherents of different and in-
compatible faiths to remain, so long as they continue to exist
and to hold beliefs, under the impression that their own
understanding of the universe is true; for they never meet an
experiential crux which either verifies or falsifies their faith.
This is a not always acknowledged feature of the pictures
adopted both by the non-eschatological religions and by most
atheistic and naturalistic theories. On the other hand, in
another possible picture of the universe, or rather family of
pictures painted by the different eschatological religions, the
future development of human experience will narrow down
the options until eventually only one faith is compatible with
the facts and it becomes irrational to hold any contrary view.
Thus it is affirmed in Christianity, in Islam, in one type of
Judaism and perhaps in one type of Buddhism that the
universe has a certain definite structure and is moving
towards a certain definite fulfilment such that in the light of
that fulfilment it will be beyond rational doubt that the
universe has the particular character that it has.
Both types of universe are logically possible. If Christianity
is true we are living in a universe of the latter type, in which
religious faith is ultimately verified; and since we are now
investigating the rationality of the Christian belief in God we
shall want at this first fork to take the verifiability-of-faiths
118
option in order to explore it further and to see where it leads.
Travelling along this path, then, we now meet a second
fork in the road, offering two rival conceptions of the
relations between the different religions. Along one path we
affirm the ultimate compatibility of the plurality of religious
faiths, whilst along the other path we deny this. The latter,
incompatibility thesis leads us to the following picture: it is
at the moment rational for adherents of different religions,
whose experience is such that they cannot help believing as
they do, to hold their respective beliefs. But — still assuming
the verifiability-of-faiths thesis — it will eventually cease to
be possible for rational persons to adhere to rival and
incompatible understandings of the universe. For according
to this option in its strongest form, there is one true faith and
many false ones — this view corresponding of course to the
traditional dogmatic stances of the eschatological religions,
such as Christianity and Islam. There is however a specifically
Christian reason for abandoning this stance. This is that belief
in the redeeming love of God for all his human creatures
makes it incredible that the divine activity in relation to man-
kind should have been confined to those within the reach of
the influence of the Christian revelation. The majority of the
human beings who have existed since man began have lived
either before or outside the historical influence of Tesus of
Nazareth. Thus the doctrine that there is no salvation outside
historic Christianity would in effect deny the universal love
and redeeming activity of God-
Any modification of that traditional claim soon leads us
over onto the alternative path, at the end of which lies the
conclusion that the different forms of religious experience,
giving rise to the different religions of the world, are properly
to be understood as experiences of different aspects of one
immensely complex and rich divine reality. If this is so, the
beliefs of the different religions will be related to a larger
truth as the experiences which gave rise to those beliefs are
related to a larger reality.
The further exploration of this possibility would take us
beyond our present necessarily limited inquiry. I have argued
that when on the basis of his own compelling religious
experience someone believes in the reality of God, he is
believing rationally; and I have added the rider that when we
119
set alongside this argument the fact of the plurality of
religions and their forms of religious experience, we are led to
postulate a divine reality of which the different religions of
the world represent different partial experiences and partial
knowledge. This latter possibility remains, however, to be
adequately developed and examined.

120
Notes

INTRODUCTION

1. For an account of this analysis see below, pp. 82-3.

2. 'Systematic Theology', I (Chicago: University of
Chicago Press, 1951) p. 237; (London: Nisbet, 1953) pp.
262-3. Cf. J o h n E. Smith, 'Experience and God' (New York
and London: Oxford University Press, 1968) pp. 118-20.
3 'The Existence of God', Russell 'Why I am Not a
Christian' (London: Allen & Unwin, 1957) p. 151.
4. Parts of my Introduction to 'The Existence of God' are
reproduced, with the publishers' permission, in this Intro-
duction and in Ch. 2 (c) and Ch. 7 (a).

CHAPTER 1

1. One could distinguish between a design argument with

a basis restricted to the fact that the universe exhibits order,
being a cosmos rather than a chaos, and one which goes
beyond this to stress the teleological character of the
evolution of the universe in producing man and the values of
which he is conscious. I shall not however discuss these two
varieties separately, since the strongest design arguments
comprehend both considerations.
2. Plato, 'Laws', X.
3. Cicero, 'On the Nature of the Gods', II v, trans. C. D.
Yonge (London: Bell, 1911) p. 50.
4. A series of lectures on Christian apologetics founded
by a bequest from Robert Boyle (1627-91).
5. 'Philosophical Theology', II (Cambridge University
Press, 1930).
6. 'Does God Exist?' (London: Macmillan, 1945).
7. 'Human Destiny' (London and New York: Longmans,
Green, 1947).
8. 'Metaphysics' (Englewood Cliffs, N.J., and London:
Prentice-Hall, 1963).
121
 9. 'Natural Theology', ch. 1.
10. Whether Paley's point constitutes an adequate reply to
Hume's will be considered presently, in note 25.
11. 'Natural Theology', Ch. 1.
12. Ibid.
13. 'Natural Theology', ch. 3.
14. Ibid.
15. William Derham, 'Physico-Theology', 9th ed. (London,
1737) pp. 257-8.
16. William Derham, 'Astro-Theology', 8th ed. (London,
1741) pp. 124-5.
17. Arthur I. Brown, 'Footprints of God' (Findlay, Ohio:
Fundamental Truth Publishers, 1943) p. 102.
18. Cf. N. Kemp Smith, introduction to 'Hume's Dialogues
concerning Natural Religion' (Oxford: Clarendon Press,
1935) app. C.
19. For a survey of theories about whom Cleanthes repre-
sents (for example Butler, or Berkeley), see Anders Jeffner,
'Butler and Hume on Religion' (Stockholm: Acta Uni-
versitatis Upsaliensis, Studia Doctrinae Christianae, No. 7,
1966) pp. 131 f.
20. Kemp Smith, p. 76. Hume's writings on religion are all
in the same sceptical vein and style as that of Philo's speeches
in the 'Dialogues'. Further, many of Philo's specific criticisms
of the design argument occur in the 'Enquiry concerning the
Human Understanding' (sect. XI). It is further evidence of
Hume's desire to avoid direct formal responsibility for these
criticisms, that here too he pretends to report them from a
'conversation with a friend who loves sceptical paradoxes'.
(Humes' 'Enquiries', ed L. A. Selby-Bigge, 2nd ed. (Oxford:
Clarendon Press, 1902) p . 132.) Burton F. Porter (in 'Deity
and Morality' (London: Allen & Unwin, 1968) pp. 39-40)
lists works in which Hume has been identified with Philo and
Cleanthes respectively.
2 1 . Pt II: Kemp Smith, p. 178.
22. Pt VII: Kemp Smith, p. 219.
23. Pt II: Kemp Smith, p. 183.
24. Pt IV: Kemp Smith, pp. 199-200. It is possible that
Paley's fifth point (p. 4 above) was intended as a reply to this
argument of Hume's — probably not however en-
countered in Hume's own writings (on this, see the next
122
note) but as used by someone else. But if so, no effective
reply is made. For Hume was not suggesting that e.g. metal
may form itself into watches by spontaneous motion, but
rather that on the cosmic scale self-ordering may well be a
property of matter. For we see in nature mind arising out of
matter (since the human brain is formed within a biological
process), as well as matter being ordered by mind.
25. Pt VIII: Kemp Smith, pp. 227-8. This is an argument
against which Paley's fourth point (p. 4 above) may possibly
have been directed. It would indeed be absurd, as Paley says,
to suggest that since the matter composing a watch must take
some form or other it is antecedently as likely to constitute a
watch as anything else. But this is no answer to Hume's
point, which is simply that any universe (as distinguished
from a chaos) must be orderly and to that extent 'as though
designed'. It therefore seems to me unlikely that Paley can
have studied the 'Dialogues' for himself, although he may
well have met elsewhere versions of some of Hume's
arguments which had entered into the stream of public
discussion. But this is conjecture; I know of no direct
evidence.
26. Julian Huxley, 'The Evolutionary Process' in
'Evolution as a Process', ed. Julian Huxley, A. C. Hardy and
E. B. Ford, 2nd ed. (London: Allen & Unwin, 1958) p . 4.
27. Ibid., p. 3. For a detailed account for the layman, see
J o h n Maynard Smith, 'The Theory of Evolution', 2nd ed.
(Penguin Books, 1966).
28. 'An Enquiry Concerning Human Understanding', sect.
XI: Selby-Bigge, p. 136. 'Dialogues', pt. V: Kemp Smith, p.
208.
29. Pt V: Kemp Smith, p. 207.
30. Pt V: Kemp Smith, p. 207.
3 1 . Pt V: Kemp Smith, p . 209.
32. Pt II: Kemp Smith, p. 185.
33. Alvin Plantinga, 'God and Other Minds' (Ithaca, N.Y.:
Cornell University Press, 1967) p . 101.
34. Cf. N. Kemp Smith, 'A Commentary to Kant's
"Critique of Pure Reason" ', 2nd ed. (London: Macmillan,
1923) p. 539, n. 3.
35. 'Human Destiny', p. 36.
36. Ibid., p. 202.
123
 37. Ibid., p . 189.
38. Ibid., pp. 34-6.
39. Wallace I. Matson, 'The Existence of God' (Ithaca,
N.Y.: Cornell University Press, 1965) pp. 102-11.
40. Ibid., p. 106.
4 1 . Ibid., p . 107.
42. This research is described in A. O. Oparin, 'The Origin
of Life on the Earth', trans Ann Synge, 3rd ed. (Edinburgh:
Oliver & Boyd, 1957). See also J. Marquand, 'Life: Its
Nature, Origins and Distribution' (Edinburgh: Oliver & Boyd,
1968).
43. Sidney Fox (Director of the Institute of Molecular
Evolution, University of Miami), 'In the beginning . . . life
assembled itself, 'New Scientist', 27 Feb 1969, p. 450.

CHAPTER 2

1. Though on the day on which this was written it was

reported ('Guardian', 26 July 1968) that the Nederduits
Gereformeerde Church, the largest of the three Dutch
Reformed Churches in South Africa, has called for an
immediate halt to the teaching of evolution in schools and
universities!
2. 'Philosophical Theology', II 79.
3. Ibid., p. 8 1 .
4. Ibid., p. 82.
5. Ibid., p. 83.
6. Ibid., p. 84.
7. Ibid., p. 85.
8. Ibid.
9. Ibid., p. 86.
10. Ibid., p. 87.
11. Ibid., p. 89.
12. Ibid., p. 93.
13. Ibid., p. 103.
14. Ibid., p. 113.
15. 'Metaphysics', p. 96.
16. Ibid., p. 96-7.
17. Ibid., p. 99.
18. Ibid., p . 100.
124
 19. Ibid., pp. 100-1.
20. Ibid., p. 101.
2 1 . Smith, 'The Theory of Evolution', p. 308.
22. For a slightly different criticism of Taylor's arguments,
see J a n Narveson, 'On a new argument from design', 'Journal
of Philosophy', LXII 9 (29 April 1965).
23. 'Philosophical Theology', II 245.
24. 'Theism', 'Encyclopaedia Britannica' (1962) XXII 50.
25. 'Towards Belief in God' (London S.C.M. Press, 1942)
p. 112.
26. Cf. Arthur Pap, 'Elements of Analytic Philosophy'
(New York: Macmillan, 1949) pp. 199-200.
27. 'Philosophical Theology', II 87.
28. Ibid., II 88.
29. Ibid., I 283.
30. 'Leviathan', pt 3, ch. 32.
31. 'Critique of Pure Reason', trans. Kemp Smith
(London: Macmillan, and New York: St Martin's Press, 1933)
p. 520.
32. Ibid.

CHAPTER 3

1. Leibniz, 'On the Ultimate Origination of Things',

'Philosophical Writings', ed. Mary Morris (London: Dent, and
New York: Dutton, 1934) p . 32.
2. Aristotle, 'Physics', VII 1. Cf. 'Metaphysics', XII.
3. 'Summa Theologiae', ia 2, 3, trans. Timothy
McDermott, O.P., new Dominican translation of the 'Summa
Theologiae' (London: Eyre & Spottiswoode, and New York:
McGraw-Hill, 1964) II 13-15.
4. These three kinds of motus were distinguished by
Aristotle in 'Physics', V 226a.
5. Anthony Kenny, 'The Five Ways' (London: Routledge
& K e g a n P a u l , 1969) ch. 2.
6. Ibid., p. 19.
7. Ibid., p. 2 1 .
8. Ibid., p. 22.
9. Cf. 'Summa Theologiae', Ia, q. 46, art. 2; 'Summa
contra Gentiles', II, chs 31-8.
125
 10. A. D. Sertillanges, 'Sources de la Croyance en Dieu'
(Paris: Perrin, 1905) pp. 64-72; R. Garrigou-Lagrange, 'God,
His Existence and His Nature', trans. Dom Bede Rose (St
Louis, Mo., and London: Herder, 1934-6) I 264 f.; E. L.
Mascall, 'He Who Is: A Study in Traditional Theism' (London
and New York: Longmans, Green 1943) ch. 5.
11. 'Summa Theologiae', I 15.
12. Cf. Kenny, 'The Five Ways', p. 40.
1 3 . ' H e Who Is', p. 46.
14. 'Aquinas' (Penguin Books, 1955) p . 118.
15.'Aquinas', pp. 118-19.
16. 'Summa Theologiae', 1 1 5 .
17. In the version in the 'Summa Contra Gentiles' (I, ch.
15) Aquinas omits this stage of his argument.
18. Antony Flew, 'God and Philosophy' (London and New
York: Hutchinson, 1966) pp. 101-2.
19. The principle of sufficient reason, 'in virtue of which
we hold that there can be no fact real or existing, no state-
ment true, unless there be a sufficient reason, why it should
be so and not otherwise, although these reasons usually
cannot be known to us': Leibniz, 'Monadology', 32.
20. Russell, 'Why I am not a Christian', p. 152.
2 1 . Copleston, 'Aquinas', p . 124.

CHAPTER 4

1. 'Critique of Pure Reason', trans. Kemp Smith, p. 29.

2. 'Fundamental Principles of the Metaphysics of Morals',
ch. 1, trans. H . J . Paton, 'The Moral Law' (London:
Hutchinson, 1947) p. 6 1 .
3. 'Critique of Practical Reason', trans. L. W. Beck (New
York: Liberal Arts Press, 1956) p . 133.
4. Ibid., pp. 114-15.
5. Ibid., p . 119.
6. Ibid., p. 130.
7. Ibid., p . 130.
8. Ibid., p. 129.
9. Ibid., p. 133.
10. Ibid., p. 126.
11. 'The Theory of Good and Evil' (Oxford: Clarendon
Press, 1907) II 208.
126
 12. Ibid., 11212-13.
13. Rashdall adds in a footnote: 'Or at least a mind by
which all Reality is controlled.'
14. Ibid., 11211-12.
15. 'A Grammar of Assent', ed. C. F. Harrold (New York
and London: Longmans, Green, 1947) p. 83.
16. London: Allen & Unwin, 1954.
17. 'A Free Man's Worship', in 'Mysticism and Logic'
(London: Longmans, Green, 1918) pp. 47-8. In 1962 Russell
commented on this early essay that whilst he regarded its
style as 'florid and rhetorical', nevertheless 'my outlook on
the cosmos and on human life is substantially unchanged':
'Autobiography' (London: Allen & Unwin, 1969) III 172-3.
18. Margaret Knight, 'Morality — Supernatural or Social?',
'The Humanist Outlook', ed. A. J. Ayer (London: Pember-
ton, 1968) p. 47.
19. Of course if life has become a burden to him — if, for
example, he is suffering the agony of a painful terminal
illness — it may be rational for a man to abandon his life. But
this is not the case that I am posing as a difficulty for
humanist philosophy.
20. I owe this argument to Dr H. Stopes-Roe, and the
counter to it to the Rev. Michael Goulder.

CHAPTER 5

1. 'Critique of Pure Reason', trans. Kemp Smith, p. 509.

2. Kant's argument about the dependence of the cosmo-
logical argument upon the ontological has been discussed,
with differing results, by several philosophers: J . J . C. Smart,
'The Existence of God', in 'New Essays in Philosophical
Theology', ed. Flew and Maclntyre (London: S.C.M. Press,
and New York: Macmillan, 1955); T. Johnston, 'A Note on
Kant's Criticism of the Arguments for the Existence of God',
'Australasian Journal of Philosophy', (1943); William H.
Baumer, 'Kant on Cosmological Arguments', 'The Monist'
(1967).
3. This name appears to have been given to the argument
by Kant, the term 'ontology' having begun in the seventeenth
century to be used to denote the 'science of being' and
127
having been popularised in the eighteenth century by
Christian Wolff. By an ontological argument Kant apparently
means an a priori argument leading to an existential con-
clusion.
4. It is clear from the references in Anselm's works that
he had studied the writings of St Augustine b o t h closely and
extensively and accorded to them a very special authority.
His high respect, almost veneration, for Augustine is
expressed in the preface to the 'Monologion', where Anselm
says: 'after frequent consideration, I have not been able to
find that I have made in [the 'Monologion'] any statement
which is inconsistent with the writings of the Catholic
Fathers, or especially with those of St Augustine.' (Trans.
S. N. Deane, 'St Anselm' (La Salle, 111.: Open Court, 1958)
p.36.)
5. 'On Free Will', II vi 14, trans. J . H. S. Burleigh in
'Augustine's Earlier Writings' (London: S.C.M. Press, and
Philadelphia: Westminster Press, 1953) p. 144.
6. After long intellectual struggles, 'suddenly one night
during matins the grace of God illuminated his heart, the
whole matter became clear to his mind, and a great joy and
exultation filled his inmost being.' ('Eadmer's Life of St
Anselm', I, 19, trans. R. W. Southern (London: Nelson,
1962) p. 30.)
7. Augustine summarises his argument as follows: 'You
admitted for your part that if I could show you something
superior to our minds you would confess that it was God,
provided nothing existed that was higher still. I accepted
your admission and said it would be sufficient if I demon-
strated that. If there is anything more excellent than wisdom, '
doubtless it, rather, is God. But if there is nothing more
excellent, then truth itself is God. Whether there is or is not
such a higher thing, you cannot deny that God exists, and
this was the question set for our discussion.' ('On Free Will',
II xv 39, Burleigh's trans., in 'Augustine's Earlier Writings',
pp. 159-60.)
8. 'Pros.' 2. Within the same chapter he uses four variant
forms: aliquid quo maius nihil cogitari potest, id quo maius
cogitari nequit, id quo maius cogitari non potest and aliquid
quo maius cogitari non valet. Elsewhere he uses yet other
minor variations, such as quo nil maius valet cogitari ('Pros.'
128
5.) It is thus clear that no one precise form of words was
essential to Anselm, and in discussing his argument I shall
follow his example and use a variety of phrases.
9. Psalms 14:1 and 5 3 : 1 .
10. Elsewhere he says, 'But I do not mean physically great,
as a material object is great, but that which, the greater it is,
is the better or more worthy.' ('Monologion', ch. 2.)
11. For example in 'Pros.' 14 and 18 Anselm uses the
phrase quo nihil melius cogitari potest, and in various other
places (e.g. 'Pros.' 3, 'Responsio', 8, 'Monol.' 4 and 15) he
moves between melius and maius, or bonum and magnum, in
a way which indicates their interchangeability within his
argument.
1 2 . ' P r o s . ' 15, Charlesworth's trans., 'St Anselm's
Proslogion' (Clarendon Press, 1965) Cf. 'Responsio', 9 and
'Monol.' 64 and 65.
13. Karl Barth rightly insists upon the negative character
of the formula. He says: 'All that the formula says about this
object is, as far as I can see, this one thing, this one negative:
nothing greater than it can be imagined; nothing can be
imagined that in any respect whatsoever could or would
outdo it; as soon as any conceives anything which in any
respect whatsoever is greater than it, in so far as it can be
conceived at all — then he has not yet begun to conceive it or
has already ceased.' ('Anselm: Fides Quaerens Intellectum'
(1931), trans. Ian W. Robertson (London: S.C.M. Press,
1960) p. 75.)
14. 'Pros.' 5 (Charlesworth's trans.).
15. There is a challenging discussion of this point in C. D.
Broad, 'Religion, Philosophy and Psychical Research'
(London: Routledge & Kegan Paul, 1953) pp. 177-80.
16. 'St Anselm's Proslogion', p . 57.
17. Leibniz's main contribution to the investigation of the
ontological argument was to point out that it is an
incomplete demonstration, 'which assumes something that
must still be proved in order to render it mathematically
evident; that is, it is tacitly assumed that this idea of the
all-great or all-perfect being is possible, and implies no con-
tradiction'. (Leibniz, 'New Essays Concerning Human Under-
standing', IV, ch. 10, trans. A. G. Langley (London:
Macmillan, 1896) p . 504.) Leibniz himself held that 'We have
129
the right to presume the possibility of every being, and
expecially that of God, until some one proves the contrary'
(Ibid.), and that the onus lying upon the critic to demon-
strate a contradiction in the concept of God has not been
discharged. He accordingly concluded that 'according to the
present state of our knowledge we must judge that God
exists' (Ibid.). In appendix X he goes further and seeks to
demonstrate the indemonstrability of the proposition, with
regard to any two perfections, that they are incompatible.
18. 'Pros.' 2 (my italics): McGill's translation in 'The
Many-faced Argument', ed. Hick and McGill (New York:
Macmillan, 1967, and London: Macmillan, 1968).
19. 'Republic', 509.
20. It is a consequence of the equation of being with good-
ness than an absolutely evil being cannot exist, for to be
absolutely devoid of goodness would be also to be absolutely
devoid of being. It follows (as C. K. Grant pointed out in his
'The Ontological Disproof of the Devil', 'Analysis', XVII (Jan
1957) 71-2) that there cannot be an absolutely evil devil. The
theologians who worked within the neoplatonist frame
accepted this implcation, holding that the devil is good qua
existent and qua powerful, etc., though wholly evil morally.
21. M . J . A. O'Connor 'New Aspects of Omnipotence and
Necessity in Anselm', 'Religious Studies', IV (1968) 136.
22. Ibid.
23. 'The Many-faced Argument', ch. 3.
24. Gaunilo's 'Pro Insipiente', 6, trans. McGill in 'The
Many-faced Argument', pp. 22-3.
25. 'Reply', 3, trans. McGill, ibid., p. 23.
26. 'Meditations', V, trans. Haldane and Ross, 'The Philo-
sophical Works of Descartes', I (Cambridge, 1911) p. 182.
27. Ibid.
28. Ibid., I 180-1.
29. Ibid., I 181. The mountain and valley example is not a
very good one, for there could be a mountain rising out of a
plain, with no valley associated with it.
30. Ibid.
31. Ibid.
32. Ibid., II 186.
33. Ibid.
34. Ibid., II 228.
130
 35. B 620-4 and B 625-30.
36. Kant's 'Critique of Pure Reason', trans. Kemp Smith,
p. 502.
37. Ibid., p. 504.
38. Ibid., p. 505.
39. Ibid., pp. 504-5.
40. Bertrand Russell and A. N. Whitehead, 'Principia
Mathematica', I (Cambridge, 1910) Introduction, ch. 1 and
pt I,, sect. B, p. 14; Bertrand Russell, 'The Existential Import
of Propositions', 'Mind' (1905) 398-401; 'On Denoting',
'Mind' (1905), reprinted in 'Readings in Philosophical
Analysis', ed. Herbert Feigl and Wilfred Sellars, (New York:
Appleton-Century Crofts, 1949); 'The Philosophy of Logical
Atomism' (1918) V and VI, in 'Logic and Knowledge', ed.
R. C. Marsh, (London: Allen & Unwin, 1956); 'Introduction
to Mathematical Philosophy' (London: Allen & Unwin,
1920) ch. 16; 'History of Western Philosophy', (London:
Allen & Unwin, 1946) pp. 859-60.
4 1 . Cf. G. E. Moore, 'Is Existence a Predicate?' in 'Philo-
sophical Papers' (London: Allen & Unwin, 1959) pp. 124-6.
42. Cf. P. Geach, 'Form and Existence' in 'Proceedings of
the Aristotelian Society' (1954-5) 262-8. Reprinted in Geach,
'God and the Soul' (London: Routledge & Kegan Paul, 1969)
pp. 54-60.

CHAPTER 6

1. E.g. J . N. Findlay, 'Can God's Existence be Dis-

proved?', 'Mind' (April 1948), reprinted in 'New Essays in
Philosophical Theology', ed. Flew and Macintyre, and else-
where; J . J . C. Smart, 'The Existence of God' in 'New Essays
in Philosophical Theology', pp. 38-9; and K. E. M. Baier, 'The
Meaning of Life' (Canberra, 1957) p. 8. Although, I think,
certainly mistaken, their idea is not absurd; for Avicenna had
previously used a concept of logically necessary being, from
which it seems that Aquinas was converted by Averroes'
arguments by the time that he wrote the 'Summa contra
Gentiles' and the 'Summa Theologiae'. (See Guy Jalbert,
'Necessite et contingence chez S. Thomas D'Aquin et ses
predecesseurs' (Ottawa, 1961).)
131
 2. Smart, in 'New Essays in Philosophical Theology' p.
40.
3. 'Responsio', 4: 'The Many-faced Argument', p. 28. Cf.
'Resp.' 1.
4. This paragraph embodies material from my article,
'Necessary Being', 'Scottish Journal of Theology' (Dec 1961).
A closely related conception of necessary being has recently
been related by Anthony Kenny to Aristotle's doctrine (in
his 'De Interpretatione') that propositions are not timelessly
true or false but may become true or cease to be true. Thus
'Theatetus is sitting' is true when Theatetus sits down but
becomes false when he stands up. And since God is defined as
an eternal being, without beginning or end, it follows that
there is no time at which 'God exists' is false. Hence 'God
exists' is a proposition which is true at all times and which is
in this sense necessarily true. (Anthony Kenny, 'God and
Necessity' in 'British Analytical Philosophy', ed. Bernard
Williams and Alan Montefiore (London: Routledge & Kegan
Paul, and New York: Humanities Press, 1966) pp. 147-51.)
Cf. P. M. Hutchings, 'Necessary Being and Some Types of
Tautology', 'Philosophy' (Jan 1964) 4-8; and Peter Geach in
Anscombe and Geach, 'Three Philosophers' (Oxford:
Blackwell, 1961) pp. 114-15, and 'God and the Soul', p. 77.
5. 'The Many-faced Argument', p. 355.
6. Cf. 'The Many-faced Argument', p. 356.
7. 'Recent Discussions of Anselm's Argument' in 'The
Many-faced Argument', pp. 41-2.
8. 'Anselm: Fides Quaerens Intellectum', pp. 132 f.
9. 'Philosophical Review', LXIX (Jan 1960), reprinted in
Malcolm's 'Knowledge and Certainty' (Englewood Cliffs,
N.J.: Prentice-Hall, 1963) and elsewhere.
10. '[Anselm's] second proof employs the . . . principle
that a thing is greater if it necessarily exists than if it does not
necessarily exist. Some remarks about the notion of
dependence may help to make this latter principle
intelligible. Many things depend for their existence on other
things and events. My house was built by a carpenter: its
coming into existence was dependent on a certain creative
activity. Its continued existence is dependent on many
things: that a tree does not crush it, that it is not consumed
by fire, and so on. If we reflect on the common meaning of
132
the word " G o d " . . . we realize that it is incompatible with
this meaning that God's existence should depend on any-
thing . . . . To conceive of anything as dependent upon
something else for its existence is to conceive of it as a lesser
being than God.' ('Knowledge and Certainty', pp. 146-7.)
11. 'Knowledge and Certainty', p. 149.
12.1 am still using 'eternal' as short for 'eternal and
independent'.
13. 'Knowledge and Certainty', p. 150.
14. Ibid., pp. 157-8.
15. 'Man's Vision of God' (Chicago, 1941; currently
republished by Archon Books, Hamden, Conn.) 9; 'The Logic
of Perfection' (La Salle, 111.: Open Court, 1962) ch. 2;
'Anselm's Discovery' (La Salle, 111. Open Court, 1965); 'The
Formal Validity and Real Significance of the Ontological
Argument', 'Philosophical Review' (1944); 'The Logic of the
Ontological Argument', 'Journal of Philosophy', (1961);
'What did Anselm Discover?', 'Union Seminary Quarterly
Review' (1962); and numerous smaller contributions.
16. Added at Hartshorne's request in the reprinting of this
passage in 'The Many-faced Argument' (p. 335).
17. 'The Logic of Perfection', pp. 50-1.
18. Ibid., p. 53.
19. 'Responsio', 4: McGill, p. 28.
20. E.g. 'Of those who claim to demonstrate that the
argument is a mere sophistry, the majority appear to have
read the first page or so (Chapter II), or at least a paraphrase
of it in some history, but one would be hard put to it in most
cases to furnish evidence that they had read more.'
('Anselm's Discovery', p. 1 1 . Cf. pp. 16-17.)
2 1 . The redundant terms, in brackets, although odd, seem
harmless.
22. 'Knowledge and Certainty', p. 150.
23. James F. Ross, 'Philosophical Theology' (New York:
Bobbs-Merrill, 1969) p. 173.
24. Ibid., p. 174.
25. Ibid., p. 176.

133
 CHAPTER 7

1. For a modern papal reaffirmation of the position that

'human reason can, without the help of divine revelation and
grace, prove the existence of a personal God by arguments
drawn from created things', see Pope Pius XII's encyclical
'Humani Generis' (1940), esp. paras 2, 3, 25 and 29.
2. See e.g. J . Oliver Buswell, 'What is God?' (Grand
Rapids, Michigan: Zondervan, 1937), and Robert E. D. Clark,
'The Universe: Plan or Accident?' (Philadelphia: Muhlenberg
Press, and London: Paternoster Press, 1961).
3. E.g. W. R. Matthews, 'Studies in Christian Philosophy',
2nd ed. (1928) and 'Theism' in 'Ency. Brit.' XXII 50.
4. See Abraham Heschel, 'God in Search of Man: A
Philosophy of Judaism' (New York: Jewish Publication
Society of America, 1955) pp. 246 f.; Martin Buber, 'Eclipse
of God' (New York: Harper & Row, 1952, and London:
Gonancy 1953) ch. 8.
5. E.g. J o h n Baillie, 'Our Knowledge of God' (London:
Oxford University Press, 1939) ch. 3, sect. 10.
6. 'Statement and Inference' (Oxford University Press,
1926)11853.
7. 'God and Evil: a study of some relations between faith
and morals', 'Ethics' (Jan 1958) 82.
8. 'Dialogues concerning Natural Religion', V.
9. 'Pensees', ed. Leon Brunschvicg, 430.
10. T. S. Eliot, 'Burnt Norton', I.
11. 'Metaphysical Beliefs' (London: S.C.M. Press, and New
York: Allenson, 1957) p. 197. Maclntyre makes the same
point in 'Difficulties in Christian Belief (London: S.C.M.
Press, 1960) p. 77.
12. J . H. Newman, 'A Grammar of Assent' (1870) ch. 4.
13. The exercise of this capacity is well described by C. S.
Peirce as the unfolding of what he calls 'the humble
argument' for the reality of God. See his 'Collected Papers'
(Harvard University Press, 1934) 6.467 and 6.486.
14. This view is developed in my 'Faith and Knowledge',
2nd ed. (Ithaca, N.Y.: Cornell University Press, 1966, and
London: Macmillan, 1967).
15. For a discussion of the notion of 'coerciveness' in sense
experience and religious experience, see Donald F. Henze,
134
'Faith, Evidence, and Coercion', 'Philosophy' (Jan 1967) and
J o h n Hick, 'Faith and Coercion', 'Philosophy' Quly 1967);
and D. R. Duff-Forbes, 'Faith, Evidence, Coercion', 'Austral-
asian Journal of Philosophy' (Aug 1969).
16. This is developed more fully in my 'Sceptics and
Believers' in 'Faith and the Philosophers', ed. J o h n Hick
(London: Macmillan, and New York: St Martin's Press,
1964).
17. Abraham H. Maslow, 'Religions, Values, and Peak-
Experiences' (Columbus, Ohio: Ohio State University Press,
1964).

135
Select Bibliography
(Works in foreign languages are listed under their English
translations when such exist.)

WRITINGS DISCUSSING TWO OR MORE OF

THE ARGUMENTS

Textbook expositions and discussions

H . J . Paton, 'The Modern Predicament' (London: Allen &

Unwin, and New York: Macmillan, 1955).
Geddes MacGregor, 'Introduction to Religious Philosophy'
(Boston' Mass.: Houghton Mifflin, and London:
Macmillan, 1959).
J o h n Hick, 'Philosophy of Religion' (Englewood Cliffs, N.J.,
and London: Prentice-Hall, 1963).
H. D. Lewis, 'Philosophy of Religion' (London: English
Universities Press, and New York: Barnes & Noble, 1965).
Thomas McPherson, 'The Philosophy of Religion' (New
York: Van Nostrand, 1965).
Frederick Ferre, 'Basic Modern Philosophy of Religion' (New
York: Scribner's, 1967, and London; Allen & Unwin, 1968).
J . Hospers, 'An Introduction to Philosophical Analysis' 2nd
ed. (Englewood Cliffs, N.J.: Prentice-Hall, and London:
Routledge & Kegan Paul, 1967) ch. 5.

Works advocating some of the arguments

A. E. Taylor, 'Theism', 'Encyclopedia of Religion and Ethics',
ed. Hastings, xii (Edinburgh, 1921, and New York:
Scribner's, 195 1).
, 'Does God Exist?' (London and New York: Macmillan,
1945).
J o h n Laird, 'Theism and Cosmology' (London: Allen &
Unwin, 1941).
, 'Mind and Deity' (London: Allen & Unwin, 1941).
G. H. Joyce, 'The Principles of Natural Theology' (London
and New York: Longmans, Green, 1951).
Richard Taylor, 'Metaphysics' (Englewood Cliffs, N.J., and
136
 London: Prentice-Hall, 1963) ch. 7.
A. C. Ewing, 'Two "Proofs" of God's Existence', 'Religious
Studies' (1965).
James F. Ross, 'Philosophical Theology' (New York: Bobbs-
Merrill, 1969).

Criticisms of the arguments

J . S. Mill, 'Three Essays on Religion' (London: Longmans,

Green, 1874).
C. D. Broad, 'Arguments for the Existence of God', i and ii,
'Journal of Theological Studies' (1939), reprinted in
'Religion, Philosophy and Psychical Research' (London:
Routledge & Kegan Paul, and New York: Harcourt, 1953).
J . J . C. Smart, 'The Existence of God', 'New Essays in Philo-
sophical Theology', ed. Flew and Maclntyre (London:
S.C.M. Press, and New York: Macmillan, 1955).
Bertrand Russell and F. C. Copleston, 'The Existence of God
— a debate', in Russell, 'Why I am not a Christian'
(London: Allen & Unwin, 1957). (Not included in U.S.
edition.)
Walter Kaufmann, 'Critique of Religion and Philosophy'
(New York: Harper, 1958, and London: Faber & Faber,
1959).
Wallace I. Matson, 'The Existence of God' (Ithaca, N.Y.:
Cornell U.P., 1965).
Antony Flew, 'God and Philosophy' (London and New York:
Hutchison, 1966).
Alvin Plantinga, 'God and Other Minds' (Ithaca, N.Y.: Cornell
U.P., 1967).

THE DESIGN ARGUMENT

Works defending some form of design argument

William Paley, 'Natural Theology' (1802), abridged ed. by F.

Ferre (Indianapolis: Bobbs-Merrill, 1963).
F. R. Tennant, 'Philosophical Theology' ii (Cambridge and
New York: Cambridge U.P., 1930) ch. 4.
A. E. Taylor, 'Does God Exist?' (London and New York:
137
 Macmillan, 1945).
Lecomte du Noiiy, 'Human Destiny' (New York: Longmans,
Green, 1947).
Peter Bertocci, 'Introduction to the Philosophy of Religion'
(Englewood Cliffs, N.J.: Prentice-Hall, 1951) ch. 11-15.
Robert E. D. Clark, 'The Universe: Plan or Accident?'
(Philadelphia: Muhlenberg Press, and London: Paternoster
Press, 1961).
Richard Taylor, 'Metaphysics' (Englewood Cliffs, N.J., and
London: Prentice-Hall, 1963) ch. 7.

Works criticising the design argument

David Hume, 'Dialogues concerning Natural Religion' (1779),

critical ed. by N. Kemp Smith, 2nd ed. (Edinburgh and
New York: Nelson, 1947).
Clarence Darrow, 'The Story of My Life' (New York:
Scribner's, 1932) ch. 4 4 .
Wallace I. Matson, 'The Existence of God' (Ithaca, N.Y.:
Cornell U.P., 1965).

Discussions of various aspects

D. L. Scudder, 'Tennant's Philosophical Theology' (New

Haven, Conn.: Yale U.P., and London: Oxford U.P.,
1940).
Antony Flew, 'Hume's Philosophy of Belief (London:
Routledge & Kegan Paul, and New York: Humanities
Press, 1961) ch. 9.
Anders Jeffner, 'Butler and Hume on Religion' (Stockholm:
Diakonistyrelsens Bokforlag, 1966).
J . Marquand, 'Life: Its Nature, Origins and Distribution'
(Edinburgh: Oliver & Boyd, and New York: W. A. Benjamin,
1968).
Norman Goldhawk, 'William Paley: or the Eighteenth
Century Revisited', Providence, ed. Maurice Wiles
(London: S.P.C.K., 1969).

138
 THE COSMOLOGICAL ARGUMENT

Works defending the cosmological argument

Thomas Aquinas, 'Summa Theologiae', pt i, Q. 2, art. 3.

R. Garrigou-Lagrange, 'God, His Existence and His Nature',
trans. Bede Rose, 2 vols (St Louis, Mo., and London:
Herder, 1934-6).
R. P. Phillips, 'Modern Thomistic Philosophy', ii (West-
minster Md: Newman Press, 1935).
Austin Farrer, 'Finite and Infinite' (London: Dacre, 1943, 2nd
ed. 1959, and Naperville, 111.: Allenson, 2nd ed., 1964).
E. L. Mascall, 'He Who Is' (London and New York:
Longmans, Green, 1948).
Fernand Van Steenberghen, 'Ontology', trans. M . J . Flynn
(New York: Wagner, and London: Herder, 1952).
Dom Illtyd Trethowan, 'An Essay in Christian Philosophy'
(London and New York: Longmans, Green, 1954).
Samuel M. Thompson, 'A Modern Philosophy of Religion'
(Chicago: Henry Regnery, 1955) pt 6.
Victor Preller, 'Divine Science and the Science of God: A
Reformulation of Thomas Aquinas' (Princeton: Princeton
U.P., 1967).

Works criticising the cosmological argument

David Hume, 'Dialogues Concerning Natural Religion' (1779)

pt 9.
Paul Edwards, 'The Cosmological Argument', 'The Ration-
alist A n n u a l ' ( 1 9 5 9 ) .
W. E. Kennick, 'A New Way with the Five ways', 'Austral-
asian Journal of Philosophy' (1960).
Anthony Kenny, 'The Five Ways' (London: Routledge &
Kegan Paul, and New York: Schocken, 1969).

Discussions of various aspects

Ronald Hepburn, 'From World to God'. 'Mind' (1963).

, 'Cosmological Arguments for the Existence of God',
'The Encyclopedia of Philosophy', ii (New York, 1967).
139
William H. Baumer, 'Kant on Cosmological Arguments', 'The
Monist' (1967).
Ninian Smart, 'Philosophers and Religious Truth', 2nd ed.
(London: S.C.M. Press, 1969) ch. 4.
Michael Durrant, 'St Thomas' "Third Way" ', 'Religious
Studies'(1969).

MORAL ARGUMENTS

Works advocating some form of moral argument

Immanuel Kant, 'Critique of Practical Reason' (1788) ii, ch.
2, trans. L. W. Beck (New York: Liberal Arts Press, 1956).
J o h n Henry Newman, 'A Grammar of Assent' (1870).
Hastings Rashdall, 'The Theory of Good and Evil' (Oxford
U.P., 1907).
, 'Philosophy and Religion' (London: Duckworth, 1909).
, 'God and Man' (Oxford U.P., 1930).
W. R. Sorley, 'Moral Values and the Idea of God' (Cambridge
U.P., 1918).
C. C . J . Webb, 'Divine Personality and Human Life' (London:
Allen & Unwin, 1920).
D. M. Baillie, 'Faith in God' (Edinburgh: T. & T. Clark,
1927, and New York: Hillary, 1964) ch. 5.
J o h n Baillie, 'The Interpretation of Religion' pt ii, chs 5-7
(Edinburgh: T. & T. Clark, 1929, and Nashville, Tenn.:
Abingdon, 1965).
A. E. Taylor, 'The Faith of a Moralist' (London: Macmillan,
1930).
W. G. de Burgh, 'From Morality to Religion' (London:
Macdonald & Evans, 1938).
Austin Fairer, 'A Starting-point for the Philosophical
Examination of Theological Belief, 'Faith and Logic', ed.
Basil Mitchell (London: Allen & Unwin, 1957, and New
York: Humanities, 1958).
H. P. Owen, 'The Moral Argument for Christian Theism'
(London: Allen & Unwin, 1965).

Works criticising moral arguments

C. A. Campbell, 'Selfhood and Godhood' (London: Allen &
Unwin, and New York: Humanities, 1957).
140
W. G. Maclagan, 'The Theological Frontier of Ethics'
(London: Allen & Unwin, and New York: Humanities,
1961).
Ronald Hepburn, 'Moral Arguments for the Existence of
God', 'The Encyclopedia of Philosophy', v (New York,
1967).
Margaret Knight, 'Morality — Supernatural or Social?', 'The
Humanist Outlook', ed. A . J . Ayer (London: Pemberton,
1968).

Discussions of various aspects

S. Korner, 'Kant' (London: Penguin, 1955) ch. 7.

H . J . Paton, 'The Modern Predicament' (London: Allen &
Unwin, and New York: Macmillan, 1955) ch. 21.
J o h n R. Silber, 'Kant's Conception of the Highest Good as
Immanent and Transcendent', 'The Philosophical Review'
(1959).
A. C. Ewing, 'The Autonomy of Ethics', 'Prospect for Meta-
physics', ed. Ian Ramsey (London: Allen & Unwin, and
New York: Philosophical Library, 1961).
D. A. Rees, 'Metaphysical Schemes and Moral Principles',
'Prospect for Metaphysics', ed. Ian Ramsey.
Burton F. Porter, 'Deity and Morality' (London: Allen &
Unwin, and New York: Humanities, 1968).

THE ONTOLOGICAL ARGUMENT

Classical treatments

Anselm, 'Proslogion', chs 2-4, 'Reply' by Gaunilo, and

'Response' by Anselm: Latin and English on facing pages,
ed. and trans. M . J . Charlesworth (Oxford U.P., 1965).
Descartes, 'Meditations', v, and 'Objections' i, ii and v with
Descartes' 'Replies', trans. Haldane and Ross, 'The Philo-
sophical Works of Descartes', ii (Cambridge U.P., 1912,
and New York: Dover, 1934).
Spinoza, 'Ethics', pt i, props 7-11, trans. R. H. M. Elwes
(London: Bohn's Philosophical Library, 1883-4).
Leibniz, 'New Essays Concerning Human Understanding', iv,
ch. 10 and app. x, trans. A. G. Langley (London and New
141
 York: Macmillan, 1896); 'Monadology', sects 44-5, trans.
R. Latta (Oxford U.P., 1898).
Kant, Critique of Pure Reason: Transcendental Dialectic', ii,
ch. iii, sect. 4, trans. N. Kemp Smith (London: Macmillan,
and New York: St Martin's Press, 1933).
Hegel, 'Lectures on the Philosophy of Religion', appendix,
trans. E. B. Speirs and J . B. Sanderson (London: Kegan
Paul, 1895).

Historical

C . C . J . Webb, 'Anselm's Ontological Argument for the

Existence of God', 'Proceedings of the Aristotelian
Society' (1896).
R. Miller, 'The Ontological Argument in Anselm and
Descartes', 'The Modern Schoolman' (1954-5 and 1955-6).
'Spicilegium Beccense', Proceedings of the Congres Inter-
national du IXe Centenaire de l'arrivee d'Anselme au Bee
(Paris: Vrin, 1959).
Deiter Henrich, 'Der ontologische Gottesbeweis: sein
Problem und seine Geschichte in der Neuzeit' (Tubingen:
J . C . B. Mohr, 2 n d e d . , 1967).
S. M. Engel, 'Kant's "Refutation" of the Ontological
Argument', 'Philosophy and Phenomenological Research'
(1963).
Alvin Plantinga, 'Kant's Objection to the Ontological
Argument', 'Journal of Philosophy' (1966).
A. C. McGill, 'Recent Discussions of Anselm's Argument',
'The Many-faced Argument: Recent studies on the
Ontological Argument for the Existence of God', ed. J o h n
Hick and A. C. McGill (New York: Macmillan, 1967, and
London: Macmillan, 1968).

Important varying interpretations

Etienne Gilson, 'Sens et nature de Pargument de Saint

Anselm', 'Archives d'histoire doctrinale et litteraire du
moyen age' (1934).
142
Anselm Stolz, 'Anselm's Theology in the Proslogion', trans.
A. C. McGill, 'The Many-faced Argument'.
Karl Barth, 'Anselhi: Fides Quaerens Intellectum', trans. Ian
Robertson (London: S.C.M. Press, and New York: Meridian,
1960).

The Hegelian use of the argument

Edward Caird, 'Anselm's Argument for the Being of God',

'Journal of Theological Studies' (1899).
W. E. Hocking, 'On the Ontological Argument in Royce and
Others', 'Contemporary Idealism in America', ed. C.
Barrett (New York: Macmillan, 1932).
R. G. Collingwood, 'Philosophical Method' (Oxford U.P.,
1933) ch. 6.
Gilbert Ryle, 'Mr Collingwood and the Ontological
Argument', 'Mind' (1935); 'Back to the Ontological
Argument', 'Mind' (1937).
E. E. Harris, 'Mr Ryle and the Ontological Argument', 'Mind'
(1936).
Aime Forest, 'St Anselm's Argument in Reflexive
Philosophy', trans. A. C. McGill, 'The Many-faced
Argument'.

The second form of the argument

Charles Hartshorne, 'Man's Vision of God' (New York:

Harper, 1941).
, 'The Logic of Perfection' (La Salle, 111.: Open Court,
1962).
, 'Anselm's Discovery' (La Salle, 111.: Open Court, 1965).
, 'The Formal Validity and Real Significance of the
Ontological Argument', 'Philosophical Review' (1944).
, 'The Logic of the Ontological Argument', 'Journal of
Philosophy' (1961).
, 'What Did Anselm Discover?', 'Union Seminary
Quarterly Review' (1962).
J . O. Nelson, 'Modal Logic and the Ontological Proof for
God's Existence', 'Review of Metaphysics' (1963).
J . B. Cobb, ' "Perfection Exists": A Critique of Charles
Hartshorne', 'Religion in Life' (1963).
143
Julian Hartt, 'The Logic of Perfection', 'Review of Meta-
physics' (1963).
David Pailin, 'Some Comments on Hartshorne's Presentation
of the Ontological Argument', 'Religious Studies' (1968).
Norman Malcolm, 'Anselm's Ontological Arguments', 'Philo-
sophical Review' (1960), reprinted in 'Knowledge and
Certainty', (Englewood Cliffs, N.J.: Prentice-Hall, 1963).
Discussion by R. E. Allen, R. Abelson, T. Penelhum, A.
Plantinga, P. Henle and G. B. Matthews, 'Philosophical
Review' (1961).
T. P. Brown, 'Professor Malcolm on "Anselm's Ontological
Arguments" ', 'Analysis' (1961).
J o h n Hick, 'A Critique of the "Second Argument" ', 'The
Many-faced Argument'.

Other important modern discussions

J . N. Findlay, 'Can God's Existence be Disproved?', 'Mind'

(1948), reprinted in 'New Essays in Philosophical
Theology', ed. Flew and Maclntyre (London: S.C.M. Press,
and New York: Macmillan, 1955).
Robert J . Richman, 'The Ontological Proof of the Devil',
'Philosophical Studies' (1958).
Nicholas Rescher, 'The Ontological Proof Revisited', 'Austral-
asian Journal of Philosophy' (1959).
William Alston, 'The Ontological Argument Revisited', 'Philo-
sophical Review' (1960).
J . Schaffer, 'Existence, Predication and the Ontological
Argument', 'Mind' (1962), reprinted in 'The Many-faced
Argument'.
Alvin Plantinga, 'God and Other Minds' (Ithca, N.Y.: Cornell
U.P., 1967) chs 1-3.
M . J . A. O'Connor, 'New Aspects of Omnipotence and
Necessity in Anselm', 'Religious Studies' (1968).
Charles Crittenden, 'The Argument from Perfection to
Existence', 'Religious Studies' (1968).

144
THE IDEA OF NECESSARY BEING IN THE ONTOLOGICAL
AND COSMOLOGICAL ARGUMENTS

A. N. Prior, 'Is Necessary Existence Possible?', 'Philosophy

and Phenomenological Research' (1955). R. L. Franklin,
'Necessary Being', 'Australasian Journal of Philosophy'
(1957).
P. JE . Hutchings, 'Necessary Being', 'Australasian Journal of
Philosophy'(1957).
, 'Necessary Being and Some Types of Tautology',
'Philosophy' (1964).
Richard Robinson, 'Necessary Propositions', 'Mind' (1958).
J o h n Hick, 'God as Necessary Being', 'Journal of Philosophy'
(1960).
, 'Necessary Being', 'Scottish Journal of Theology'
(1961).
T. Penelhum, 'Divine Necessity', 'Mind' (1960).
W. E. Abraham, 'Is the Concept of Necessary Existence Self
Contradictory?', 'Inquiry' (1962).
J . F. Ross, 'God and Logical Necessity', 'Philosophical
Quarterly'(1961).
Guy Jalbert, 'Necessite et Contingence chez saint Thomas
d'Aquin et chez ses Predecesseurs' (Ottawa: Editions de
l'Universite d'Ottawa, 1961).
Anthony Kenny, 'Necessary Being', 'Sophia' (1962).
, 'God and Necessity', 'British Analytical Philosophy', ed.
Bernard Williams and Alan Montefiore (London:
Routledge & Kegan Paul, and New York: Humanities
Press, 1966).
R. Puccetti, 'The Concept of God', 'Philosophical Quarterly'
(1964).
Patterson Brown, 'St Thomas' Doctrine of Necessary Being',
'Philosophical Review' (1964).
J. N. Findlay, 'Some Reflections on Necessary Existence',
'Process and Divinity', ed. William Reese and Eugene
Freeman (La Salle, 111.: Open Court, 1964).
Alvin Plantinga, 'Necessary Being', 'Faith and Philosophy',
ed. Plantinga (Grand Rapids, Michigan: Eerdman, 1964).

145
 RATIONAL THEISTIC BELIEF WITHOUT PROOFS

J o h n Baillie, 'Our Knowledge of God' (London: Oxford U.P.,

and New York: Scribner's, 1939).
, 'The Sense of the Presence of God' (London: Oxford
U.P., and New York: Scribner's, 1962).
H. Richard Niebuhr, 'The Meaning of Revelation' (New
York: Macmillan, 1941).
Erich Frank, 'Philosophical Understanding and Religious
T r u t h ' (London and New York: O.U.P., 1945).
J o h n Hick, 'Faith and Knowledge' (1957), 2nd ed. (Ithaca,
N.Y.: Cornell U.P., 1966, and London: Macmillan, 1967).
H. D. Lewis, 'Our Experience of God' (London: Allen &
Unwin, and New York: Macmillan, 1959).
Austin Farrer, 'Faith and Speculation' (London: Black, and
New York: New York U.P., 1967).
Alvin Plantinga, 'God and Other Minds' (Ithaca, N.Y.:
Cornell U.P., 1967).
Diogenes Allen, 'The Reasonableness of Faith' (Cleveland,
Ohio: Corpus Books, 1968).

146
Index

Aitken, H. D., 103 Existence, vii, viii

Anselm St, ix, 69 f., 84 f., 87 f., 95-6,
97, 128, 129, 132 Faith, 102 t , 104 t , 108,114
a posteriori, viii, ix Farmer, H. H., 27
a priori, viii, ix, 68 Findlay.J. N., 131
Aquinas, St Thomas, 1, 37, 38 f., 79, First cause, 103
85 First cause argument, 42 f.
Aristotle, vii, 132 Flew, Antony, 46
aseity, 86-7 Fox, Sidney, 124
Augustine, St, 31, 69, 73, 128 Freedom, 104
Avicenna, 131
Garrigou-Lagrange, R., 126
Baier, K. M., 131 Gassendi, Pierre, 80
Baillie, John, 134 Gaunilo, 75, 76 f.
Balbus, Lucilius, 1 Geach,P., 131
Barth, Karl, 90-1, 129 God: definition, vii; possibility of, xiii,
Baumer, W. H., 127 99-100
Beauty, 20 Goulder, M., 127
Bible, 102 Grant, C. K., 130
Bonaventura, 79 Guye, Charles-Eugene, 15
Boyle Lectures, 1, 18, 121
Bridgewater Treatises, 1, 18
Hartshorne, Charles, 70, 9 1 , 93 f.
Broad, C D . , 129
Henze, D. F., 134
Brown, Arthur I., 122
Heschel, A., 134
Buber, Martin, 134
Hinduism, 117
Buddhism, 117, 118
Hobbes, Thomas, 30
BusweU, J. O., 134
Humanism, 59 f.
Hume, David, 2, 4, 7 f., 20, 2 7 , 4 7 , 8 5 ,
Caterus, 92 103, 110, 122
Charlesworth, M.J., 72-3 Huxley, Julian, 11
Christianity, vii, 119
Cicero, 1, 7 Improbability, 14 f.
Contingency, argument from, 43 f. Islam, vu, 117, 118, 119
Copleston, F. C , xi, 43, 51
Cosmological argument, 37 f., 68 Jalbert, Guy, 131
Jeffner, Anders, 122
Darwin, Charles, 2, 10-11, 18 Jesus of Nazareth, 112, 115, 119
Derham, William, 3, 5 f. Johnston, T., 127
Descartes, Rene, ix, 70, 79 I, 92, 110 Judaism, vii, 101, 118
Design argument, 1 f., 68
Designer, 103
Du Nouy, Pierre Leconte, 2, 14 f. Kant, Immanuel, 14, 33-4, 53 f., 64,
Duns Scotus, 79, 97 f. 68, 70, 81 t , 85, 92, 127
Kenny, Anthony, 40-1, 132
Eadmer, 128 Kierkegaard, S., 101
Eliot, T. S., 134 Knight, Margaret, 62-3
Evil, 30-1
Evolution, 11, 19, 25-6 Leibniz, G. W., 37, 5 3 , 129-130
147
McGill, A. C , 76, 90 Probability, xii, 13-14, 27 t, 101;
Maclntyre, Alasdair, 105 alogical, 29 f.
Malcolm, Norman, 70, 91 f. Proof, ix f.
Marquand, J., 124 Protestantism, 101
Mascall, E. L., 43
Maslow, Abraham, 117 Rashdall, Hastings, 57 f., 127
Matson, Wallace I., 16 f. Rationality, 108 f.
Matthews, W. R., 27, 134 Ray, John, 3
Moore, G. E., 131 Religious experience, 109 f., H I f.,
Moral arguments, 53 f. 117 f.
Morality, 20-1, 31, 53 f. Roman Catholicism, 101
Motion, argument from, 38 f. Ross, James F., 97 f.
Mystery, 34 f. Russell, Bertrand, viii, xi, 5 1, 58, 60 f.,
82-3,85, 131
Narveson, Jan, 125
Sartre, J.-P., 51
Natural belief, 47, HO
Sertillanges, A. D., 126
Necessary being, 46, 84 f., 103, 132-3
Smart, J. J. C , 85-6, 127
Neoplatonism, 7 1 , 73
Smith, John E., 121
Newman, J. H., 58-59, 106, 107
Smith, J. M., 123, 125
O'Connor, M. J., 74 f. Smith, N. Kemp, 8
Ontological argument, ix, 68 f., 104; Solipsism, 110, 112-113
'first form', 70 t ; 'second form', Stoics, vii
84 f. Stopes-Roe, H. V., 127
Oparin, A. O., 124 Sufficient reason, principle of, 37,43

Taylor, A. E., 2
Paley, William, 2, 3 f., 14, 18, 122, 123 Taylor, Richard, 2, 21 f.
Paley's Watch, 3 f. Teleological argument, see Design
Pascal, B., 104 argument
Peirce, C. S., 134 Tennant, F. R., 2, 14,18 f., 27, 28-9
Plantinga, Alvin, 13-14 Tillich, Paul, viii
Plato, vu, 1,73
Pope Pius XII, 134 Whitehead, A. N., 131
Porter, Burton F., 122 Wilson, J. Cook, 102-3
Prime mover, 38 f. Wolff, C , 53. 128

148