The Library
SCHOOL OF THEOLOGY
AT CLAREMONT
WEST FOOTHILL AT COLLEGE AVENUE
CLAREMONT, CALIFORNIA
BULK
The urgent need to save paper has had
a marked effect upon size.
DO NOT BE MISLED
BEFORE THE WAR NOW
this book would have the same book
looked like this > looks like this
Vii
| WAR
the number of words is EXACTLY the same
45112
HE WHO IS
BOOK
PRODUCTION
WAR ECONOMY
STANDARD
THIS BOOK IS PRODUCED IN
COMPLETE CONFORMITY WITH THE
AUTHORIZED ECONOMY STANDARDS
se = : : : te ne 4S ey
eee NE A i .
sh
=a - ee
Se : ; ee. = iS iNe Goes
3 = bile RUE
; 7 Hs
Ss : ti
Se cae}
ie
ve at
. - = - ;
¥ sy t
ie = : oe
HE WHO IS
A STUDY IN TRADITIONAL THEISM
BL
200
M3a By
1443 E. L. MASCALL, B.D.
é
Sub-Warden Ms Scholae Cancellarii, Lincoln
Priest of the Oratory of the Good Shepherd
Les purs philosophes disent sans cesse: recommengons; le
philosophe chrétien peut seul dire: continuons.
A. D, SERTILLANGES
Ita ergo philosophandum est in fide.
Joun oF St. THomas
LONGMANS, GREEN AND CO.
LONDON << NEW YORK < TORONTO
-LONGMANS, GREEN AND CO. LTD.
OF PATERNOSTER ROW
43 ALBERT DRIVE, LONDON, S.W.19
NICOL ROAD, BOMBAY
17 CHITTARANJAN AVENUE, CALCUTTA
364 MOUNT ROAD, MADRAS
LONGMANS, GREEN AND COQ.
55 FIFTH AVENUE, NEW YORK
LONGMANS, GREEN AND CO.
215 VICTORIA STREET, TORONTO
Theol ogy ie;brary
SCHOOL OF THEOLOGY
AT GLAREMONT
Cee ee
First published 1943
Reprinted 1945
CODE NUMBER: 45112
PRINTED IN GREAT BRITAIN
BY WESTERN PRINTING SERVICES LTD.) BRISTOL
el i
Paes
timens Dominum ipsa lau
= ae des Fit ees Es s
CONTENTS
CHAPTER PAGE
PREFACE . : ‘ ; 5 ; ; 5
L. INTRODUCTION . : ‘ P : ‘ <
II. Tue MEANING oF ‘‘GopD”’ : : : ;
III. EXPERIENCE AND REVELATION . : j , a4
Iv. Tue TRADITIONAL APPROACH
(1) SELF-EVIDENCE OR PROOF? ; é : 30
Ns Tue TRADITIONAL APPROACH
(2) Tue Five Ways ; : : : 40
VI. THE TRADITIONAL APPROACH
(3) Irs StcnirIcANCE AND VALIDITY d : 57
VII. INTELLECT AND INTUITION ‘ P : : 83
Additional Note A: CERTAINTY AND CERTITUDE . 93
VIII. Gop AND THE Wor tp: Analogia Entis : ; 95
Additional Note B: Gop AND THE WORLD IN THE
THEISM OF JAMES WARD 11g
TX. Tue Divine ATTRIBUTES . 5 : : : 116
‘TRANSCENDENCE AND IMMANENCE : ‘ ; 126
XI. THE CosmMoLocy oF WHITEHEAD : : ‘ 150.
XU. Tue Cosmic TELEOLOGY OF TENNANT 2 : 161
Additional Note C: Tot Morar ARGUMENT AND THE
PROBLEM OF EvIL ; : : : : 179
Additional Note D: NATURAL THEOLOGY AND THE
DoctTRINE OF THE TRINITY . 184
XIII. CONCLUSION : : : : : : Z IQI
BIBLIOGRAPHY . E ; i 5 ; E 200
INDEX OF PROPER NAMES . : : Z : 208
vi
PREFACE
HE primary negate of
this book is purely academic. It is’
an attempt to restate and reassess the traditional Christian
approach to the fundamental question of_natural._theology,...
namely the question of the existence of God and of his relation
to the world, and then, in the light of the conclusions arrived at,
to examine some important allied problems and to review some
of the more notable discussions of natural theology of recent
years.
But, from the very nature of the case, a secondary purpose is
inevitably involved of a very practical kind. For it is altogether
impossible, unless one has achieved a quite inhuman degree of
intellectual detachment, to ignore the tremendous repercussions
which one’s intellectual convictions about the existence of God
and his relation to his creatures are bound to have upon the very
deepest levels of one’s personal life. And, if anything is clear about
the predicament in which man finds himself at the present day, it
is that, both in its individual and in its corporate aspects, human
action is profoundly influenced by the assumptions which men
make about the nature of ultimate reality. In the realm of
individual piety this is too obvious to need remark, but in the
corporate realm it is hardly less evident. I would advise anyone
who has doubts about it to study such a penetrating discussion of
the influence of ideas upon society as that which is given by
Dr. Demant in his book The Religious Prospect, or even merely to
refer to the Communist Manifesto and Mem Kampf and then
consider the state of Europe to-day. And all the time that I have
been writing this book the thoroughly practical implications of its
subject have never been far from my mind. I have, however,
tried to resist temptations to digress from the main theme, and\
I offer it simply as an essay in the basic problem of natural |
theology.
I have described this book as “‘A Study in Traditional Theism,”’
and in view of this and of the frequency with which the word
“‘traditional”’ occurs in it, the use which is made of the term may
seem to demand both definition and justification. I mean by
traditional theism the doctrine about God and the universe which,
deriving from the impact made upon the Greco-Roman world by
vii
Vili PREFACE
the Christian Church and passing by way of Augustine, received,
as regards its main features, a coherent formulation in the thir-
teenth century in the works of St. Thomas Aquinas. I do not suggest
that the last word on everything has been said by the Angel of the
Schools, nor in this use of the word ‘‘traditional’’ do I wish to
deny that other trains of thought within Christendom might well
lay claim to the word in their different ways. M. Gilson has
demonstrated the existence in the brepee ES itself of a more
narrowly Platonic and Augustinian line, which was crystallized
in the writings of St. Thomas’s great Franciscan contemporary,
St. Bonaventure, and the mere mention of the names of Duns
|
/ Scotus and William of Ockham is sufficient to disprove any
notion that a medieval western Christian philosopher was ipso
facto a Thomist.. Even within the post-Reformation Roman
| Church the Cartesian theology of Malebranche and the French
Oratorians might not illicitly claim to constitute.a tradition.
Nor, in the strict sense of the term are the great Spanish Jesuit
theologians to be described as Thomists. Furthermore, it must be
remembered that the Western Church does not form the whole of
historic Christendom, and that in the east of Europe a divergence
appears at least as early as the time immediately after St. John
of Damascus, which knows not Aquinas and is gravely suspicious
of Aristotle and Augustine. Nevertheless, it may be maintained
that, for Western Christianity, a position which, at least in its
broad outlines, was that of the Angelic Doctor achieved a central
position which was not seriously challenged until the day when
Descartes shut himself up with his stove and indulged in his orgy
of universal doubt, and which has shown in recent years a remark-
able power of revival and of competence to come to grips with the
pressing problems of contemporary life and thought. The term
“scholastic”? is too wide for it, as reference to any book on the
history of medieval philosophy will show; on the other hand, the
| term ‘“Thomist”’ might suggest too restrictive and reactionary a
resort to the ipsissima verba of the Angelic Doctor; and, while I
should be ready to use the latter term if it was interpreted in a
broad sense, I have on the whole preferred the word “‘traditional.”’
“I hope, in view of this explanation, that I shall not appear to be
——__
begging a question in doing so. In any case, it will be seen that
I do not consider Thomas locutus, causa finita as the last judgment to
be passed on any theological problem ; though my approach might
be summed up in the words, Thomas locutus, causa incepta, ——~
PREFACE ix
~ By << adifional theism,” then, I mean a doctrine about God
a and the world whose main assertions are :
(1) That the human mind can, from the consideration of finite
beings, arrive, without appeal to “‘religious experience”’ or “‘reve-
lation,’ at a sure knowledge of the existence of a God whose
primary character is that of self-existent Being. ——
(2) That the finite world derives its existence and its persistence
from a free act of will upon the part.of God, to whom it is alto-
gether unnecessary and who would. be in every respect complete
without it, but whose concern with it is none the less a manifesta-
tion of the deepest condescension and love.
I am aware that from the time of Kant onwards the arguments
which claimed to establish this position have been frequently held
to be fallacious, and that, for example, most Gifford Lecturers
would deny it. I would, however, reply that, in his long article
on “Theism’’ in the Encyclopaedia of Religion and Ethics, a philo-
sopher of the eminence of Professor A. E. Taylor has criticized
very devastatingly the Kantian objections, and that, in the radi-
cally altered intellectual atmosphere of the present day, an
increasing number of philosophers of the first rank are to be found
who are prepared to defend in all its main points the traditional
approach. The names of Gilson and Maritain are sufficient in
this connection. ©
I am also aware that it is very generally believed that any
return to the general philosophical attitude of the Middle Ages
involves a complete rejection of the discoveries and theories of
modern science and a retreat to a pre-Copernican view of the
universe. There is, however, no necessary connection between
medieval science and medieval philosophy. M. Maritain has
admitted that ‘‘the endeavour of the ancients resulted, so far as
the science of the phenomena of nature was concerned, in a
resounding failure, at least with regard to matter and movement,’’!
and he has remarked that the prodigious invention of physico-
mathematical science ‘‘has changed the face of the world, and
given rise . . . to that terrible misunderstanding which, for three
hundred years, has created a quarrel between modern science and
the philosophia perennis.”’? But the book from which these words
are quoted should be sufficient to show that the task of dis-
1 Degrees of Knowledge, E.T., p.5
2 Ibid., p. 52. Cf. Réflections sur i isebatigathe, ch. vi.
x pea PREFACE
entangling the philosophy from the science of St. Thomas is far
from impossible and that the philosophia perennis is quite capable of
assimilating, and furthermore of putting in its right place, the
great body of scientific discovery of the post-Renaissance period.
Nor were the medievals quite as backward as we sometimes are
inclined to think. Albertus Magnus believed in the sphericity of
the earth, Nicholas of Autrecourt believed in the finite velocity of
light, 2 Buridan anticipated the notion of momentum$ and William
of Ockham the first of Newton’s Laws of Motion.* And, whatever
\may be the popular assumption, the Middle Ages were anything but
rrationalist ;their defect was rather to place too much reliance
reason. Nobody could describe Professor Whitehead as a
bigoted medievalist ; yet it is he who has told us that the faith of
the Middle Ages was based upon reason, while it is the later
__period that has based its reason on faith.®
And here perhaps a word of apology may be needed for the
extent to which, in the course of my argument, I have referred to
the works of modern French-speaking Roman-Catholic: philo-
sophers. I can only say that, in my opinion, the intellectual
revival in Roman-Catholic circles in France and Belgium is one
of the most significant cultural phenomena of the present century,
and that Anglicans have at least as much to learn from them as
from the more widely publicized Protestant thinkers of Germany. In
thephilosophical spheresuch writers as Gardeil, Garrigou-Lagrange,
Sertillanges, Maritain and Gilson have a great deal to say to us;
and we may add, in the theological sphere, the names of Mersch,
de Lubac, Congar and Charlier. And if it be objected that their
discussions are out of harmony with the attitude of the main body
of Anglican theology in recent years, this may be admitted so long
as we add that Anglican theology in recent years has itself departed
from its own tradition. If we go back behind the liberal movement
to the great Anglican divines of the seventeenth century we find
that, whether they agree or disagree with their Roman-Catholic
contemporaries, at least they speak the same language; indeed
without some knowledge of the thought and terminology of
scholasticism it is for the most part impossible to follow them.
For an example of a precisely worded scholastic argument of great
subtlety one need only read the discussion of the compatibility
of divine foreknowledge with human freedom which is given by
1 De Coelo et Mundo, II, iv, 3. % See Gilson, La Philosophie au Moyen Age, p. 276.
3 Ibid., p. 286. 4 Thid., p. 285. 5 Science and the Modern World, ch. iv.
pe PREFACE xi
_ the seventeenth-century Dean of Peterborough, Thomas Jackson;
he says precisely what Aquinas says, and says it quite as well. All
this would, of course, be mere archeology if the traditional
approach had been conclusively proved to be fallacious ;but I do
not think that will be found to be the case.
I have tried, therefore, in this book both to restate and to
reinterpret the traditional approach to theism. How far my
restatement agrees with St. Thomas’s own thought only an expert
Thomist scholar presumably could judge. I do not think that in
any case it matters very much; St. Thomas certainly will not mind.
And how far my restatement is itself valid it is not for me to say.
I will, in concluding this preface, only anticipate one criticism:
namely, that I have given very inadequate consideration to two
very important questions: (a) the divine attributes, (b) the problem
of analogy.
As regards the former of these I do not think that I have any-
thing to add to the discussion given by such a theologian as, for
example, Fr. Garrigou-Lagrange. As regards the latter, it is, I
should say, one of the most difficult and at the same time most
important problems for contemporary philosophy. The whole
question of the relation between words and things, which is
exercising the logical positivists so much, is bound up with it, and
if they are right most of our statements not only about God but
about other beings as well are meaningless. Fortunately for the
theologian, the logical positivists’ position seems to be crumbling
from within; at least we may claim the right to go on talking
philosophy until they have settled with the rebels in their own
ranks. But I should heartily agree as regards the need for a more
thorough investigation of analogy than it has yet received; I
cannot feel that the great work of the Abbé Penido, with all its
excellences, has said the last word on the matter. Partly the
problem is in the realm of language, partly it is in the realm of
ontology ; and analogy itselfisan analogical notion. The doctrine ~
of analogical predication is in the former realm; the doctrine of
analogia entis, as formulated, for example, by Przywara, is in the
latter; and the relation between the two presumably depends in
some way upon the truth that the intellect is the faculty whose
object is being. But non omnia possumus omnes, and I am at least
conscious_of the problem even if I do not know how to solve it.
| To-sum up, then, this book is put forward as a small contribu-
a
|tion to.the‘reconstruction of Anglican theology. One thing at least
}
\
\
xii PREFACE
is clear in the theological turmoil of the present time: that, as a
living force, theological liberalism has had its day. Professor
Whitehead asserted only a few years ago that “‘the defect of the
liberal theology of the last two hundred years is that it has confined
itself to the suggestion of minor, vapid reasons why people should
continue to go to church in the traditional fashion”’ ;1 and, if this
be felt to be too sweeping a condemnation, we may at least affirm
that the theological attitude which has, to quote the introduction
to a well-known series of modern books,? prided itself on its “‘claim
to be in harmony with modern thought,” will have little contribu-
tion to make to the rebuilding of society in a world which is daily
manifesting itself as being less and less in harmony with the
Christian religion. The immediate danger in the world of
religious thought is that, in reaction from liberalism, reason should
find itself asphyxiated by a wave of sheer irrationalism. It was the
glory of medieval theology that it assimilated all that was best in
contemporary movements of thought without being assimilated
by them ; that was no doubt the ambition of the liberal theology of
the last two centuries, but it was one which it failed to achieve.
Nevertheless, the attitude which lies behind the present book is
far from being one of hostility to all that liberal theology stood for.
The liberal aim, namely the commendation of the Christian Faith
to the contemporary world, was highly laudable, but the cost
which liberalism was prepared to pay in its unsuccessful attempt
to actualize it was too high. Ifthe claim of reason to a rightful, if
limited, recognition is to be justified, theology must return to its
traditional position and apply itself to its traditional task, with a
clear understanding of the demarcation between the spheres of
faith and reason and of the relation between them. It is because
I have tried to examine the problem of the existence of God and
of his relation to the world with this conviction in mind that I have
called this book “A Study in Traditional Theism.”
I have given quotations as exactly as possible, but have every-
where spelt pronouns referring to the Deity with a small initial.
To avoid unnecessary repetition I have generally omitted honorific
prefixes from the names of living scholars after the first mention
1 Adventures of Ideas, p. 218. 2 The Library of Constructive Theology.
a PREFACE xiii
in any context; it will, I hope, be understood that this is done for
the sake of brevity and not through lack of respect.
My gratitude is due to Professor J. S. Boys Smith for his kind-
ness in reading this book in manuscript and to Mr. Basil S. Moss
for correcting the proofs. How much the book owes to the intel-
lectual stimulation provided by my fellow members of the staff
and by the students of the Schole Cancellarii will be evident to
any of them who may chance to read it. I must also thank my
friends at All Saints’ Church, Margaret Street, London, for their
hospitality and forbearance during the later stages. Acknowledg-
ments are due to the Delegates to the Clarendon Press for permis-
sion to incorporate in Chapter VI, in expanded form, the substance
of an article which appeared in the Journal of Theological Studies
for July—October 1942, and to the following for allowing me to
quote passages from other works: The Syndics of the University
Press, Cambridge; The Oxford University Press; The Society
for Promoting Christian Knowledge; Messrs. Sheed and Ward;
Messrs. Hodder and Stoughton; Messrs. Frederick Muller; The
Centenary Press; Messrs. James Nisbet and Co.; and Messrs.
Burns, Oates and Washbourne.
CHAPTER I
INTRODUCTION
OGICALLY and essentially, the doctrine of God is the funda-
mental doctrine of the Christian Religion, for, according to
its teaching, everything other than God depends upon him and
exists for his glory. ‘‘The Catholic Faith is this,” declares the
Athanasian Creed, “that we worship one God in Trinity and
Trinity in Unity.”’ This does not mean, however, that the truth
of the triune being of God is the first thing of which most of us
become conscious in our life as Christians. One of the drawbacks
of being a mere creature is that you see everything the wrong way
round ; you look at things from man’s standpoint and not from
|
God’s. The order in which things ultimately exist, the ordo essendi,
is usually the precise opposite of the order in which we come to
know them, the ordo cognoscendi;! and this is specially true of that
which is of all beings the most fundamental, namely God himself.
|
If we were brought up in a Christian home, our first réligious
contact was with the practices and objects of Christian devotion:
the crucifix or picture above our bed, the prayers which were first
said for us and which later on we learnt to say for ourselves, the
structure and furniture of our parish church. Then we learnt
about the events of our Lord’s life and about his teaching, and
only—if ever—when we began to study the Catechism were we
given any systematic instruction about the nature of God. That
is to say, we passed from the practice of Christian devotion to the
study of the person of Christ, from that to some understanding of
| God. The logical order is the reverse of this: God comes first ;then
Christ, who is God incarnate in human flesh; and last of all, the
\faith and devotion of the Church which Christ founded. And this
is, in fact, the order adopted by both the Apostles’ and the Nicene
Creed, which begin with God the Father, then summarize the
facts of the Incarnation and of Redemption, and only at the end
mention the inspired Scriptures, the Church and Baptism. As
Karl Barth says, the orderis not genetical but essential. In saying,
AG Professor Whitehead remarks that the “identification of priority in logic with
priority in practice has vitiated thought and procedure from the first discovery of
mathematics and logic by the Greeks.”’ (Process and Reality, p. 75.)
2 Credo, p. 40.
F
Qs : HE WHO Is
then, that the doctrine of God is the fundamental doctrine of the
Christian Religion, we are not suggesting that it is what comes
chronologically first in the normal education of the Christian soul ;
it must, however, be put in the first place if we are tryingtomake
a systematic study of the Christian Faith and to see how its various
parts fit together in a coherent and articulated whole.
The doctrine of God is thus of the most immense and basic
importance. It has also, at any rate in this country, been most
shamefully neglected. It has only too often been assumed that,
however much English people may differ about the Church and
the Sacraments and even about the person of Christ, they all (with
the exception of a few avowed atheists) inherit, as by a kind of
birthright, at least the essential elements of the Christian doctrine
of God. The consequence is that we find ourselves trying to
persuade people that Christ is God when their knowledge of God
is practically non-existent, and that Christ is present in the Blessed
Sacrament when they have only the haziest knowledge of what
the Church believes about Christ.
There may have been some excuse for making this assumption
in the last century, when the secularization of life and thought was
less advanced than it is now; there is very little excuse for it to-day.
And even in the last century it was a dangerous assumption to
make, for it cannot be taken for granted that any doctrine of God
claiming the name of Christian will form a satisfactory basis on
which to erect the superstructure of the Church’s sacramental life
and practice. That is to say, there is not just one uniform doctrine
of God held without variation by all who profess and call them-
selves Christians, upon which Catholic Christology and sacra-
mental doctrine can be superimposed.in successive layers. On the
contrary, while it is a matter for gratitude that much truth is held
jin common by Christians of all denominations, there is a specific
Catholic doctrine of God, which differs in important ways from
‘the doctrine taught by liberal Christianity and even more from
the vague sentiment of a shadowy and beneficent power behind
the visible and tangible universe which is all that the ordinary
Englishman has in mind when he admits, with considerable
embarrassment, that of course he does really believe in God. And
since, as we have seen, the doctrine of God is the basis upon which
all other Christian doctrine rests, any error that has been allowed
to creep into a man’s belief about God will distort his understand-
ing of every other Christian truth. If his idea of God is wrong, his
- INTRODUCTION 3
idea of Christ will be wrong, since Christ is God incarnate ;and
his ideas of the Church and the Sacraments will be wrong, since
the Church is Christ’s Body and the Sacraments are the instru-
ments of his action upon the human soul. It is a common experi-
ence of those engaged in Christian apologetics that the difficulties
that trouble people concerning Christian belief and practice turn
out, on careful investigation, in the great majority of cases to rest
upon a misunderstanding of some element in the Christian doc-
trine of God. If English people are ever to be won back to the
Faith of historic Christendom, it will very largely be as the result
of an apologetic based upon a clear and reasoned exposition of the
fundamental Fact of the triune and creative God. For, as
Dr. W. G. Peck has said, “You cannot have a valid human mean-
ing in a universe which is meaningless; and you cannot maintain
the Christian respect for persons when you have dismissed the
Christian doctrine of God.’’}
It was said a few lines back that the ordinary Englishman has
some sort of vague sentiment which he would on occasions describe
as his belief in God, and indeed one of the most universal charac-
teristics of human beings is the possession of such a belief. It is a
fairly safe generalization that all human beings, unless they are
blinded by prejudice or sophistication, have a conviction, though
often a very obscure conviction, of the existence of something
which as a matter of fact is God, even though they may not them-
selves know what it is and even though they may express this
conviction in a self-contradictory or ridiculous way. This may be
illustrated by one or two examples which are none the less typical
for probably being apocryphal. The farmer who complained that
that there dratted providence had given him a bad harvest and
no mistake, but thanked heaven that there was one above who
would see justice done in the end ; the Member of Parliament who,
having supported the admission of Jews to the House, opposed the
admission of atheists on the ground that we all of us believe in some
sort of a something somewhere; and the Hyde Park orator who
ended his recital of the crimes of organized religion with the words,
“Well, if that’s Christianity, thank God I’m an atheist’’—all these
were testifying to their belief in the existence of a being to whom
a Christian would have to attribute the name of God even if they
were not ready to do so themselves. And this belief is as common
in so-called polytheistic primitive tribes as in civilized nations.
1 The Salvation of Modern Man, p. 83.
B
4 HE WHO IS
After his very full discussion of this question in his book, Religions
-of Mankind, Otto Karrer writes: “Our conclusion must be the
following: There is a ‘consensus generis humani,’ an agreement of
mankind so far as our present knowledge extends, in the belief that
there exists an absolute and supreme Being above ourselves which
has ordered the universe and human life in particular. Mankind
as a whole is conscious of being bound in return... . by a power
which is above all and mightier than itself.’ And again, ‘in this
widest sense it is true to say that history knows of no people godless
and devoid of religion, though here and there particular groups,
schools of thought or governments may combat religion.”? And
Fr. W. Schmidt has argued that the primitive religion is the
“religion of the High God” and that nature-myths, fetichism,
ghost-worship, animism, totemism and magic are absent, or almost
absent, among the earliest peoples.2, Among both savages and the
civilized this one God tends to be kept very much in the back-
ground and the conception of his nature is often vague to the
point of nebulosity; nevertheless, the belief is there, and, while
Christian theologians might well hesitate to make it the basis of a
formal argument for the existence of God, the universality of its
occurrence is impressive.+*
Historically speaking, the Christian doctrine of God arises from
two main sourcef | Jewish religion, with its culmination in Jesus
of Nazareth, and the philosophy of the Greco-Roman world. For
the Christian Church emerged from the self-contained milieu of
Judaism, with its rigid, and indeed almost fanatical monotheism,
into the syncretistic culture of the Roman Empire with its Gods
many and Lords many. And the history of the Christian con-
troversies of the first five centuries is very largely the story of the
adaptation of the categories of Greek thought to the Christian
revelation and of the successful resistance on the part of the
Church to all attempts to distort her Gospel in the process.
The contrast between the Jewish and Greek approaches to the
1 Op. cit., E.T., p. 80. 2 Ibid., p. 81.
3 The Origin and Growth of Religion, E.T., passim. A.short summary of W. Schmidt’s
views will be found in his two essays on “‘‘The Religion of Earliest Man” and ‘““The
Religion of Later Primitive Peoples” in Studies in Comparative Religion, Vol. I. He main-
tains that the essential monotheism of the most primitive peoples degenerates into an
Earth-mother and moon religion in the matriarchal-agrarian culture, while remaining
substantially itself in the large-family cattle-raising culture. In the later cultures,
formed by the fusion of these three primary cultures, the religious conceptions and
observances become correspondingly mixed. Cf. Mr. G. K. Chesterton’s amusing
stories of the experiences of missionaries given in The Everlasting Man, p. tor f.
4 Cf. St. ‘Thomas Aquinas, S. Theol., I, ii, 1 ad 1; S.c.G., IIL, xxxviii.
INTRODUCTION 5
problenrof existence is very largely the contrast between effective
and formal causality. We might perhaps say that, while the
Greeks were interested to know what sort of thing the world was,
the Jews wanted to know what was the power behindit. But even
to say “the Jews wanted to know” is to attribute to them an
interest in philosophy which was, except where they became
hellenized, foreign to their character. For the Jews were quite
certain that they did know, since God had revealed himself to
them. In his mighty redemptive acts, when he had delivered them,
first from slavery in Egypt and then, centuries later, from captivity
in Babylon, and in his disclosure of himself in the teaching of the
great line of prophets—of Moses, Elijah, Amos, Hosea, Isaiah,
Jeremiah, Ezekiel and the rest—he was manifested, not as a mere
principle of philosophic explanation, nor as just the basis of man’s
moral aspirations, but as the Living God, who had made heaven
and earth out of nothing, who had chosen the Jews out of all the
nations of the earth to be his own peculiar people, who was
righteous and faithful and who demanded righteousness and faith-
fulness from his servants, who scourged them for their sins and
pardoned them on their repentance, who rode upon the thunder-
clouds and yet dwelt in the hearts of men, and whose throne upon
earth was the Holy of Holies in the Temple of Jerusalem. And to
those Jews who were Christians the climax of the revelation and
the fulfilment of the prophecies had come in the person of Jesus
of Nazareth, who by his death and resurrection had overcome
death and wrought the supreme deliverance, and had sent his
Church into the world filled with the Spirit—the very breath—of
the Living God.
The God_of the Jews was living, personal and creative; he was
supreme Being and transcendent Act. And his claims were
inescapable and paramount. “Hear, O Israel, the Lord our God is
one Lord, and thou shalt love the Lord thy God with all thine
heart and with all thy soul and with all thy might.” “Before me
there was no God, neither shall there be any after me. I, even I,
am the Lord, and beside me there is no saviour.... I am the
first and the last, and beside me there is no God.” The revelation
of God’s true narne and nature had been made to Moses at the
burning bush, in words which became the basic text of the
|Christian doctrine of God—of what M. Gilson has called “the
|metaphysic of Exodus.” His name is Jehovah—“‘I am that I am”’
& “TI will be that I will be.’ ‘Thus shalt thou say to the chil-
6 , HE WHOIS _
dren of Israel, I AM hath sent me unto you.... I am Jehovah,
and I appeared unto Abraham, unto Isaac and unto Jacob as
God Almighty, but by my name I AM I was not known unto
them.”
In contrast to the Jews, Hie Creckeave eatin mein
of any
historical sense, and the Jewish conception of a Great Day at the
end of time “when the Living God would finally vindicate his
supremacy over his creation and destroy the powers of evil was
entirely alien to them. Their interest was not in what was happen-
ing in the course of history but in what was going on around them.
How was the multiplicity of the world to be explained? What was
its principle of unity? Who could solve the problem of the many
and the one? What was the world’s formal cause, its unifying
principle? The answers given to these questions were many and
diverse, and there were almost as many philosophies as philo-
sophers, but perhaps we shall not be giving too inaccurate a picture
of the attitude of Greek philosophy if we say that, when it believed
in God at all, it tended to conceive of him as the principle of form
and beauty rather than as a creative living being.1_ Whether he
was personal, and if so in what sense, was at least doubtful; he
could hardly be supposed to be interested in the world, and he
influenced it probably through the attraction which his perfection
had for it rather than through any deliberate effect exercised by
him upon it. God and the prime matter of which the world was
composed were very probably co-eternal, and his effect upon it
was to impose form and order upon this pre-existent matter; or
perhaps it was some kind of emanation from him. Anyhow, there
was no question of creation, in the sense of making the world out
of nothing—that would be a most unphilosophical idea! To the
Greeks God was the great Thought behind the universe; not as
to the Jews, the great Act. For nothing ever really changed, or
perhaps nothing ever really stood still; in either case there was a
complete absence of the Jewish insistence on the contrast between
the changing, dependent and yet real world and the changeless,
sovereign and creative God.
There had, of course, been some contact between Greek and
Jewish thought, especially in the neighbourhood of the city of
Alexandria, with its large colony of Jews and its extraordinarily
1 The preoccupation of the Greeks with the nature of the universe rather than with
the cause of its existence becomes very clear from the reading of such.an account as
that given by M. Maritain in his Introduction to Philosophy, Part I, ch. ii-iv.
- al
INTRODUCTION ,,
cosmopolitan setting. Indeed, this provided Christian thought
with at least one of the ideas that it needed for the proper formula-
tion of its doctrine.! But in spite of this the clash of the Jewish
religion of Christianity with the world of Greek thought was like
the impact of flint upon steel—and fire was kindled by it.
Notwithstanding the work of the great Christian apologists and
Fathers, the final synthesis did not appear until the thirteenth
century, and then in the west of Europe. The immediate cause
was the production at various times in the previous two centuries
of some of the works off Aristotle in inaccurate double or triple
translations from the Arabic, often through the medium of
Hebrew.” The first reaction to this resurrected pagan on the part
of Western Christendom was one of fear; in 1215 his works were
condemned by the statutes of the University of Paris. But, largely
through the labours of one supremely great and saintly intellect,
Aristotle’s thought was saved for the Christian Church. St.
Thomas Aquinas, with his clear delimitation of the spheres of
philosophy and theology and his doctrine of an ordered and
organic relation between them, achieved the final synthesis of the
Judeo-Christian revelation with Greek philosophical thought. It
is a little ironical that it was the irreligious Aristotle who became
for Christians the Philosopher par excellence rather than the much
more devoutly minded Plato, to whom the majority of the Fathers
had leaned. The answer is really very simple. It is that Aristotle _
had no religion to speak of, and therefore could be given one,
‘while Plato had one, and it was largely false. In the “baptism”
of Aristotle by St. Thomas, Greek philosophy found its culmina-
tion and its true home. And the Aristotelian arguments for a God _
who drew the world to him by the sheer force of his beauty
without, in all probability, any consciousness that the world even
{ existed were transformed, in the hands of the Angelic Doctor, into
/ the Five Proofs of the existence of a living and loving ecertial
i
|
)
Creator and Preserver of heaven and earth.
S. —_
1 Namely, the idea of Logos or “Word.”
2 Cf. W. Turner, History of Philosophy, p. 320.
CHAPTER II
THE MEANING OF “GOD”
HE first of the Thirty-nine Articles of the Church of England
sums up the Christian doctrine of God in the following words :
There is but one living and true God (unus est vivus et verus
Deus), everlasting, without body, parts or passions, of infinite
power, wisdom, and goodness, the Maker and Preserver of all
things both visible and invisible. And in unity of this Godhead
there be three Persons, of one substance, power, and eternity;
the Father, the Son, and the Holy Ghost.
This article deals first with the unity and then with the trinity of
God, in the two sentences of which it is composed. The second
of these falls outside the scope of this book, the first 1s its primary
- concern.
It is immediately clear that this sentence, and the similar state-
ments which we might find in other confessions, gives us a descrip-
tion rather than a definition of God. It catalogues the most
important truths about God which the Christian religion holds,
but it makes no attempt to show that they are consistent with one
another or that all of them are logical consequences of some one
more fundamental truth. And, indeed, in the strictest sense of the
word “ definition’’—what is technically called “‘logical definition”’
—it is impossible to give a definition of God. For logical definition
proceeds by the method of genus and differentia; it singles out the
being (or beings) with which it is concerned from some larger
class by attributing to it some specific character which is felt to
belong peculiarly to it. Thus, when we define man as a rational
animal, we single him out from the class of animals by means of
the differentiae “rational” ; if we define King George IV as the
first gentleman of Europe, we single him out from the class of
European gentlemen by means of the differentia ‘‘first.”” Some
differentia, of course, express the real nature of a being far more
intimately than others; there is a certain arbitrariness about the
definition just given of King George IV that does not attach to
that just given of man. Some philosophers have held that there is
for every being some definition—the essential definition—that does
8
hi THE MEANING OF “GOD” 9
not merely enable us to pick it out from the whole aggregate of
beings of all kinds, but also expresses its essential nature as a
specific entity, though they would add that the task of framing
such a definition is usually impossible since the inner essence of
things is not in general accessible to the human mind. “Rational
animal,” they would say, is what we mean by “‘man,”’ but no one
would say that ‘‘the first gentleman in Europe”’ is what we mean
by “King George IV,” even if the definition is accurate and
adequate. We are not, however, concerned with this distinction
here, but simply with the fact that God cannot be defined by
genus and differentia in any way.
But, it may be objected, is this ee Has not God been des-
cribed by such phrases as “the Supreme Being”? and “the
Heavenly Father,” and does not this imply that he can be singled
out from the genus “‘being”’ by the differentia “supreme” or from
the genus “father” by the differentia ““heavenly’?? As a mere
matter of verbal logic this might seem to be so, but we must not
be led astray by words. We cannot lump together in one genus
God and everything else, as if the word “‘being”’ applied to them
all in precisely the same sense, and then pick out God as the
supreme one. For if God is the Supreme Being, in the sense in
which Christian theology uses the term, “being” as applied to
him is not just one more instance of what “being”? means when
applied to anything else. So far from being just one item, albeit
the supreme one, in a class of beings, he is the source from which
their being is derived; he is not in their class but above it. Nor, to
take the other example, can we lump together in one class God
and all other fathers, and then pick out God from among them
as the one who happens to be heavenly. So far from being one
item in the class of fathers, he is, as St. Paul told the Ephesians,
the Father from whom every fatherhood in heaven and on earth
ne
isnamed.! In the technical phrase, when we apply to God a term ~
/
j
H
which is normally used of other beings, we are using it not univo-
cally but analogically; for he is not just one member of a class
with them, but their ground and archetype.
~St.. Thomas puts this point more strictly philosophically by first”
/remarking that if God were in any genus it would be that of being,
/ and then showing that being cannot be a genus. For, he argues,
+—following Aristotle, the differences which determine a genus to the
species contained in it cannot themselves be members of the genus.
1 Eph. iii, 15.
ioe HE WHO IS
(For example, when we define man as a rational animal, the
- difference ‘‘rational’? must be distinct from the generic essence
“animality,” otherwise it could not differentiate it.) But every
difference must be an instance of being, or else it would
be simply
non-existent. Therefore, bebeing cannot bea” genus.1 This con-
clusion has, of course, veryfar-reaching-consequenees. It is the
basis of the doctrine of the analogy of being, which permeates the
whole of the Thomist metaphysics. It involves, for example, that,
since being is not a genus that could be determined to its instances
by the addition of differentie, it must be determined to them by
its own inner dynamism. As M. Maritain has written, “What is
primarily known . . . is being. But nothing can be added to it
extrinsically to differentiate it, for all its differentiations issue from
its own depths, as some one or other of its modes.’”?
If, however, we cannot, in the strict sense, give a definition of
God, can we find some nominal or quasi-definition, that is to
say, some form of words which will apply to God and to God
alone; and, if more than one such form of words is available, is
there one that expresses more fully than any other what God really
is? What, that is to say, is the meaning of ‘‘God’’? Or, in more
accurate philosophical language, what is the formal constituent of
deity? What is God’s metaphysical essence? What_really1 makes
God God? Par
This was the question that was exercising the mind of a fat and
silent young Italian boy in the thirteenth century, who startled his
teacher by suddenly bursting out with the question, ‘What is
God?’ Years after he gave his own answer in the assertion that
the most proper name of God is He who 1 is. But this is to antici-
pate.
Many of the Greek philosophers would no doubt have replied
in effect that the formal constituent of God was thought, and
St. Augustine is following in their footsteps when in his Confessions
he rises to God as subsistent Truth. In the late thirteenth century
the Franciscan Duns Scotus regarded infinity as the primary
) divine attribute; many have given priority to intelligence or to
' goodness; te like the modern Russian, Nicolas Berdyaev, to
freedom. Many Christians would no doubt feel that, in view of
the plain declaration of the Beloved Disciple, the fundamental
1§. Theol., I, iii, 5. Cf. S.c.G., I, xxv. 2 The Degrees of Knowledge, E.T., p. 259.
3 Cf. R. P. Phillips, Modern Thomistic Philosophy, TI, p. 305 f.; M. C. D’ Arcy, Thomas
Aquinas, p. 165 f.: R. Garrigou-Lagrange, Dieu, son existence et sa nature, E.T., II, p. 3 f.
r THE MEANING OF “GOD” II
fact about God is that he is love, and it will be well to pause for a
: moment to ask whether this is so.
Certainly we must admit that, from the point of view of
Christian devotion, the fact that God is Love is all-important.
“God is Love,” wrote St. John to his.flock, ‘‘and he that dwelleth
in love dwelleth in God and God in him”’; and a whole host of
mystics and saints—Augustine, Bernard, Bonaventure, John of the
Cross, to mention but a few—have found this to be the basic and
all-inclusive truth about their experience of their Creator and
Redeemer. But the attribute that is primary from the point of
view of devotion may not necessarily be primary from the point of
view of theology or of philosophy. For God has other attributes,
such as power, wisdom and justice, and it does not seem possible
to derive these from the fact of his love. Is there not something
even more fundamental, from which love and all these other
attributes can be deduced? The Thomist tradition has answered
J
this in the affirmative when it has asserted that the formal
constituent of deity is Being.
But before we look more closely at this it will be worth while to
consider how St. Anselm of Canterbury in the twelfth century
dealt with the question. In the second chapter of his ‘‘ Address to
God concerning his Existence,” to which he gave the title Proslo-
gion, he gives the famous definition of God on which the whole of
his subsequent discussion is based: ‘‘Of_a truth we believe that
thou art something than which nothing greater can be thought, aliquid quo
majus nihil cogitari potest.” We must see precisely vwhat this means.
Xirst we must notice that the word is cogitart, “thought,” not
‘imagined.’ St. Anselm does not limit his idea of God by the
capacity of the human mind. A simple illustration may help here.
If I shut my eyes and try very hard to picture a polygon with a
large number of sides, there is a limit tomy power of visualization.
I may be able to distinguish between a heptagon and an octagon,
but beyond that the images which I form will probably be vague.
My visual imagination has a limit, depending, of course, upon my
personal make-up; suppose it is reached when the number of sides
is eight. The most multilateral polygon that I can imagine is then
an octagon, but I can easily think of a polygon with nine sides or
ten, or for the matter of that with ten thousand. In this particular
sphere my powers of conception are unlimited, while my powers
of imagination are certainly not. In a somewhat similar way, in
St, Anselm’s definition of God we are not concerned with what we
12 HE WHO IS
‘can imagine but with what can be thought; God is not limited
by the
capacities of the human mind.1 ar oevine
ondly, the- words aare “that than which nothing greatercan
be hought,” _not. “that than which nothing”“greater actually
exists.” > Again, a simple parallel may
m make this ciear. Somewhere
“in thé world (or, perhaps, somewhere in the remote depths of
starry space) there is an elephant which is heavier than any other
elephant now alive. I do not know where it is, nor how much it
weighs; nor, I imagine, does anyone else. But its existence is
indubitable, unless indeed there are several elephants of precisely
the same weight, which is very unlikely. Suppose it weighs x tons.
It is the elephant than which no heavier elephant exists. But it is
not the elephant than which no heavier elephant can be thought,
for I can easily conceive the existence of one whose weight is
x + 1 tons, and that will be heavier. So, to return to St. Anselm,
he is not simply applying the name of God to the greatest being
that actually exists, for that might for all we know be finite. God,
for him, means that than which no greater can be thought, and
j
f
if so the question immediately arises whether God, as so defined,
~-—-exists or not.
In the third place, it may be asked whether, after all, the defini-
tion really means anything. God, St. Anselm has told us, means
that than which nothing greater can be thought, but what does
the word ‘greater’? mean when applied to God in relation to
creatures? It obviously does not mean ‘“‘larger,” in the sense of
spatial extension, or “‘heavier”’ in the sense of physical mass; it
obviously does mean ‘‘more wise”’ and “‘more powerful,” and it
presumably means an infinity of things besides, of which we have
no idea. Now there is a real question as to whether the definition
in fact has a meaning, and we shall consider it later when we come
to deal with St. Anselm’s attempt to prove that God exists; it is
not, however, what we are concerned with here. The present
difficulty disappears if we accept—as we must, if any statements
that we make about God mean anything—the scholastic doctrine
of analogical predication, according to which the terms of human
speech when applied to God neither mean precisely what they
mean when applied to finite beings nor are completely meaningless.
1 The geometrical illustration used must not be taken to imply that we can form an
adequate concept of God, as we can of the ten-thousand-sided polygon. It is the notion
of God’s existence rather than the conception of his nature with which we are con-
cerned. As St. Thomas says, we know that God iiss rather than what he is. i We shall
return to this point later.
<4 THE MEANING OF “GOD” 13
It will now be clear that St. Anselm’s approach to God, like
that of the later theologian Duns Scotus, sees God’s formal con-
stituent as his infinity. St. Thomas, however, while he readily
accepts the Anselmian definition as giving a true description of
God,! sees God’s fundamental attribute as that¢of.self-subsistent
being.? “Ant inthis heisnotmerely philosophizing inthe abstract.
His starting-point is the ‘“‘metaphysic of Exodus,” the revelation
of the name of God as “I am that I am.” Ego SUM qui sum, ait:
sic dices filiis Israel, Qui est misit me ad vos. And the conception of
God as ipsum esse subsistens, subsistent beingi
itself, is fundamental to
his whole discussion of the.divine nature.’ It draws into a unity
all the other attributes and operations of God: simplicity, perfec-
tion, goodness, infinity, immutability, eternity, unity, his character
as Prime Mover, as Uncaused Cause, as Sufficient Reason, as
Perfect Pattern and as Final End of all things. It involves that, if
God does exist, his existence is identical with his essence. It means
that heisnot:merely the ens maximum, the greatest being that exists,
but the maxime ens, that which completely. is...So far from being,
as might appear at first sight, an arid and lifeless philosophical
abstraction, St. Thomas’s definition of God is fertile of all the
fullness of Catholic devotion. From it we derive the great truths
that Godis Life and Love.and_ Power, and if, as we shall see, the
full content of these truths only appears when the arguments of
philosophy are supplemented by the revelation which God has
given of himselfin his Son Jesus Christ, it is none the less true that
natural theology, as St. Thomas develops it, is something much
more than a sterile philosophical speculation.
1 §.¢.G., I, xi; S. Theol., I, ii, 1 ad 2. We may add that the great seventeenth-century
Anglican divine, Thomas Jackson, while taking infinity as the primary attribute of
God, insists that it is infinity in being (Works, V, p. 22 f.). He goes on to assert the
absolute identity of the divine essence and attributes and then argues for the infinity
_of the various attributes, immensity, eternity, wisdom, ee, and so on.
Gs2.** The self-same thing which God is, is his existence.” ($.c.G., I, xi.)
3 It should be noted that St. Thomas does not state that self-existent being is the
S formal constituent of deity. It is, however;-widély held that this was his view.
CHAPTER III
EXPERIENCE AND REVELATION
HYdo we believe God exists? This is an extraordinarily
difficult question to answer, for several reasons. In the
first place, it depends very much upon us. It may quite well be
the case that no two people have come to believe in God in pre-
cisely the same way, and that no two people would give exactly
the same set of reasons to justify their belief. But this does not
mean that their reasons are necessarily unsound, for there may be
quite a number of different arguments leading to a particular
conclusion and all of them may be perfectly valid. If, however, we
try to set up a body of argument which we feel would convince
a perfectly reasonable man, we shall probably lay ourselves open
to the charge of artificiality. For the perfectly Reasonable Man is
as much an abstraction as is the Economic Man or the Average
Man, as Mr. A. P. Herbert has amusingly demonstrated in one of
his ‘‘ Misleading Cases.’ ‘‘ Devoid, in short, of any human weak-
ness, with not one saving vice, sans prejudice, procrastination,
ill-nature, avarice, and absence of mind, as careful for his own
safety as he is for that of others, this excellent but odious char-
acter,” said Mr. Justice Marrow, “‘stands like a monument in our
Courts of Justice, vainly appealing to his fellow-citizens to order
their lives after his own example.... All solid virtues are his,
save only that peculiar quality by which the affection of other
men is won.” And, in fact, he is a myth.
As a matter of experience, most Christians have acquired their
belief in the existence of God in extremely elaborate ways, which
have varied widely from case to case, and which it is most difficult
to disentangle into their component parts in such a way as to
display a logical and coherent argument. This is not, however, as
discreditable to Christian belief as it might seem. For the same
thing is true about almost all the beliefs by which our lives from
day to day are governed. Very few married men, for example,
could give a perfectly watertight answer to the question, ‘‘ How do
you know that your wife really loves you?” Presumably, in most
instances, if he was really forced to it, a man could put up some
sort of case for the fidelity of the lady in question, but it is very
14
EXPERIENCE AND REVELATION 15
doubtful whether it would be sufficient to convince a court of law,
still less a professional logician ; and it would in all probability bear
only the remotest resemblance to the process which had actually led
up to the establishment of his home.
It is very much the same in regard to belief in the existence of
God. The trouble is not that there are no reasons, but that there
are so many, and that it is impossible to discuss all of them at once.
Furthermore, many of them are of the practical type which we
constantly use in the affairs of common life but which are difficult
to state in a clear and logical form. Nevertheless, it is important
to see what rational grounds can be alleged for belief in God, if
only for the sake of rebutting the contrary arguments of unbelievers.
First, Jhowever, it may be well to inquire why any arguments
should be necessary.
For, to revert to the parallel which we have just used, there
might be some point, on occasion, in asking a man for his reasons
for believing in his wife’s affection, but we should hardly ask him
why he believed that she existed. If we did, and were able to
convince him that our question was serious, he would presumably
reply, ‘‘Why, there she is. Can’t you see her?”’ And it is the
existence of God, and not any of his attributes, that we are con-
cerned with at the present moment. Why do we need to ask for
arguments and proofs for this?
The answer pretty plainly is that, while we have certain natural
faculties for perceiving material objects, such as trees and stones
and wives, we have no natural faculty for perceiving spiritual
beings.1 Ttis indéed true that some naturally devout Christian
thinkers have asserted the existence in man of what, by a rather
distant analogy, they have sometimes called a “religious sense,”’
but very few of them have claimed for it the immediacy of know-
ledge that is claimed for the physical senses. It is possible, we may
admit, to assert that the bodily senses are open to error, but very
few, even among philosophers, have claimed that sensible experi-
ence is always and entirely delusion. ‘The fact is that it seems to be
commonly accepted that, while the evidence of the bodily senses
needs definite and convincing arguments for its rejection, the
evidence of any alleged religious sense needs arguments for its
justification. We might sum this up by saying that, whether there
is a specific religious sense in man or not, it is certainly not as
immediate and sure in its operation as are the senses of the body.
1 The view that we have such a faculty is called ‘“‘ontologism.”
16 HE WHO Is
If you want to know whether there is a tree in the garden or not,
you merely have to go there and look, and as often as you turn
your eyes in the right direction you will see it. It is impossible to
produce at will a similar direct apprehension of God.
Having made the foregoing provisos let us then see what
grounds can be urged for belief in the existence of God. We may
divide the spheres in which man has claimed a knowledgee of divine
realities intg.tthree} and“éach of them hasbeen claimed~as~pro-
viding the main~basis for belief in God’s existence. These are
religious experience, revelation and reason.
~<Tn recent years a number of Christian apologetists have tried to
find the main justification for belief in God in the sphere of
“‘religiofis experience,” which they have sometimes interpreted as
meaning a man’s whole experience of life as possessing a religious
quality, and sometimes as meaning the experience of certain com-
paratively rare moments in which the soul is filled with a peculiar
and almost indescribable feeling of the certainty of divine reality.
As examples of discussions upon these lines we might instance
Otto’s famous book, The Idea of the Holy, with its characterization
of the essential content of religious experience as that of a mysterium
fascinans et tremendum, and Canon Lindsay Dewar’s more recent
work, Man and God.
Now those who claim to have had religious experience are
agreed in ascribing to it fwo_qualities. The“first) is that it is
uniquely intimate and convincing, the second is that it is, in its
essence, ineffable. The consequence is that, while it is extremely
impressive to the person who has it, it is impossible adequately to
describe it to anyone else and so to make it the basis of an argu-
ment. Either the other person has had it, in which case its very
intimacy will have convinced him already, or else he has not had
it and its incommunicability will make it impossible to convince
him of its genuineness. The most that could be done would be to
try to persuade him to put himself in the frame of mind in which
he might perhaps experience it for himself, but even this raises
difficulties, for it is pretty certain that religious experience cannot
be turned on at will. The result of this is that, while religious
experience, especially in those higher mystical forms which are
known as the spiritual betrothal and the spiritual marriage, may
well carry such conviction to the experient that anything in the
nature of arguments for the existence of God seems to be
ludicrously unnecessary, it does not seem capable of forming the
4
EXPERIENCE AND REVELATION hieg/
basis of arguments to convince
the outsider. As Professor Gilson
has said, with reference to the great philosopher of religious
experience, ““After reading W. James, I still want to know if my
religious experience is an experience of God, or an experience of
myself.”? There is a good deal to suggest that all that can be
- communicated in words—and even this only with great difficulty
—is some of the psychological concomitants of the experience—
states of feeling and the like—and that these can be paralleled in
certain conditions of mental exaltation which are anything but
religious in their origin, and which may arise from sheer sensuality
or from the administration of drugs. The so-called “anesthetic
experience” is well known.
Even those who claim to have had a convincing experience of
divine reality themselves are often extremely reluctant to admit
the authenticity of similar experiences in others. Dr. E. W. Barnes,
_in his Gifford Lectures on Scientific Theory and Religion, describes
_ his own experience in the following words:
\
*‘Four or five times in life, the first time when I was a boy some
fourteen years old and the last time at the age of thirty-three, I
have felt, enjoyed and wondered at a sudden exaltation which
seemed to carry with it an understanding of the innermost nature
of things. So vivid has been the experience that I could to-day go
to the exact spot in the street of the Oxfordshire village where the
flash of revelation first came. Always such experience has
occurred in sunshine and out-of-doors, never in church. Always
it has been unexpected. Always I have been alone. There has
never been ill-health as an exciting cause. On the last occasion,
which still remains vivid, I sat down in the early afternoon on a
piece of bare turf in a fern-covered moor near thesea. Iremember
that I was going to bathe from a stretch of shingle to which the
few people who stayed in the village seldom went. Suddenly the
noise of the insects was hushed. ‘Time seemed to stop. A sense of
infinite power and peace came upon me. I can best liken the
combination of timelessness with amazing fullness of existence
to the feeling one getsin Watching the rim of a great silent fly-
wheel or the unmoving surface of a deep, strongly flowing river.
Nothing happened: yet existence was completely full. All was
clear. I was in a world where the confusion and waste and loss
inseparable from time had vanished. At the heart of the world
there was power and peace and eternal life. How long this blank
trance, so full and so empty, lasted I cannot say. Probably a very
1 Reason and Revelation in the Middle Ages, p. 97.
18 HE WHO IS
short time indeed. It passed, leaving me neither tired in body nor
mentally irritable. The memory remains. And it is because an
inexplicable quality of supreme significance attaches to it that it
remains precious.”?
Later on in the same chapter Dr. Barnes quotes from a letter
written by a church worker describing her experience in a convent
chapel where the Blessed Sacrament was reserved :.
‘Immediately I entered I was gripped by a sense of a Holy
presence, the Presence of God, so that I trembled as I knelt. Then
I had to go lower on my knees. I could not help myself falling on
myface. Then I was flooded with the love of God as well as awed
by his nearness. I could not move for a long time. When at
length I felt I must rise and leave the Chapel, I trembled so that
I could scarcely walk. From that time my happiest hours of
communion with our Lord were spent in that presence. I
approached it in fear and trembling each time. ... I cannot say
that now I have such exceptional experience.’’?
And here is Dr. Barnes’s comment:
“Of course, the crisis of the churchworker is easily explained as
the result of suggestion operating in the lower unconscious levels
of her mind. She had heard much argument as to the value of
‘reservation’; and, unknown to herself, had in her unconscious
mind accepted the doctrine of a spiritual presence connected
with the consecrated elements. The sudden uprush of latent
convictions had a memorable splendour; but significantly there
was no repetition of the experience after the period of crisis had
passed. ‘The new belief with its emotional atmosphere was
within a few days firmly established.’’?
1 p. 620. 2 p. 634.
3 p. 634. With the above experiences we might compare the following: “I was not
more than eighteen when an inner and esoteric meaning began to come to me from
all the visible universe, and indefinable aspirations filled me. . . . There was a deeper
meaning everywhere. The sun burned with it, the broad front of morning beamed
with it; a deep feeling entered me while gazing at the sky in the azure noon and in the
star-lit evening.” “I was aware of the grass blades, the flowers, the leaves on haw-
thorn and tree: I seemed to live more largely through them, as if each were a pore
through which I drank.... I was plunged deep in existence, and with all that
existence I prayed.”’ (Richard Jefferies, The Story of My Heart, pp. 199, 14, 15, qu. by
E. I. Watkin, Philosophy of Mysticism, p. 376.) Cf. also the experience at a Procession
of the Blessed Sacrament described by Mr. Alfred Noyes in The Unknown God, p. 304 f.
As an experience of a very peculiar type, which raises the question of the subjective
element acutely, we might instance the three appearances of the Divine Sophia to
Vladimir Solovyev. (See Karl Pfleger, Wrestlers with Christ, p. 223 f.)
EXPERIENCE AND REVELATION 19
One is tempted to wonder what a sceptical psychologist might be
able to do in the way of accounting for Dr. Barnes’s own experi-
ence, but the above passages have been quoted simply in order to
illustrate the fact already mentioned, that religious experience
may be absolutely convincing to the experient and entirely uncon-
vincing to somebody else.
One of the most trenchant criticisms of the argument from
religious experience that has been given in recent years will be
found in Dr. F. R. Tennant’s great work, Philosophical Theology;
it is in substance repeated in his Philosophy of the Sciences.2, In the
argument from experience, he asserts,
“the point that is of epistemological import, is the assertion that
numinous Objectivity Reality
or is cognized with zmmediacy like
that ofsensatio [sc. the act of sensation], and as distinct from
objectivity of the imaginal or the ideal order. A distinct faculty,
not included in such as are contemplated in the psychology of
natural knowledge, is thus affirmed; for valuation and emotion
are not cognition of existents. Now, in sensatio, the particular
quale of the impression is given; but there seems to be no cor-
responding or quasi-impressional quale presented, in alleged
apprehension of the numinous Reality in the numinous pheno-
menon or thing.... And certainly the clearer conceptions of
the numinous, characteristic of more highly developed religion,
owe their definiteness to discursive thought. The vague original
suggests the imaginal or ideal, rather than the underived such as
we encounter in the concrete percept. . . . The numinous Real is
indeterminate enough to enter equally well into a multitude of
diverse mythologies and religions; it therefore seems to partake
of the nature of the vague generic idea, rather than to be com-
parable with an underived and ‘perceptual’ object.’’®
Again he writes:
‘*The imaginal and the ideational are objective. And emotional
attitude can be evoked by such objects as well as by the perceptual,
the ‘feelings’ being as profound and intense in the one case as in
the other. The numinous... must be objective, but it is a
further question, whether it is not conceptual; whether it is also
Real; or only ideal, imaginal, or even illusory.’’*
1 J, ch. xii. 2p. 167 f. 3 Phil. Theol., I, p. 309.
4 Phil. Theol., 1, p. 310. For Tennant, “objective” (with a small “‘o”) means
having the status of the object of a mental act, but not necessarily having existence
outside the mind concerned in it,
Cc
20 HE WHO IS
**We may,” he admits,
“believe in the Beyond, or in God, on less direct grounds reached
by more circuitous paths ; and then reasonably interpret numinous
or religious experience in terms of the theistic concept and world-
view: on the way back, so to say, as distinguished from the way
out. But the short cuts of‘immediacy,’ often pursued since the
downfall of rationalistic proofs, seem to owe their seductiveness
and their appearance of being other than ‘no thoroughfare,’ to
the prevalence of the two confusions that have just been
mentioned,’’+
namely, the assumption that what seems to be immediate is really
so upon reflection and analysis, and the ignoring of the fact that
the objective includes the imaginal, etc., as well as the Real.
In dealing with specifically mystical experiences Tennant dis-
tinguishes between those that are strictly ineffable and those that
admit of translation, partly or wholly, into terms of ordinary
imagery and knowledge. Of the former he remarks that, by their
very nature, they cannot be made the basis of rational argument,
while, in the case of the latter, the imagery adopted is almost
invariably derived from the mystic’s past history and beliefs and
so cannot belong to the alleged exterior object of the experience.
Tn neither instance, therefore, can we argue from the experience
‘to the God who is asserted as its object; the heart of the experience
\is strictly ineffable and therefore cannot be discussed.
~ Tennant’s conclusion, therefore, is as follows:
“What is called ‘the truth of religion’or ‘the validity of
religious experience, ’cannot be established by the ipse dixit of that
type of experience. Ifit is tobeestablished, itmust beasreasonable
inference from discursive ‘knowledge’ about the world, human
history, the soul with its faculties and capacities; and above all,
from knowledge of the interconnexions between such items of
knowledge. Thence alone are derived the notions of the
numinous, the supersensible, the supernatural, and the theistic
idea of God. . . ..Knowledge of God, on this view, is in the same
case with knowledge of the soul, of other selves, and of the
Reality behind the sensible ‘worlds’ of individual experients.’’?
1 Phil. Theol., I, p. 311.
~ 2 Phil. Theol., I, p. 325. One or two comments on Tennant’s remarks on mysticism
seem to be called for. One is that, in Catholic mystical theology, “revelations” are
discouraged rather than sought after; it is supernatural charity, rather than revela-
tions, that unites the soul to God, and in the spiritual marriage extraordinary mystical
phenomena almost invariably cease. And the part played by the mystic’s own mind
EXPERIENCE AND REVELATION 7441
As will be seen later on, we shall have some criticisms to pass
upon Tennant’s own approach ; Ibut we are at one with himin his
rejection of religious experience as providing the primary tbasis for
a-rational approach to the existence. of God. We shall see in a
later chapter that the argument from experience is not to be alto-
gether excluded, and we may note in passing that it forms one of
the strands of the ‘‘threefold cord’? in Professor A. E. Taylor’s
essay on “The Vindication of Religion,’’! but we shall maintain
that religious experience can only be properly assessed against the
background of a theology which has taken as its starting-point
the existence of finite beings; then we shall be able to express
approval of the statement of Fr. D’Arcy that the argument from
experience may well turn out to be “nothing but the old argu-
ment from contingency looked at from inside instead of from
eae
NUS
outside.’”?
Before leaving this question it may be worth remarking that
Catholic theology has been on the whole very reluctant to admit:
that, even in authentic religious experience, there is anything that
could validly be described as a direct apprehension or an imme-
diate knowledge of God. The only exception that it would make
would be in the rare case of mystical union, properly so called, and
then it would say that the soul was apprehended by God, that it
was seized upon by him, rather than that he was apprehended by
it. And even then it would hesitate to say that the soul saw him
in his essence, for it is separated from him both by the gulf that
divides the finite from the infinite and by its union with the body,
from which, even in ecstasy, it cannot be altogether severed. And
it would be even more emphatically insisted that i in non-mystical
since our knowledge of God here on earth is always through his
in providing theiimagery of revelations is recognized quite definitely; Poulain gives
thirty-two cases of mystics of accepted sanctity whose revelations contained errors.
(Graces of Interior Prayer, E.T., p. 321 n.) Again, Christian prayer as practised by the
mystics is far from being, as Tennant apparently thinks (p. 322), a technique of self-
stupefaction resulting in a pathological state akin to the hypnotic; and it is important
to distinguish, as Maréchal does (Studiesin the Psychology of the Mystics, E.T., p. 170 f£.),
between a simplification of the content of the mind which is impoverishing and one
which is enriching.
Furthermore, Tennant seems to be exaggerating when he writes that “a person’s
convincedness as to the truth of a dogma is a sufficient condition of its efficacy in
ministering to his spiritual life, and that the truth of the credendum is then a superfluous
condition.” (Phil. of the Sciences, p. 179.) Conviction indeed supplies a strong subjec-
tive dynamic, but surely true beliefs are likely to be more spiritually fruitful than false
ones.
1 Essays Catholic and Critical, p. 70 f. 2 See ch. viii and p. 92, infra.
22 HE WHO IS
effects.1, This is not, however, to deny that we can really know
him, for knowledge by effects is real knowledge, and God’s effects
upon the soul in which he dwells by grace can be amazingly
intimate. ‘“‘No man hath seen God at any time”’—this is indeed
true—but “‘he that dwelleth in love dwelleth in God and God in
him,”’ and the promise of the Incarnate Son is, “If any man love
me, my Father will love him, and we will come unto him, and
make our abode with him.”
1 Cf. §. Theol., I, xii, 11; S.c.G., III, xlvii. With the growth in recent years of the
view that God gives to all human beings at least a remote and general vocation to
contemplative prayer (see, e.g., Garrigou-Lagrange, Perfection chrétienne et contemplation)
a tendency has appeared in some theologians to admit that in this life it is possible
without any extraordinary grace to have a direct, though obscure, knowledge of God by
pure species and not by the intermediation of sensible things. Dom John Chapman
Outlines such a theory in the paper, “What is Mysticism?” printed at the end of his
Spiritual Letters. He argues (1) that the human soul is radically capable of the angelic
mode of cognition “‘by intelligible effects” and (2) that this mode of knowledge of
God, which was possessed by Adam before the Fall but lost as a result of the Fall, can
be so far restored by grace as to deliver us at least partially and on occasion from the
tyranny of sense. It must be noted, however, that it is very obscure and so could
hardly form the material for a rational argument for the existence of God and that, in
any case, it depends upon the healing power of divine grace and does not inhere in
human nature as fallen. Chapman claims to find support for his view in St. Thomas.
Cf. the essay by Mr. Christophe: Dawson, ‘‘On Spiritual Intuition in Christian
Philosophy,” in Enquiries into Religion and Culture.
It is worth noting that, while the approach to God by reason gives solid grounds for
belief in a God who is personal, it is much more difficult to derive this conviction from
“religious experience.”’ A mysterium fascinans et tremendum is not obviously personal, nor
indeed is the “power and peace and life” that Dr. Barnes experienced in the village
street ; still less obviously personal is the object of Richard Jefferies’s experience quoted
in the footnote on p. 18. And in mystical experience in the strict sense there does not
seem to be any direct apprehension of God as personal, except possibly in the ultimate
degree of the spiritual marriage. The Christian mystics, of course, know that God is
personal,.and this, combined with the fact that they are livingin intimate moral and
sacramental union with Christ, leads them to apprehend their experience as that of a
personal God, But psychologically considered there does not seem to be for the most
part a direct awareness of personality, as the impersonalistic theology of most non-
Christian mystics shows. The Christian explanation of this is, of course, that God so
exceeds the capacity of the human mind that the more the soul is brought face to face
with God the less it is able to distinguish his features; it is “‘blinded with excess of
light.” Hence the phenomenon of the “Dark Night,’ which only ceases in the com-
pletely integrated state of the spiritual marriage in which the soul attains a kind of
foretaste of the Beatific Vision. Mr. Aldous Huxley has the ingenious, but_perverse,
theory that the reason for the Dark Night is that God is really impersonal, and that it
takes the soul a long time to get used to this; he claims support for his view from the
fact that Hindu mystics, who have never really believed in a really personal God, do
not experience the Dark Night. The obvious reply is that the Hindu mystics do not
normally achieve a real mystical union with God at all; most Hindu mysticism—
perhaps bhakti is an exception—is an egocentric psychological technique. (See p. 134,
n. 1, infra.) And, in any case, the Dark Night can take very atypical forms. In the case
of certain Islamic mystics, such as al-Hallaj, the authentic features of personalistic
mysticism seem to be present. (Cf. A. Huxley, Ends and Means, ch. xiii, and Grey
Eminence, ch. iii and x; J. Maritain, Degrees of Knowledge, Part I1; E. I. Watkin,
Philosophy of Mysticism, passim, especially ch. xiv; J. Maréchal, Studies in the Psychology of
the Mystics, E.T., passim; Fr. Gabriel of St. Mary Magdalene, St. John of the Cross;
V. Elwin, Christian Dhyana.)
EXPERIENCE AND REVELATION 23
We will now pass on to consider whether the sole significant
source of our knowledge of God is to be found in divine revelation.
It must first of all be made clear that we are here using the
word “revelation” in the strict and proper sense which it has
borne in Christian theology, of the unique self-disclosure of him-
self given by God to the Jewish people and culminating in his
personal incarnation in the figure of Jesus of Nazareth, that
revelation whose record is found in the Scriptures of the Old and
New Testaments and which is preserved and mediated to the
world by the Christian Church. If God in fact exists, there is no
doubt a secondary sense which the word may bear as indicating
the part which is played by the action of God in all man’s experi-
ence of him and his works, for obviously he will not display himself
merely as a lifeless object before the gaze of men. But it is not this
“‘general revelation”’ that we are concerned with here.
Now certain theologians, particularly of the present-day neo-
Protestant school, have tidifitained with considerable vehenience
that the only knowledge of God that man can have—or, at any
rate, the only knowledge of him that will not be hopelessly per-
verted and distorted—is given to us by the deliberate and unilateral
action
of God in Jesus Christ, and that man is an entirely passive
recipient of it, accepting it by a pure act of faith as something
which his own powers are totally inadequate to approach. Thus
Professor Karl Barth in his Gifford Lectures on The Knowledge of
God and the Service of God explicitly declares himself as “‘an avowed
opponent
ofallnatural theology.” “God,” he writes,
‘fis the one and only One and proves himself to be such by his
being both the Author of his own Being and thesource. ofall
knowledge..of himself. In both these respects he differs from
everything in the world. A God who could be known otherwise
than through himself, i.e. otherwise than through his revelation
of himself, would have already betrayed, eo ipso, that he was not
the one and only One and so was not God. He would have
betrayed himself to be one of those principles underlying human
(_ systems and finally identical with man himself.’”?
And that this “‘revelation” to which Barth refers is not general
revelation is shown when he goes on to say:
“True knowledge of the one and only God . . . is based on the
fact that the one and only God makes himself known. Every-
1p. 6. 2 p. 19.
24 HE WHO IS
thing is through him himself or is not at all. He makes himself
known through himself by distinguishing himself in the world
from the world. Otherwise he cannot be known at all. He can
be.known because he arises—‘Arise, O Lord’—in human form™>-
and therefore in a way that is visible and audible for us, i.e. as
the eternal Son of God in the flesh, the one and only God in
whom we have been called to Pepa Jesus Christ. . ~_Because
~~he manifests himself thus, he makes himself. toon to_us not
through revelation ofsome sort or other, but through tthe fact
= of his self-revelat
We may indeed be grateful for so definite a reaction against the
position of Liberal Protestantism, which tended—and frequently
more than tended—to view Ghristianity and the non-Christian
religions as so many human activities upon the same level, as a
number of parallel efforts upon the part of man to attain to the
ultimate’ truth of things. And we are in the whole tradition of
Christendom in recognizing that the things that the human
reason can find out about God look very small and pale by the
side of the tremendous truths that God has revealed about himself
in Christ. Where this tradition, however, parts company with
Barth is in its insistence that, however small and pale they appear,
the deliverances of reason are none the less true in their sphere,
and that they form a necessary base upon which the structure of
revelation is erected. Barth himself is—or was at one time—
prepared
to recognize the occurrence of truth outside the sphere
of revelation.in the strict sense, and éven to concede to it, in a
secondary application, the name of revelation. But he has nothing
but contempt to offer it. ‘‘The discussion as to whether there is
not revelation also in ‘other religions’ is,” he writes, “superfluous.
We need not hesitate to grant this to them, for revelation to them
clearly means something very different.’’? With all that he has
written about the uniqueness and supreme importance of the
Revelation that is in Christ we can heartily agree. But we cannot
conclude, as he does, that truth about God acquired in other ways
is therefore irrelevant and to be despised._For all truth is ulti-
mately from_God, and something more needs to be said than that,
in comparison with the act of God, the acts of man are as nothing.
- is that, man and his natural powers being themselves the work
38205
2oo. Barth’s essay in the symposium Revelation, p. 45. A more temperate expres-
sion of this point of view is to be found in Emil Brunner. See, e.g., The Mediator, E.T.,
Book I, ch. 1.
pA EXPERIENCE AND REVELATION 25
fot God, there must be an organic relation between what man can
__ find out about God and what only God can make known to him.
It is for this reason that Catholic theology has distinguished
between the sciences of natural and revealed theology, and has
also maintained that the two are intimately connected. The
problem of their exact relation has received a good deal of atten-
tion in recent years, especially from the two distinguished French
_ writers, M. Maritain and M. Gilson.! The_precise question with
which they have been concerned iis how, if philosophy (ofwhich
.ee
natural theology isa] part) iis simply a work of human reason,there 4
canbesuch athing as a specifically Christian philosophy. How
are we to account for the difference between the natural theology
of an Aristotle and an Aquinas? If natural theology is the work
of reason and not of revelation, how can it make any difference
whether the natural ‘theologian 3is a Christian or not? The
essence of the answer that is given ‘liesin the fact that, according
to the Christian Faith, man’s natural powers are themselves
weakened by sin, and so his natural knowledge of God is, even in
its own order, clouded and distorted; one need not be a Calvinist
to recognize that man is not only spoliatus gratuitis but also vul-
neratus in naturalibus.2. The answers given by Maritain and Gilson
are, when shorn of technicalities, substantially the same. It is that
grace not only supplies perfections that lie above the level of
nature, but also restores nature to its own proper integrity; gratia
is not only elevans but sanans too. Hence, while in principle there
is a certain limited knowledge of God which is accessible to the
human reason as such, in practice it is only in the light of revela-
tion that the human reason can function adequately and obtain,
even within its own proper limits, a knowledge of God which is
free from error. Natural and revealed theology are thus in the
abstract autonomous, _ being concerned respectively with the
sphere of reason and nature and with the sphere of revelation and
grace, buti
in the concrete a true natural theology can only be
developed 1in the light of the Christian revelation. Thus Gilson
speaks of “those rational truths ...which did not enter philosophy
by way of reason.’? The relation between natural and revealed
theology was described with force and humour by the late Abbot
1 J. Maritain, De la Philosophie Chrétienne; E. Gilson, Christianisme et Philosophie, also
Spirit of Medieval Philosophy, eh 2 CE the discussion of “La Croyance et |’Auto-
nomie de la Philosophie” given by Sertillanges in Le Christianisme et les Philosophies,
epson.
"2CR. S. Theol., IL i, cix, 20. 8 Spirit ofMed. Phil., E.T., p47.
26 HE WHO Is
Chapman in one of his letters. “It is obviously not possible
in practice,” he wrote, “to disentangle the Supernatural
from the
Natural. The two are warp and woof from which our whole experi-
ence is woven. But it zs possible to do so in theory, and SCHOLASTIC
PHILOSOPHY deals oNLY with the Natural, and therefore not with
life in all its complexity as we know it, but with the world as it
~ would be without revelation and without grace (of all kinds),
-which are disturbing factors.” Again: ‘“‘The crucial instance of
this abstract nature of pure philosophy is in Natural Theology,
which is a part of philosophy ; its subject-matter is what man can
know of God without revelation and without grace; whereas it is
(really) OF FAITH that it is within the power of every man to
have divine faith (of some kind) and that no one is ever without
‘sufficient grace.’ ... Consequently, there is not and never has
been in the world such a monster as a professor of purely natural
religion. A human being falls lower or rises higher, but is never a
simply natural man.”’+
This has been a digression, but a necessary one. Its purpose has
been to justify the position that not by revelation alone, but also
by reason, man is able to attain a true but limited knowledge of
L : God. The traditional position maintains that by reason man can
- kecome convinced of the existence of God and-of-certain“of “his
attributes, ‘but that the truth of his threefold nature as Father,
~ Son and ‘Spirit jis, while not in itself contrary to reason, beyond the
\_ power of reason to discover. It is assumed by the Anglican Bishop
Butler in his famous Analogryof’ Religion, ‘and
: we shall assume it here.
It will follow that in discussing the existence of God we shall be
relying primarily upon arguments based upon human reason.
Nevertheless, in view of the close connection which we have seen
to exist between natural and revealed theology, we shall, even in
discussing the existence of God, make frequent reference to the
Christian revelation; that is to say, we shall follow the line of
Aquinas rather than that of Aristotle.
One last remark must be made before we leave this matter of
revelation and reason. One of the grounds on which traditional
theism has refused to base belief in God simply upon the fact of
revelation is that revelation itself needs rational justification. To
accept something on. the authority. of revelation is to accept it
because one isconvinced that God has said it; and this involves a
previous conviction of the existence of God. On what can this
1 Spiritual Letters of Dom Sane Chapman, p. 192 f.
EXPERIENCE AND REVELATION 27
conviction rest? Ifit is alleged to be due to an immediate experi-
ence of God, that experience itself must be vindicated as authentic
and not illusory by the use of reason, unless indeed we were to
accept the position, which we have seen reason to reject, that man
has a natural faculty by which he can directly perceive God. If,
on the other hand, it is not due to such an experience, it must be
due to rational conviction. In either case, therefore, reason has to
be appealed to, in one way or another, before we can base any-
thing upon revelation. This is why St. Thomas says that “the
existence of God and other like truths about God, which can be
known by natural reason, are not articles of faith, but are pre-
ambles to the articles; for faith presupposes natural knowledge,
even as grace presupposes nature. ... Nevertheless,’ he adds, no
doubt having in mind the ordinary believer who has not succeeded
in disentangling the various strands which have gone to produce
conviction of the existence of God, “there is nothing to prevent a
man, who cannot grasp a proof, accepting, as a matter of faith,
something which in itself is capable of being scientifically known
and demonstrated.”! To say that a man’s conviction of the
.existence of God is based upon reason is not to say that he must be
|
capable of setting it out in the form of a technical theological
argument; it does, however, mean that before he can accept
anything upon the authority of God, he must first of all have been
__ convinced that there is a God and that God has spoken.?
1 §. Theol., I, ii, 2 ad 1.
2 St. Thomas shows a very pleasing sympathy with the ordinary busy, and not
always over-intelligent, man or woman in his insistence that “the truth about divine
things which is attainable by reason is fittingly proposed to man as an object of
belief.” (S.c.G., I, iv.) “Three disadvantages,” he claims, “would result if this truth
were left sly ‘to thei inquiry of reason. One is that few men_would have knowledge
. for three reasons. Some indeed on account of an indisposition of tempera-
ment, by }reason of which many are naturally indisposed to knowledge. . Some
are hindered by the needs of household affairs. . And some are hindered by
laziness.... The second disadvantage is that those who would arrive at the
discovery of the aforesaid truth would scarcely suctged: in doing so after a long
time. . Fitst) because this truthis. so profound. . .»-Secondly, because many things
are required beforehand... . Thirdly, because at the time of youth, the mind, when
tossed about by the various : movements of the passions, is not fit for the knowledge of
so sublime a truth... . The third disadvantage is that much falsehood is mingled with
the investigations of human reason, on account of the weakness of our intellect in
forming its judgments, and by reason of the admixture of phantasms.”’
Cf. Lancelot Andrewes: ‘‘If by knowledge only and reason we could come to God,
then none should come but they that are learned and have good wits, and so the way
to God should be as if many should go one journey, and because some can climb over
hedges and thorns, therefore the way should be made over hedges and thorns; but
God hath made his way viam regiam, ‘the king’s highway.’ Many are weak natured,
and cannot take the pains that is needful to come to knowledge; and many are
detained by the affairs of the commonwealth. ee! are cut off before they come to
age to understand reason and to attain knowledge. . . If they should in any matter
8 HE WHO Is
Those who reject this position fall into one of two opposite kinds
of irrationalism. On the one hand there are the ngo-Calvinists,
who rely pom si self-eeiennng authority ofthe Word W ofGod.
Against the ic “Maritain has made
of the most eminent of com, Piofosdar Barth. Bakth,) he says,
‘‘wishes to hearken only io God and he wishes only to hearken to
God. ... Yet when he speaks, and most of all when he speaks
in order to proclaim that man must only listen to God, it is he
himself that speaks, he himself that is heard, and it is his per-
sonality which moves and stirs his listeners.) Ox the other hand,
¥l
\ |
~there is the attitude, which has been all too common in England in
recent years, according to which any real certainty in matters of
religion would deprive faith of all its merit. For this school of
thought, ‘‘ believing where we cannot prove”? becomes the essence
of religion, and sometimes the suggestion is even made that God
has deliberately concealed himself from us in order that we may
exercise the virtue of believing on insufficient evidence—a view
which is somewhat reminiscent of the famous theory of Mr. Philip
Gosse that God created the world in the year 4004 8.c. with fossils
ready made beneath the surface of the earth as a test of our faith
in the historicity of Genesis. To it we may reply that, whether or
not the existence of God is capable of proof, the merit of faith lies
not in the mere intellectual recognition of the fact that there is a —
God, but in one’s readiness to accept its implications. Mere fides
informis can only increase one’s damnation; it is in fides caritate
formata alone that any merit is to be found. There is no particular
connection between charity and uncertainty; and faith will
be driven to prove everything by reason, it would drive them into madness. No man
can make demonstration of every thing, no not in matters of the world.” (Pattern of
Catechistical Doctrine, ch. ii.)
And Joseph Butler: “It is impossible to say, who would have been able to have
reasoned out that whole system, which we call natural religion, in its genuine sim-
plicity, clear of superstition: but there is certainly no ground to affirm that the
generality could. If they could, there is no sort of probability that they would.
Admitting there were, they would highly want a standing admonition to remind them
of it, and inculcate it upon them. And further still, were they as much disposed to
attend to religion, as the better sort of men are: yet even upon this supposition, there
would be various occasions for supernatural instruction and assistance, and the
greatest advantages might be afforded by them.” (Analogy of Religion, II, i, 1.)
Compare also the definition of the Vatican Council : ‘‘ Huic divinae revelationi tribuendum
quidem est, ut ea, quae in rebus divinis humanae rationi per se impervia non sunt, in praesenti
quoque generis humani conditione ab omnibus expedite, firma certitudine et nullo admixto errore
cognosct possint.” (Sess. III, cap. ii; Denzinger, Enchiridion, 1786.)
1 True Humanism, E.T., p. 63. Cf. E. Gilson: “God speaks, says Karl Barth, and
~ man listens and repeats what God has said. But unfortunately, as is inevitable as soon
as a man makes himself God’s interpreter, God speaks and the Barthian listens and
repeats what Barth has said.” (Christianisme et Philosophie, p. 151.)
EXPERIENCE AND REVELATION 29
not be of less value if it is a leap not into the dark but into the
light.
i eenaaee the position that has been reached. We have
considered. briefly:twovof the grounds that have been urged for
belief in God—experience and revelation. The former has been
seen to possess the disadvantage, from an apologetic point of view,
that, while it may be completely convincing to those who have it,
it is incommunicable to those who have not. TheJatter raises the
question as to how itsown authenticity is to be vindicated. This
does not by any means deprive them of all value. For it may very
well be the case that a conviction of God’s existence is not to be
obtained solely by the construction of formal rational arguments.
It may rest not upon one or two logically displayed ‘‘proofs,” but
upon a whole body of considerations, made very largely uncon-
sciously and only with great difficulty disentangled and systematized.
Belief that there is a God may be acquired, not in the way in which
we deduce from the principles of Euclidean geometry that the
angles at the base of an isosceles triangle are equal, but rather in
the way—whatever that may be—in which a man is convinced
that his wife is faithful to him, that the beer which he is just going
to drink is not poisoned, or that there is such a place as Australia.
This possibility should not, however, deter us from investigating
whether or not reason is capable of demonstrating the existence
of God; and in any case, as we have seen, it is impossible to base
belief upon experience or upon revelation without questions
arising which require the exercise of reason for their answer.
Leaving experience and revelation, therefore, we will go on to the
main task of this book and inquire what can be asserted about the
existence of God from the standpoint of human reason.
CHAPTER IV
THE TRADITIONAL APPROACH
(1) SELF-EVIDENCE OR PROOF?
E now pass on to consider what can be learnt as to the
existence of God by the exercise of the human reason. As
we have already seen, even in the discussion of religious experience
and of revelation the rational element cannot be altogether
excluded. But it was there involved in a different way from that
with which we are now concerned. There it came in simply in
order to decide whether the claims which were made for religious
experience and for revelation were justified ;here, in contrast, it is
itself the instrument by the use of which the problem of God’s
existence is to be investigated.
St. Thomas Aquinas divides his discussion of thg-existence of
| God in the Summa Theologica’ into three articles Whether the
| existence of God is self-evident
?/j
(0.‘Whether it is demonstrable?
ff )And, Tastly, whether iin fact God exists?” And he discusses the same
UH dies points in greater detail in chapters x to xiii of the First Book
of the Summa contra Gentiles.
Under.the first heading he first of all explains what is meant by
self-evidence. ““Those-things,”’he tells us, ‘‘are_said to be self-
evident’ which 2are known as soon ass the terms are "known,”? and
an
part, which, he says, is known meted as soon as it is known
what is meant by whole and part. This would seem at first sight
to exclude any possibility of argument whatsoever. Ifthe existence
of God was self-evident in this sense, we should only have to make
certain that we understood what was meant by the definition of
God in order to see immediately that he existed. And, indeed,
M. Gilson has suggested that St. Thomas’s great Franciscan con-
temporary;-St:-Bonaventure,-actually..held this position. ‘For
St. Anselm,’ he writes, “the definition of God implied a content
which our thought had to unfold in order to get at the conclusion
involved in it,” but “‘for St. Bonaventure, the same definition
becomes an immediate evidence, because it participates in the
1], i. 2 8. Theol., I, ii, 1 obj. 2.
30
» THE TRADITIONAL APPROACH I
necessity of its content. ... Since the assertion of God’s existence -
is founded upon the intrinsic evidence of the idea of God, it should
suffice to place this idea before our eyes to ensure our perceiving
its necessity: if God is God, God exists; and since the antecedent
is evident, theconclusion
isevident likewise.” St. Thomas, how-
eubs TeRTEROS the ERE of self-evidence to include the case in
which a certain amount of reasoning is needed in order to render
explicit the content which the definition contains. And he is
hence led to consider under this heading the famous Ontological
Argument expounded by St. Anselm in the second and third
chapters of the Proslogion._
St. Anselm, as we have already seen, takes as his definition of
God “something than which nothing greater.can.be.thought,” and
hest:
states his argument in the following words:
“It is one thing for a thing to be in the understanding, and
another to understand that the thing really exists. For when a
painter considers the work which he is to make, he has it indeed
in his understanding ; but he doth not yet understand that really
to exist which as yet he has not made. But when he has painted
his picture, then he both has the picture in his understanding, and
also understands it really to exist... . And surely that than which
no greater can be conceived cannot exist only in the understanding.
For if it exist indeed in the understanding only, it can be thought
to exist also in reality : and real existence ismore than existence in
the understanding only. If then that than which no greater can be
conceived exists in the understanding only, then that than which no
greater can be conceived is something a greater than which can be
» conceived: but this is impossible. Therefore it is certain that
f something than which no greater can be conceived exists both in the
¢ understanding and also in reality.’’?
f) St.Thomas states the ame argument more succinctly without any
_.. change in its substance, as follows:
“By this word [sc. “God”’] is signified that thing than which
nothing greater can be conceived. But that which exists actually
and mentally is greater than that which exists only mentally.
Therefore, since as soon as the word ‘God’ is understood it exists
mentally, it also follows that it exists actually. Therefore the
proposition ‘God exists’ is self-evident.”>
> “Philosophy of St. Bonaventure, E.T.,
e ® Proslogion, ii, In this and the idlonite extract I have allowed the translation
“conceived ’ ? to stand and have not sabstinnted the reading used elsewhere in this
book, viz. ‘‘thought.” 3 §. Theol., 1, ii, 1 Obj. 2.
32 HE WHO Is
The reaction that most people feel when they are first con-
fronted with this example of the reductio ad absurdum is very similar
to that which they feel when they see a conjuror extract a rabbit
from an apparently empty hat. They cannot explain how the
rabbit got there, but they are pretty certain that the conjuror
introduced it somehow. The transition that St. Anselm makes
from the realm of verbal definition to that of concrete existence
seems too good to be true, and yet it is not at all easy to see exactly
what is wrong with it. It has been attacked by philosophers of
very different schools, of whom we may instance St. Anselm’s
nt
(contemporary Gaunilo, St. Thomas Aquinas and Kant; it has also
been many times recast and restated, as for example by Descartes
‘and Leibniz. Gaunilo objected that if the argument proved any-
‘thing at all it could equally well be used to prove the existence of a
perfect island, by simply substituting in the definition of God the
word ‘“‘island” for “‘something’’ and the words “more perfect”
for “greater,” but to this it could be replied that the very idea of
an island than which none more perfect can be conceived is a
contradiction in terms, whereas this is not evident of God as
St. Anselm defines him. The objection, does, however, at least
raise_the possibility. that the definition is self-contradictory Sieto
occ”
show that the affirmation of necessary existence is analytically
implied in the idea of God,” remarks Gilson, “would be, as
Gaunilo remarked, to show that God is necessary ‘Gf he exists, but
would not prove that he does exist.”? It was presumably” the
recognition of this that prompted Duns Scotus to re-word the
Anselmian definition in the form ‘‘ quo cogitato sine contradictione
majus cogitart non potest sine contradictione :that which can be thought
without a contradiction but than which nothing |greater can be
thought without a contradiction, ” and then to assert that when
and only when God as thus denned has been shown to be “‘think-
able’ the ontological argument is valid.”
In fact, the objections that may be brought against the argu-
ment are of three types:
It may, in the first place, be suggested as above that for all we
know the definition may be meaningless, it may be as inherently
self-contradictory as would be the idea of around’square. The
fact that, whereas we can immediately see the self-contradictori-
1 Spiritlees Phil., E.T., p. 62.
2 Op. Ox.,= d. ii, q. 2, n. 32 (Ed. Vivés, VII, p. 479). Cf. Harris, Duns Scotus,
II, p. 168: A. E. Taylor, E.R.E., XII; p. 270, s.v. 2Pheem.”
THE TRADITIONAL APPROACH 33
ness of this latter idea, we cannot immediately see any self-
contradiction in the Anselmian definition of God is irrelevant, for
it appears to be the case that there are various self-contradictory
ideas whose self-contradictoriness is not immediately apparent.
Many mathematical logicians believe that the notion of a class
composed of all non-reflexive classes is self-contradictory, but this
self-contradictoriness is not evident from the definition. It is
suspected only because a certain argument based upon the notion
leads in practice to an absurdity. A similar contradiction arises
with the adjective “heterological.”! _May not thé Anselmian
definition
of God"bé just as absurd? The suspicion receives some
support fromthe fact that there is a very similar notion to which
it is known that nothing can correspond: that is, the ‘‘schoolboy”’
notion of “infinity.” If we define “infinity” as a number than
which no greater can be conceived, we are simply using a meaning-
less set of words, for however large a number we like to conceive
we can always conceive a greater; for example, by adding the
number ‘“‘one”’ to it. And there is at least sufficient similarity
between this definition and the definition of God given by St.
Anselm to make us wonder whether perhaps that too may really
be meaningless.”
~ Professor C. D. Broad has followed up this point in much detail.
He first points out an ambiguity in the definition itself. It could
mean either “a being such that nothing more perfect than it is
logically possible” (the comparative interpretation) or ‘“‘a being
which has all positive powers and qualities to the highest possible
degree” (the positive interpretation). Now, unless all positive
characteristics are mutually compatible, the positive interpreta-
1 A class is said to be“reflexive”? ifit fulfils its own defining characteristic. Thus the
class of mathematical ideas is reflexive because it is a mathematical idéa;-while the
class of cabbages is not reflexive, because it is not a cabbage. The question is then put:
“Ts the class of non-reflexive classes reflexive or not?”’ Hither of the answers, Yes or
No, is €asily seen to lead to a contradiction.
A predicate is said to be homological orheterological according as it can or cannot
be predicated of itself. Thus the predicate “‘verbal”’ is homological, because it is itself
verbal; while the predicate “‘pink”’ is heterological, because it is not itself pink. The
question is asked: “‘Is the predicate “heterological’ heterological or not?”’. As before,
either of the answers, Yes or No, leads to a contradiction.
A discussion of this antinomy is given in F. Gonseth’s Fondements des Mathématiques,
ch. x. Cf. B. Russell, Intro. to Math. Phil., ch. xiii; M. Black, Nature of Mathematics,
opt.
2 This possibility is seized upon as the weak point in the Anselmian argument by
Professor A. E. Taylor in his discussion of St. Anselm in the article on ““Theism” in
E.R.E., XII, p. 268. As a verbally similar, but obviously meaningless notion, he
instances that of a “‘line so crooked than none crookeder can be conceived.”
34. HE WHOIS | "i
tion becomes meaningless.1_ And unless the same condition is
fulfilled the comparative interpretation is inapplicable. For, sup-
posing that A and B are incompatible characteristics, how can we
decide whether the combination ACDEF ... , from which B is ex-
cluded, is more or less perfect than the combination BCDEF ... ,
from which A is excluded? Furthermore, the positive interpre-
tation will be meaningless unless every positive characteristic
has an intrinsic maximum; and unfortunately there are some
characteristics in our experience, such as length or temperature,
which have none. If any such fall within the meaning of
“positive powers and qualities” in the definition, the definition
will then be meaningless. And, lastly, if there is nothing that
answers to the positive interpretation, it is clear that nothing could
answer to the comparative one.?
The second objection, and it is the one which St. Thomas puts
forward, is that all that the argument proves is that, if you define
God as St. Anselm does, you are bound to think of him as existing,
but that this does not prove that he exists. That.is
to say, the idea
of a necessarily existent being does not necessitate its existence.
= As Dr. R. P. Phillips
writes, “The fact that we conceive,
and must
yconceive, of God in a certain way, namely as existing of himself,
/ in no way shows that in fact there is a Being which exists of itself,
i but merely that if there is a Being to whose concept existence
attaches necessarily, he will, if he exists at all, exist necessarily.”
In the third-place, it may be urged that existence is not a quality
like other qualities, the possession of which adds to the perfection
of a being. Ought we to say that a being which exists is more
perfect than a precisely similar being which does not exist? Is
existence a quality that can be added to other qualities, and not
rather something entirely different from qualities, which, in any
actually existing being, underlies them all? Modern logicians
have with considerable cogency criticized Aristotle for treating
existenceas a predicate; and is this not precisely what St. Anselm
was doing in his argument? —
1 Tt is, we may remark, very far from obvious that, for example, infinite justice and
infinite mercy are compatible. Cf. Garrigou-Lagrange, Dieu, E.T., II, p. 108 f.
2 F.T.S., Jan. 1939, p. 19 f. I have slightly altered Professor Broad’s. wording, but
have left the argument unchanged except to question by implication his assumption
that temperature, length, etc., as we know them must be included among the positive
powers and qualities of the definition. To assume this is to deny the principle of
analogy, and Broad presumably does deny it. But even if we accept it, his objection
holds, at any rate as a possibility.
3 Modern Thomistic Philosophy, 11, p. 266.
es THE TRADITIONAL APPROACH 35
Professor Broad has stated this difficulty very forcefully in the
following words:
“The Ontological Argument professes to make a categorical
comparison between a non-existent and an existent in respect of
the presence or absence of existence. The objection is twofold.
(i) No comparison can be made between a non-existent term and
anything else except on the hypothesis that it exists. And (ii) on
this hypothesis it is meaningless to compare it with anything in
respect of the presence or absence of existence.” —
We may, then, not unreasonably share St. Thomas’s hesitation
to admit
the Anselmian argument, much as we may admire its
ingenuity. Before we leave the question, however, it will be worth
_while to make one observation. This is, that the mere occurrence
f“of the argument shows how firmly, by. the.time of St. Anselm; the
thought. ™M. Gilson has put this with his usual directness and
~ clarity:
“That no trace of it [sc. the Ontological Argument] exists in
Greek thought is quite undisputed, but it does not seem to have
occurred to anyone to ask either why the Greeks never dreamt of
it, or why, on the contrary, it was perfectly natural that Christians
should be the first to conceive it.
“Once the question is asked the answer is obvious. Thinkers like
Plato and Aristotle, who do not identify God and being, could
never dream of deducing God’s existence from his idea ;but when
a Christian thinker like St. Anselm asks himself whether God
exists he asks, in fact, whether Being exists, and to deny God is to
affirm that Being does not exist. That is why the mind of
St. Anselm was so long filled with the desiré of finding a direct
proof of the existence of God which should depend on nothing
but the principle of contradiction. .. . The inconceivability of the
on-existence of God could have no meaning at all save in a
Christian outlook where God is identified with being, and where,
(
consequently, it becomes contradictory to suppose that we think
| of him and think of him as non-existent.’’?
—
4
If now, following the example of St. Thomas, we reject the
Ontological Argument for the reasons that have been given, what
grounds can we find for believing in the existence of God? How
does St. Thomas himself go about the matter?
1 F.1T.S8., Jan. 1939, Pp. 22- 2 Spirit ofMed. Phil., E.T., p. 59.
D
36 HE WHO IS
We must remind ourselves that the definition of God which he
assumes is somewhat different from St. Anselm’s. He thinks of
God, not primarily as aliquid quo majus nihil cogitart potest, but as
God revealed himself to Moses in Exodus, as ipsum esse subsistens.
But this is equally a definition in terms of being, and indeed it is
not difficult to argue that the twonotions have really the same
content, and that either of them implies the other.. If, then, we
have grounds to suspect that St. Anselm’s notion may conceivably
be self-contradictory, will not the same objection apply to St.
Thomas’s? The answer to this is that St. Thomas’s arguments
are not ontological.” They claim,not from the mere content of the
es
| idea, but on quite other and extraneous grounds, to show that a
being corresponding to the idea does in fact exist, and therefore,
|by implication, that the idea is not self-contradictory. If he were
“not conscious of the necessity of this indirect approach, St. Thomas
could base an ontological argument for the existence of God on his
notion of God just as well as St. Anselm could on his. For no less
than St. Anselm will he admit that God’s existence is a mere
accident. ““A thing, =|* he writes,
‘can be nelf-deidenw in either_of two ways; on the ofehand,
self-evident in itself, though not to us; on the other, self-évident in
itself, and to~us. A preposition is self-evident. because the
predicate iis included in the essence of the subject, as ‘Man is an
animal,’ for animal is contained in the essence of man. If, there-
fore, the essence of the predicate and subject be known to all, the
proposition will be self-evident to all ... If, however, there are
some towhom the essence of the predicate and subject is unknown,
the proposition will be self-evident in itself, but not to those who
do not know the meaning of the predicate and subject of the
proposition.... Therefore I say that this proposition, ‘God
exists,’ of itself is self-evident, for the predicate is the same as the
subject; because God is his own existence as will be hereafter
shown. Now because we do not know the essence of God, the
proposition is not self-evident to us; but needs to be demonstrated
by things that are more known to us, though less known in their
nature—namely, by effects.’’+
,
Tn short, if we knew God in his essence, we should see that his
existence was necessary, but if we did so know him we should not
need to prove it. We may therefore agree with Professor Broad
when he writes:
1§. Theol., I, ii, te.
THE TRADITIONAL APPROACH 37
St. homas Aquinas, I think, would have held that it is neces-
sary fake
there something
is that exists, but that only God
Gc cor angels
can see thenecessity “of this fact. “Men can see only that the
agp, Tesis a necessary consequence of certain facts which,
so far as we can see, are contingent, e.g., the fact that there is
motion and qualitative change.’”
The arguments that St. Thomas Linsclf puts forward for the
existence of God are the famous Quinque Viae, which are stated in
the third article of the second question of the First Book of the
Summa Theologica. They are, in a general way, based upon
Aristotle, though they are completely transformed by the con-
ception of God in terms of being, which we have seen is entirely
due to Christianity. They are all of them arguments to the
existence of God from the existence and nature of his effects;
thereforé their scope is considerably limited. Sensible things, says
the Angelic Doctor,
“cannot lead our intellect to see in them what God is, because
they are effects unequal to the power of their cause. And yet our
intellect is led by sensibles to the divine knowledge so as to know
about God that he is, and other such truths, which need to be
cribed to the first principle. Accordingly some divine truths |
are attainable by human reason, while others altogether surpass /
ower of human reason.”’? J
And, so far from being remote and “highbrow” arguments,
they are, as Fr. Garrigou-Lagrange has pointed out, nothing more
than philosophical refinements of one broad general proof that is
used, largely unreflectively, by quite untrained people. He writes
as follows :
“The principle of this general proof, ‘The greater cannot arise
from_the less,’ condenses in effect_into_one single formula the
principles on which our five typical, proofs rest: becoming can
emerge only “from determinate ‘being ; caused being only from
uncaused being; the contingent only from thé necessary; the
imperfect,composite and multiple only from the perfect, simple
and one; order only from an intelligence. The principles of the
first three proofs |place in relief especially the dependence of the
world upon a cause, the principles of the last two insist on the
. superiority and ogra of this cause; all of them can be summed
up in this formula: ‘The greater does not arise from the less;
aon only the higher explains the lower.’
1 Examination of McTaggart’s Philosophy, I, p. 23. 2 .¢.G., I, ili.
38 HE WHO IS
‘This general proof needs to be made scientifically precise by
the five others; but, while by itself it remains rather confused, it
possesses the strength of all the others put together. In it we see
realized what theologians teach about natural knowledge of God.
‘Although the existence of God needs demonstrating,’ says
Scheeben (Dogmatics, II, 29), ‘it does not follow that its certitude
is only the result of a reflective and conscious scientific proof,
based on our own investigations or the teaching of someone else,
nor that its certitude depends on the scientific perfection of the
proof. On the contrary, the proof which is necessary for every
man in order to attain full certitude is so easy and so clear that
one hardly notices the logical procedure which it implies, and that
the scientifically developed proofs, so far from giving man his
first certitude of the existence of God, only illuminate and con-
solidate that which already exists. Moreover, since the proof, in
its original form, presents itself as a sort of demonstration ad oculos
and finds an echo in the deepest recesses of the rational nature of
man, it provides the foundation of a conviction stronger and more
unshakable than any conviction artificially obtained, and it can-
not be shaken by any scientific objection.’ ’’+
Garrigou-Lagrange goes on to expound this proof as follows:
“We know beings and facts of different orders: an inanimate
physical order (minerals), an order of vegetative life (plants), an
order of sensitive life (animals), an order of intellectual and moral
life (man). All these beings come into existence and afterwards
disappear, they are born and die, their activity has a beginning
and an end; thus, they do.not-exist_of themselves. What is their
Gals? s
“If there are beings to-day, then there must always have been
something; ‘if for one single moment, there is nothing, then there
will be nothing for evermore’ (ex nihilo nihil fit, non-existence
cannot be the reason or the cause of being; this is the principle of
causality). And it makes no difference whether the series of
corruptible beings had or had not a beginning; if it is eternal, it is
eternally insufficient: the corruptible beings of the past were as
indigent as those of to-day and were no more self-sufficient than
they. How could any one of them, which cannot even explain
itself, explain those that come after it? That would be to make
the greater arise from the less. There must therefore be, above
corruptible beings, a First Being which owes its existence to it-
self alone and which can give existence to others.’’?
- 1 Dieu, son existence et sa nature, E.T., 1, p. 252 f. The English translation of this work
is rather free, so I have given, in all extracts from it, a literal translation of my own.
The references are to the English edition, 2 Thid.
THE TRADITIONAL APPROACH 39
It is themsuccessively argued that because the world contains living
beings, the First Being must have life; because there are intelligent —
beings, It must be intelligent; because there are principles of
reason, It must be a first and immutable truth; because there are
morality, justice, charity and holiness, It must be moral, just, good
and holy Itself. Thus, _Garrigou-Lagrange _ concludes, by the «
principle that “the greater cannot arise from the less” “Thich isj \
/ only one form of the principleofcausality). wehave demonstrated, |»
/ starting from the world of our experience, that there is aa First ge
\ Being who is at the same time supreme Life, Intelligence, and Ms
\ ruth, perfect Justice and Holiness, and sovereign Good. —
CHAPTER V
THE TRADITIONAL APPROACH
(2) THe Five Ways
T must be admitted that St. Thomas himself does not explicitly
state the very general argument which has been outlinedin the
last chapter, or claim that his Five Ways are only different forms of
one more ultimate proof. On the other hand, there is one of the
_ Ways to which he obviously ascribes very special importance, so
much so that in the Summa contra Gentiles he lays almost exclusive
emphasis on it. This is the Kinetological Argument, or Argument
from_ Motion, which he collects out of Books V to VIII Of the
Physics 0of Aristotle: it occurs in substance also in Books XI and
XII of the Melanie. In his later work, the Summa Theologica,
St. Thomas stated the Argument from Motion, not in practical
isolation, but merely as the first of the Five Ways, and moreover
gave it a mich ’shorter-and less closely Aristotelian form.t This
may show some change in the Angelic Doctor’s estimate of the
argument, but there are two other factors that may have been
responsible for it. The first is that the Contra Gentiles sets out to be
a primarily philosophical work, whereas the Summa Theologica, as
its name implies, is primarily theological; the setond iis that the
Contra Gentiles was written to refute the teaching of the Muslim
| philosophers Averroes and Avicenna and hence naturally meets
‘them on their own Aristotelian ground.
In its full form the Argument cae Motion_is of very consider-
able complexity. It can be stated in either a direct or an indirect
way and is highly technical. It will perhaps be sufficient to give
the main points of the former in the text, and to relegate a sum-
mary of the full argument to a footnote.* The starting-point
of the
1 Cf. J. Rickaby, Studies in God and his Creatures, p. 33 f. A shorter form is also given
in the Compendium Theologiae, 1, iti and iv (qu. by Wicksteed, Dogma and Philosophy,
p. 281 f.), but it is not quite so compressed as in the Summa Theologica.
2 The Argument.from-Motion-is-stated by St. Thomas in ch. xiii of the First Book
of the Summa contra Gentiles. A full discussion is given in Gilson’s Philosophy of St. Thomas
Aquinas, ch. iv.
N.B.—‘ Mov yement”” means any kind of change.
“Move”? is always a transitive verb.
« “Being moved”’ does not necessarily mean “being moved by something else,”’
-“<~~but simply “being in motion.’
40
THE TRADITIONAL APPROACH 41
argumerit is the existence in the world of our experience of what
is called “motion,” by which is meant not merely change of
position in space but any kind of
change. Basing his thought upon
the Aristotelian distinction between potentiality and act, SUvayis
and évépyeia, the Angelic Doctor asserts the truth of two proposi-
tions. The first is that everything which is in motion must be put
in motion by something else, since change means therealizationin
actuality by abeing
of some property which previously existed in
it only in a potential state, and the being cannot itself possess the
power of actualizing this potentiality or it would have done so from
the start. The s€condis that we cannot go back to infinity in this
series of beings which are in motion and beings which move them.
If A is moved by B, and B by C, and C by D, and so on, somewhere
we must come to a being—let us call it Z—which is the first in the
series, and which both causes the movement of all the others and
is itself unmoved and so requires no antecedent being to move it.
“Tt is necessary,” writes St. Thomas, “to arrive at a First Mover, _
I. Drrecr Proor.
From two propositions:
1. Everything set in motion is so moved by something else.
Three proofs of this:
iFor a thing to be in motion of itself:
a. It must have the principle of its movement in itself.
b. It must be moved primarily, i.e. by its whole and not by any
part.
c. It must be divisible and have parts.
Now, since it is moved primarily, if one part were at rest the
whole would be at rest. For if one part were moved while
another were at rest, not the whole but a part would be moved
primarily. But if the quiescence of the whole depends on the
quiescence of a part, the movement of the whole would depend
on the movement of the part. Hence the whole is not moved
by itself. And this is a contradiction.
ii. By induction. Things are moved either accidentally or fer se.
a. If accidentally, they are not moved by themselves.
b. If per se, then either:
a. By force, and so not by themselves. Or
B. By nature, and so either
aa. By something in them (as animals). Or
bb. By something not in them (as heavy bodies).
Ergo.
iii. Nothing is at the same time and in the same respect both in act
and in potency.
2. It is impossible to proceed to infinity in the series of movers and things moved.
Three proofs of this:
i. All the members would move at the same time as any one of
them. But one, being finite, is moved in a finite time. There-
fore, so are the infinite number of them. Which is impossible.
ii. There would be no first mover, and so nothing would be moved,
iii. All would be instrumental movers. Which is impossible.
From 1 and 2 the existence of an immobile First Mover follows,
42 HE WHO I8&
put into motion by no other ; and this everyone unidirstands to be
God.’’!_
It is interesting to note that, while to the Middle Ages this
argument seemed to be the strongest of the five, to the modern
mind it is perhaps the least convincing. This may be due to the
habit of thought engendered by modern science, of assuming that
change, especially change of local position (which is what the
word ‘‘motion”’ most readily suggests), is something ultimate
which we can take for granted without demanding an explanation.
The fact that, whereas the pre-Newtonian astronomers felt the
necessity of postulating some force to keep the planets moving at
all, Newton stated as his First Law of Motion that any body
ees
continues in astate of rest or ofuniformmotion in a straight line except
in so far as it is acted upon by a force, is symptomatic of this.
And the tendency has gone so far that, in the present century, it is
| possible for philosophers like Bergson and Whitehead to conceive
=x .-of change as fundamentally inherent in the category of existence. .
II. Inprrect Proor.
Suppose ‘Whatever moves is moved” is true. It is then true either fer accidens
or per sé.
1. Suppose it is true fer accidens.
Then: a
i. ““Whatever moves is moved”? is true. But
“Nothing that moves is moved”’ is possible.
Therefore “There is no motion at all” is possible.
But it is impossible that there should ever be no motion at all.
Therefore the proposition is not true fer accidens.
me Alternatively :
ii.
If moving and being moved occur together per accidens,
And, as we know, being moved sometimes occurs without moving ;
Then probably, moving sometimes occurs without being moved.
Thus the proposition is not true per accidens.
2. Suppose it is true per se.
Then either:
i. A mover receives a movement of the same kind as it imparts,
which is contrary to experience. Or
ii. It receives a different kind, and either
a. We get an infinite regress, which is impossible, since the
number of different kinds of movement is clearly finite. Or
b. After a finite number of steps we get to the same kind, and
so the case is reduced to case i.
Thus the proposition is not true fer se.
Therefore it is not true at all.
me we have only proved that the First Mover is unmoved by anything outside
itsel
However: if it moved itself, then either:
i. The whole moves the whole, and it is in potentiality and in act at the same
time and in the same respect, which is impossible. Or
ii. Part of it moves another part, and so again we have an immobile First Mover.
Therefore: there exists an absolutely immobile First Mover.
L$. Theol., 1, ii, 3¢.
THE TRADITIONAL APPROACH 43
Largely as a result of this, though perhaps for other reasons as well,
the Aristotelian notions of potentiality and act have become quite
unfamiliar to the modern mind, as indeed has the whole process
of metaphysical argument. It must, in consequence, be empha-
sized that we are concerned here with metaphysical and not with
purely physical and mathematical concepts. ‘‘ Physics, beware
metaphysics’’ may have been a very necessary warning in the days
of Newton, but ‘‘ Metaphysics, beware physics”’ is not less neces-
sary to-day. And, in spite of the difficulties to which it gives rise,
the distinction of potentiality and act, or at least some modern
equivalent for it, seems to be unavoidable if we are to discuss the
extremely mysterious fact of change without subtly substituting
something else for it. For unless we are prepared to say that, if
X changes into Y, Y was potentially in X before the change, we
shall not be recognizing that X has changed at all. We shall,
instead, be assuming that X has been annihilated and that Y has
been created to take its place, and we shall be substituting for the
rich complexity of a universe which, with all its processes of
generation and corruption, of life and death, persists through time
a succession of discrete states without any real continuity.
But it
must be admitted that, like most of the concepts that Thomism
has adopted, the notions of potentiality and act do not ultimately “
explain. anything; they do, however, make it certain that we are
stating the problem that really has to be solved and are not
substituting for it an easier problem having no reference to concrete
_ reality.
/ Nevertheless, we must be prepared to recognize that, if the
fe world finds the notions of potentiality and act difficult to |
as similate, the First Way of St. Thomas will make less appeal to it _/
KCthan some of the others will, That is only to be expected.
‘Something must now be said about the sécond proposition which
St. Thomas asserts, namely, that it is impossible to go to.infinity
in the succession of movers and moved, and that. we.must finally.
arrive at an unmoved First Mover. If this is true, it might be
objected, we only arrive at a first mover which is itself a member
of the series, and therefore is nothing like the Christian idea of
God. It would bear the same relation to the other members of
the series as the integer ‘‘one’’ bears to the succeeding members
of the series of integers, “two,” “‘three”’ and the rest. Its status
is essentially the same as theirs, except that it happens to have no
predecessor. It would appear that here St. Thomas has followed
44 HE WHO IS
Aristotle in his wording of the argument in a way that does not
really express his own convictions, for the whole essence of his
position is that God is of an entirely different nature from all
other beings, that he belongs to an infinitely higher order of reality.
Garrigou-Lagrange’s remark, which we have already quoted, to
the effect that an eternal series of corruptible beings would be
eternally insufficient, suggests the clarification that is necessary,
and indeed he develops it explicitly in his discussion of the First
jee The ae is not really that we cannot have an infinite
r egress in the order ofnature,
but that such an infinite regress in
the series of moved movers would necessitate an unmoved First
Mover not inthe order of n
nature but above sees_Garrigou-Lagrange
(nites as follows?
“The second proposition, ‘We cannot go to infinity in the series
of movers which are actually and essentially subordinate,’ rests on
the very notion of causality and not on the impossibility of an
infinite and innumerable multitude. With Aristotle, St. Thomas,
Leibniz and Kant, we do not see that an infinite series of accident-
ally subordinate movers in the past would be contradictory. ...
What is repugnant is that a movement which exists in fact should
have its sufficient reason, its actualizing raison d’étre, in a series of
movers which have only the status of moved movers: if all the movers
receive the influx which they transmit, if there is no first one which
gives without receiving, the movement could never take Bice for
there would never be a cause for it.
“But there can be no need to come 6 a halt in the series of past
movers, since they exert no influence on the actual movement
which as to be explained; they are causes fer accidens. The
principle of sufficient reason (principe de raison d’étre) does not
oblige us to terminate this series of accidental causes, but only to
leave it, to rise up to a mover of another order, not itself moved,
and in Wd sense motionless, not with the Emaaobility of potentiality
which is anterior to movement but with the immobility of the
act which has no need to become because it already is (immotus
in se permanens).’*4
And, as is well known, St. Thomas caused something of a scandal
in his time by maintaining against contemporaries of the eminence
of St. Bonaventure that it is impossible to prove by reason alone
that the world had a beginning in time.’
A further point that calls for comment is St. Thomas’s identifica-
tion of the Prime Mover with God. ‘This everyone understands
1 Dieu, E.T., I, p. 264 f. 2 Cf. S. Theol., I, xlvi, 2c; S.c.G., I], xxxviii.
THE TRADITIONAL APPROACH 45
to be God.” But has he any right to? As this same point arises in
connection with each of the-Five._Ways,it will be convenient to
defer it until our outline of them is complete.
The Se ; Causality, or, as it
is sometimes galled, the Aetiglogical Argument. In the Summa
Theologica it is given equal pfominence with the <argument from
Motion ; in the Contra Gentiles it forms a kind of appendage to it.
It is important to notice that the idea of efficient causality is some-
what different from what is suggested by the words ‘“‘cause”’ and
“causation”’ as they are used in modern physical science. —It_is__
sometimes ae thatmodern science has no use for efficient
within its “Sphere se interest, that assertion may be admitted, at
any rate as regards the sciences of physics and chemistry Gesis
rather less certain in the cases of biology and psychology). But
this is quite different from admitting that efficient causality is an
unnecessary or discredited notion; the fact is that the method of
investigation of the world which physical science adopts—observa-
tion of measurable phenomena and their correlation and predic-
tion by general statements—is such as to exclude efficient causality
-from its purview, and hence renders it quite incompetent to decide
whether there is efficient causality or not. Efficient causality 1is
not_a physical concept but a_metaphysical one, and it is only
because the physical scientists of the eighteenth and nineteenth
centuries insisted on illicitly talking physics in terms of efficient
causality that their successors, having discovered that efficient
causality is not what physics is as a matter of fact concerned with,
have only too.often assumed that it is non-existent.
~ The notion of efficient causality is inevitable as soon as we
“inquirerenot merely How things happen, but why. Events.are caused
~by other events, things have effects upon one another. And the
Second Way proceeds by two “steps. which are closely similar _to
those inthe FirstsTt begins fr
from the_assertion that in the world
there is an order of efficientt causality, an interrelation, sometimes
simple, sometimes very complex, of Causes and effects. And it then=,
argues that we cannot proceed to infinity in the sequence of
efficient causes; there must be a First Cause, which is_itself
uncaused_but_is, directly or indirectly, the cause of everything
except itself. And to this, concludes St. Thomas, ‘everyone gives
the name of God.”
46 HE WHO IS
All that was said in discussing the question of an infinite regress
in connection with the_First Way will, with obvious verbal
changes, apply here. But one or two remarks may be added. We
are, it must be noted, not primarily concerned with causes in fier,
but with causes in esse, that is, not with the causes which bring
things into existence but with those that keep them in being. It
must be stressed that a cause is needed for the continued existence
Pa a being just as much as for its original production. If the
/question Whyndid it begin?” needs answering, so does the
-
J
f
question ‘‘ Why.does.it.go.on?”’, and just as it would be maintained
“that unless something had produced it it could not have begun to
be, so also it must be maintained that unless something was
preserving it it would collapse into non-existence. It was the
neglect of this consideration that led the eighteenth-century deists
to be satisfied with the idea of a God who had created the world
at some date in the remote ipa but had been since then nothing
a chain ofcauses stretching rae into the past,but about a chain
of causésexistiiig it the present and each depending on the one
_beyond: And, just as with the First Mover, so here with the First
“Efficient Gove we are not arguing to a first member of the series
having the same status as all succeeding members except for the
fact of being the first; if the number of beings in the world be
admitted to be infinite (though this isa” suppositionthat St.
Thomas ‘would have been reluctant to concede’), so that the causal
chain might have an infinite number of members, it will still be
nécessary to postulate a Cause outside the series for the existence
of the series itself.2 For an infinite number of insufficient beings
will still be insufficient; you cannot get sufficiency out of insuffi-
ciency by multiplying by infinity. The argument in its essence is
simply this: that in the world we find any number of causés of
things, but. they all demand causes for themselves. We must
therefore either give up philosophizing altogether or admit the
existence of a Cause which does not require a cause for itself. And
to this, in the words of St. Thomas, “everyone gives the name of
God.”
The Third Way.‘is frequently called the Cosmological Argu-
ment, thoughthat name may quite reasonably be appliéd-to-all
1 Cf. S$. Theol., I, vii, 4c.
2 St. Thomas himself makes this point. Cf. S. Theol., I, xlvi, 2 ad 7.
? THE TRADITIONAL APPROACH 47
the five or, at any rate, to the first three, since they all start from
a consideration of the cosmos, the world of which we ourselves are
part. In the Summa-Theotogicait immediately follows the Aetio- _
logical Argument; in the Contra Genitles it does not appear in the
chapter-on the existence of God, but, in a simpler and more
satisfactory form than in the Summa Theologica, in the chapters
“That God is eternal’’ (I, xv) and “That God is to all things the
cause of being” (II, xv). It holds a similar position in the
Compendium Theologiae in the chapter “‘Quod Deum esse per se est
necessarium”’ (I, vi), where it occurs, as an alternative to an argu-
ment from God’s immutability, to prove that God’s being is
necessary. It is generally called the argument e contingentia mundu.
Its general form is very similar to that of the preceding two, but it
raises some special questions.
It first establishes as a preliminary thesis the existence of some
necessary being or beings. In the_world, it asserts, there are
clearly many beings whose existence is possible but not necessary.
This is shown by the fact that they are subjecttogeneration and
corruption, they come into existence and then pass away, so there
cannot be in the nature of things any necessity for them to exist;
of any one of them it is true that, since at one moment it exists and
at another it does not, it can either be or not be. Now, it is argued,
it is impossible for the totality of existent being to be of this type.
For, if it were, we should merely have to go back far enough in
time to come to a moment when nothing existed. But, if at any
moment there had existed absolutely nothing, it is clear that
nothing could have subsequently come into existence, and there-
fore nothing would exist to-day. This would contradict the fact
of experience from which we began. Therefore the hypothesis
which we assumed must be false. That is to say, at any moment
/Athe existence of some being or beings is necessary.*
~~T The exact nature of the argument in the Third Way is to some degree a matter-of
uncertainty. I assume that the true reading of a crucial sentence in the proof as stated
in S. Theol., 1, ii, 3, is Impossibile est autem omnia quae sunt, talia (sc. possibilia esse et non
esse] esse, as in the best uncials, and not Impossibile est autem omnia quae sunt talia, semper
esse, as in most of the printed editions (I am indebted to Fr. Victor White, O.P., for this
information). The latter reading involves a complete non-sequitur; Fr. Garrigou- -
Lagrange is reduced to mistranslating it by “Sil n’y a que des étres contingents, il est
impossible qu’ils existent depuis toujours” (Dieu, p. 270; cf. E.T., I, pp. 293-4). Gilson
also takes the second reading when he writes, “‘J/ est impossible que toutes les choses de ce
genre existent toujours” (Le Thomisme, p. 82); the E.T. ignores the toutes and reads, “It is
impossible that things of this kind should always exist” (Phil. of St. Thomas Aquinas,
p. 85). Similarly, Dr. R. L. Patterson ignores the word omnia, when he writes, “‘It is
impossible that such things should always exist.’? (Conception of God in the Philosophy of
Aquinas, p. 51.) |
48 HE WHO Is _
Having established this fact, the argument proceeds as follows.
Such a necessary being either has its necessity caused by another
or it has not. Now, if the former alternative holds, we shall have
an infinite regress similar to those previously considered. There-
fore, there must be some necessary thing whose existence is not
caused by another, but which, while it causes the necessity of
others, has its own necessity in itself; that is, in other words, some-
thing whose existence is involved in its essence. And “this all men
speak of as God.”
The presentation of the argument in terms of time is not really
necessary to it; in the form given in the Contra Gentiles and the
Compendium Theologiae no reference to time occurs and the two
steps are telescoped into one. Dr. R. P. Phillips states it in this
shorter form, as follows:
“Tt is clear that any existing being which can cease to exist does
not contain in itself the reason of its own existence, and must
therefore derive its reason of being from something else; and, in
the long run, from a being which exists of itself; for we-cannot
It must further be noted that the argument does not assert that, if every individual
existent comes into being and passes away, there must have been a moment when none
of them existed. If we admit the possibility of an actually infinite manifold of things
(though, as we have seen, St. Thomas would have been reluctant to do this; ef.
S. Theol., I, vii, 4), and suppose that No. 1 has existed for one year, No.2 for two years
‘and generally No. n for n years, then each of them will have had a beginning and yet
there will never have been a moment when none of them existed. What is asserted is
that, if existent-being-as-a-whole came into being and passed away, there must have
been a moment when nothing existed; this will still be true if we suppose existent-
being-as-a-whole to have different compositions at different times. The upshot is thus
that, at any moment, the existence of something is necessary, and the argument goes on
from this.
It is also important to notice the force of the second step in the argument. Having
decided that at any moment the existence of some thing is necessary, we want to know
whether this particular thing exists just because something must exist and this happens
to be the thing that does, or whether it exists because its existence is intrinsically
necessary. The former alternative, by the regress-argument, involves the existence of
some other being whose existence is intrinsically necessary; the second alternative
furnishes such a being at the start. Therefore, in any case an intrinsically necessary
being must exist.
Professor A. E. Taylor, in his long article on ““Theism” in £.R.E., Vol. XII, gives
a rather different interpretation of the argument. Presumably taking the second of the
above readings of the text, he apparently falls into the fallacy mentioned in the second
paragraph of this footnote and asserts as the conclusion of the first step of the argument
that ‘“There must be ‘something in things’ which is necessary, i.e. INCAPABLE OF NOT
EXISTING”? (op. cit., p. 270, n. 2, small capitals not in original). He then takes the
second step as serving simply to prove that there is one ultimately necessary being.
It may be added that the question as to the possibility of an actually infinite manifold
is not the out-of-date medieval subtlety that it might appear. The theory of infinite
number as developed by Mr. Bertrand Russell depends upon the axiom that the
number of things in the world is infinite (cf. Intro. to Math. Phil., ch. xiii), an axiom
which logicians are by no means unanimous in accepting. (Cf. M. Black, Nature of
Mathematics, p. 104 f.)
os THE TRADITIONAL APPROACH 49
proceed to infinity in a series of beings which derive theirreason__ -
of being from some other. To suppose that some contingent
being, or the series of such beings, is eternal, does not in any way
account for their existence, or relieve us of the necessity of
demanding a necessary beingas_the_ cause of such eternal
existence. Even iftheseries is eternal, itiseternally insufficient.”
Before we go on to» consider the last two of the Five Ways, it
may be as well to see just where we have got bets ae i Jie
similarity of the three arguments
a ey EY which we have considered
Senate : ra,
e evident. Each of the om. some1echaracteristic es *
beingg such “as Wwe experience to the. existence of a First_Being a A,
. possessing a corresponding. ‘attribute. The first.“one argues from ~~
“the-occurrence of motion (i.e. change) to a First Unmoved Mover,
the sec m the dependence of finite things to a First Cause,
the t m their contingency to a First Necessary Being.
Furthermo é, each of them establishes its case by denying the
possibility of an infinite. regress. But, however-similar, they” arénot
identical, for they concentrate upon different elements in finite
being. Dr. Phillips, in the manual from which we have just
quoted, remarks that “though both the first and second ways
argue from causation, the first applies it to the transition from
potency to act in the becoming of things, while the second applies
it to the being of things which demands a cause for its preservation
and continuance.” Again, while “the second way starts from the
observation of causes which continuously bestow on things a cer-
tain kind of being,” the third starts from ‘‘the observation of
things which do not possess being of any kind of themselves, and
so must receive being simply speaking, and not only a particular
‘kind of being.” Thus, he says, the third way “‘has a wider basis
than the second, for it starts with all finite being, not only with
those that are causes fer se. Similarly, the basis of this second way
is wider than that of the first, which deals only with causes
secundum fieri, while the second deals also with causes secundum
C580,2
{ Modern Thomistic Philosophy, 11, p. 285. Dr. Phillips goes on to state, as a subsidiary
considération, the argument in the form in which we have rejected it in the last note.
He assumes the usual reading of the disputed sentence, and then argues that if all
beings individually came into being and passed away, there must have been a moment
when none of them existed. This involves a transition from the distributive to the
collective sense of omnia and is a subtle form of the fallacy of composition. Neither he
nor Garrigou-Lagrange nor Gilson apparently sees the flaw which the argument
contains when it is stated in this way.
2 Modern Thomistic Philosophy, II, pp. 294, 295.
50 HE WHO IS
~ One difficulty which will be felt by many to the form in which
the arguments are stated is that the notion of causation as a simple
linear sequence is very much over-simplified. Even in the Middle
Ages it must have been obvious that no event ever has only one
efficient cause, and the view of the universe that modern science
has disclosed to us shows it as a vast system of interconnected
operations and entities in which linear causal sequences are rarely
to be discerned and, when they can be disentangled, are abstract
and rough approximations to the real causal nexus. The criticism
may be admitted, but it does not really affect the argument.
Indeed, one may suspect that St. Thomas—and perhaps also
Aristotle—was merely presenting a simplified picture for the sake
of clearness, leaving his readers to see that this simplification did
not make any substantial difference to the argument. For a net-
work, however complicated it may be, of entities which are in
themselves insufficient is just as insufficient as an infinite linear
sequence of them would be; and just as in this latter case, suffi-
ciency cannot be obtained from insufficiency by adding complexity
to it. Professor A. E. Taylor has expounded this point very
adequately in a well-known essay. He writes as follows:
‘*T can explain the point best, perhaps, by an absurdly simpli-
fied example. Let us suppose that Nature consists of just four
constituents, A, B, C, D. We are supposed to ‘explain’ the
behaviour of A by the structure of B, C, and D, and the inter-
action of B, C, and D with A, and similarly with each of the
other three constituents. Obviously enough, with a set of
‘general laws’ of some kind we can ‘explain’ why A behaves as
it does, if we know all about its structure and the structures of
B, CG, and D. But it still remains entirely unexplained why A
should be there at all, or why, if it is there, it should have B, QC,
and.D as its neighbours rather than others with a totally different
structure of their own. That this is so has to be accepted as a
‘brute’ fact which is not explained nor yet self-explanatory.
Thus no amount of knowledge of ‘natural laws’ will explain
the present actual state of Nature unless we also assume it as a
brute fact that the distribution of ‘matter’ and ‘energy’ (or
whatever else we take as the ultimates of our system of physics)
a hundred millions of years ago was such and such.... As M.
Meyerson _putsit, we only get rid of the ‘inexplicable’ at one point
at the price of introducing it again somewhere else. Now any
attempt to treat the complex of facts we call Nature as something
which will be found to be more nearly self-explanatory the more
~ THE TRADITIONAL APPROACH 5t
of them we know, and would become quite self-explanatory if we
only knew them all, amounts to an attempt to eliminate ‘bare
fact’ altogether, and reduce Nature simply to a complex of ‘laws.’
n other words, it is an attempt to. manufacture _particular
( existents out of mere un universals, and. therefore must_end_in
ure. 21
It must be admitted that not all modern scientists would agree
to Professor Taylor’s further remark that ‘‘the actual progress of
science bears witness to this.” Professor Eddington, in particular,
has made a very valiant effort to demonstrate that the whole
physical universe, including such apparently contingent charac-
teristics as the number of ultimate particles that it contains, is an
inevitable outcome of the way in which the human mind is bound
to arrange and classify any impacts that it receives through the
medium of sensory stimulation. He has expounded this philosophy
of “selective subjectivism” in his Tarner Lectures on The Philo-
sophy of Physical Science. But it may be doubted whether what he
presents as a demonstration of the entire inevitability of the
physical universe really amounts to more than an inverted
exposition, illustrated in the most fascinating way from the extra-
' ordinarily beautiful theories of modern mathematical physics, of
the Aristotelian doctrine of the intelligibility of being. And it may
be seriously doubted whether, as a matter of fact, he has succeeded
in eliminating the factor of “‘brute fact” from the universe. The
ordinary reader of the work just mentioned receives from it some-
what the same impression of baffled but unconvinced mystification
that is produced by the Ontological Argument, with the difference:
that, whereas St. Anselm produces an apparently previously non-
existent rabbit out of an empty hat, Professor Eddington uses his
hat in order to reduce a very substantial rabbit to non-existence.
And a very acute critic of the lectures referred to has pointed out
that there is, in fact, a considerable element of the purely empirical
in Eddington’s treatment which is obscured by an equivocal use
of terms and by a pseudo-Kantian epistemology for which there
seems no sound basis.?
1 “The Vindication of Religion,” in Essays Catholic and Critical, Pp. 53-
2 E. F. Caldin on “Eddington, Physics and Philosophy” in Blackfriars, March 1940.
_ Professor Whitehead remarks that ‘“‘the four dimensions of the spatio-temporal :
: eeometrical axioms, even’ the mere dimensional character of the
continuum—apart from the particular number of dimensions—and | the fact of
i measurability .. . are additional to the more basic fact of extensiveness.” (Process and
Reality, p. 127; ‘cf. Part II, ch. iti, sec. iv; and Modes of Thought, p. 212.) See the
criticisms of Eddington’s earlier expositions (which he does not answer in The Philosophy
E
52 HE WHO Is
We will therefore pass on to the consideration of St. Rentiae S
Fourth Way, leaving one or two further important points raised
by the first Three Ways forlater discussion when the exposition of
the Five Ways is complete.
- The Fourthh Wayiis the Argument from Degrees of Being
or from
ultiplici also known as the Henglogical Argument. In
the Contra “Cmniiles iit is stated with extreme brevity, relying upon
two_places
in Aristotle,.as follows.t Those things which excel as
true excel as beings.- But there is something supremely true,
because, of two false things, one is falser than the other and there-
fore one is truer than the other, and this is by approximation? to
& that which is simply and supremely true. Hence there is some-
~thing that is supremely being. And this we call God.
The presentation in the Summa Theologica is rather more elabor-
ate. Among beings, some are more and some are less good, true,
noble, and so on. But more and less have a meaning only in so
far as things approximate in the quality under consideration to _
that which possesses the quality in the supreme degree, which “‘has
all of it that there is”, as we might say. And this being that
possesses the quality in the supreme degree must be the cause of
its occurrence in other beings in lesser degrees ; as for instance fire,
which is the supreme degree of heat, causes all other heat. ‘There-
ke concludes St. Thomas, “‘there must also be something which
is to all beings the cause
of their being, goodness, and every
€ other
\_ perfection :and this we-call-God-"s
Several comments seem to bé Called for. In the first place, the
rather trivial illustration from heat and fire must not oetaken too
seriously. If it was intended to be more than a remote parallel,
St. Thomas would presumably conclude that God was not only
perfectly true, good and the rest, but also perfectly hot. It is only
to such qualities as truth, goodness and nobility, which of their
nature demand something directly akin to them as their cause
that the argument applies, in distinction from those impure perfec-
of Physical Science) by Professor L. Susan Stebbing in Philosophy and the Physicists,
especially ch. iv. See also the important and very critical review of The Philosophy of
Physical Science by R. B. Braithwaite in Mind, Oct. 1940, and Eddington’s reply in
Mind, July 1941. Cf. Professor C. D. Broad’s remark, in criticism of Eddington on this
point, that “no valid argument can derive a singular conclusion from premisses which
are all universal.”’ (Philosophy, July 1940, p. 312.)
1 S.c. G., I, xiii. The references are to Metaphysics II and IV.
2-i.e, § ‘nearness.” The truer thing is the nearer to the supremely true.
8 §.\Theol. I, ii, 3.
- THE TRADITIONAL APPROACH 53
tions whose nature prevents them from applying directly to God,
such as this very instance of heat. Goodness, so the argument
claims, demands as its cause a God who is good; while heat,
though it necessarily demands a God whose knowledge of possible
being includes an idea of heat, does not demand a God who is hot
as its cause, but only a God who can create. And this leads on to
a further observation.
The question is sometimes raised whether the argument, assum-
ing it is valid at all, proves the existence of an absolute good,
truthand the like or only.a. relative-one.- That is to say, does it
prove a good than which no greater good can be conceived, or
only a good than which no greater good exists? And are we not,
in either case, confronted with a difficulty? For if the argument
asserts an absolute maximum, are we not subtly falling into the
Ontological Argument, which we have seen reason to reject? And
if it asserts a relative maximum, is not that merely a finite being
of a very high degree.of perfection and so something less than God?
The second alternative must obviously be rejected if the argument
is an argument for the existence of God, as it claims to be, but it
does not appear that the first alternative is necessarily equivalent
to the Ontological Argument. For that argument claimed to
, deduce the existence of God from the mere concept of him, whereas
the argument now under review does not claim to deduce the
existence of maximum good from the mere concept of goodness.
~On the contrary, its starting-point was the concrete existence of
things which, albeit imperfectly, are good. There is no transition
from the order of concepts to that of existence. The only question
is whether the argument can in fact prove what it claims to;
whether the concrete existence of finite good necessarily implies
the concrete existence of an absolute maximum good. (And
similarly, of course, with the other types of perfection.) It is here
that the addition which the form of the argument in the Summa
Theologica makes to that in the Contra Gentiles is of importance, for
it asserts the existence of a supreme degree of each perfection not
merely as the pattern or exemplar of the lesser degrees, but as their
creative cause. And because it is creative, that is because it is
capable of bestowing being, it must, so it is implied, be self-existent
1 Cf. S. Theol., I, xlv, 7¢: ‘‘Some effects represent only the causality of the cause, but
not its form; as smoke represents fire. Such a representation is called a trace (vestigium).
Other effects represent the cause as regards the similitude of its form, as fire
generated represents fire generating ;and a statue of Mercury represents Mercury ;3 and
this is called the representation of image.”
54 HE WHO IS
~ being itself. It is not merely that the idea of finite perfection
Penile tbeciate TERE the-idea) of infinite
pe
its model,/,but.that the existence of. finite. perfection. implies” the
eXistence ofof infinite perfection as its cause. There will be more to
say about this later on, but we will leave it here for the present.
It is perhaps just worth while to add that, although St. Thomas
does not include beauty among the perfections enumerated, the
argument will obviously apply to it, too. It is therefore not true
to suggest, as is sometimes suggested, that St. ‘Thomas has no place
for esthetic. considerations
in his. Conception of God. The reason
for his omission is far more probably that he did notConsider
beauty as a distinct “perfection, from_thosewhich he. instances.
“The beautiful isthe same as the good,” he says elsewhere, ‘“‘and
they differ iinaspect only. 2 “Beauty,” writes a modern Domini-
can, “is the good known and the known. loved, ” and he quotes
tae Gill to the effect that beauty is “a union of the true and the
good, and the faculty which has beauty for its object is the whole
and undivided mind.”® St. Thomas, then, will agree with St.
Augustine’s acclamation of God as pulchritudo tam antiqua et tam
nova,* as the eternal beauty which is the cause of all beauty in
created things.
The Fifth Way is the Argument from Design, or, as it is also
called, the Teleological Argument, though it is, at least in the
form given to it in the Summa Theologica, very substantially different
from the arguments to which that name has been given since St.
Thomas’s time. It does not start, as Paley did, from the appear-
ance of design manifested in living organisms, nor, as many
present-day writers do, from a similar appearance alleged to be
evident in the course of nature as a whole or in processes, either
organic or inorganic, extending over centuries or millennia. It
simply asserts that, when we are confronted with any being, how-
ever insignificant, the question, ‘What is its purpose?”’ is just as
valid and necessary as the question, ‘‘ What began it?” or “Why
does it go on?” In other words, the very existence of a being, as
an ontological fact, necessitates the existence of a final cause for it
just as much as that of an efficient cause. If there were only
efficient causes-and—no—final_causes, nothing could. come into
1 Cf. the discussion of Professor Broad’s objections, p. 57 infra.
2S. Theol., II 1, xxvii, 1 ad 3; cf. I, v, 4.ad 1.
3 Kenelm Foster, O.P., on “Created Holiness,” in Blackfriars, Feb. 1941,
© Confessions, X, xxvii.
THE TRADITIONAL APPROACH 55
existence, Final and efficient causality are equally involved in the
being. (Incident ally, we may note that the Fourth
Way asserted that formal causality is involved as well.!) Now, in
some cases, as in that of intelligent beings, it might be alleged that
the internal nature of the being itself provided the final cause;
conceivably I myself decide what my actions are for, and so I have
no
need to postulate God as my final cause. But,
at,whether this is
true or not of some beings—and_the admission need_not in any
case be too readily 2made, since we are concerned with the final
se not merely of actions but of being—it certainly is not true of
all “We see,” writes St. Thomas, “that things which lack intelli-
“gence, such as natural bodies, act for an end, and this is evident
from their acting always, or nearly always, in the ’same way, so as
to obtain the best result. Hence it is plain that not fortuitously,
but designedly, do they achieve their end. Now whatever lacks
intelligence cannot move towards an end, unless it be directed by
some being endowed with knowledge and intelligence; as the
arrow is shot to its mark by the archer. Therefore some intelligent
eing exists by whom all natural things are directed to their end; i
and this being we call God.’’?
At is fairly clear that, ifit is taken in isolation, this argument is
is
(
A insufficient to put the existence of God beyond all shadow of doubt.
“The teleological nature of beings in their separate particularity,
the purposiveness.which they show in simply going on existing and
operating as the sort of beings that they are, might conceivably be
accounted for by supposing that each of them was provided with a
spiritual guardian or angel to direct its operations. In the Contra
Gentiles the argument is put in a rather less strictly metaphysical
and superficially more “‘modern”’ form; though this last adjective
might be questioned, since the reasoning is based upon that of the
great eighth-century Greek theologian, St. John of Damascus.
**We see,”’ it tells us, “‘that in the world things of different natures
accord in one order, not seldom and fortuitously, but always or
1 The Aristotelian, distinction of the four,kinds efcause isi thus explained by Dr.
Phillips.:
“There are ... four ways in which the cause may pass into the effect ;for it may he.
thaty which.
v the effect is produced, and we have an efficient cause; or that fox.whose
sake theeffect is een and we have a fitial cause ;or that out of which it is produced,
and wehavea mate Atisé; or that which makes the effect to be of aparticular kind,
and we have a farnaeneee>
“A man building a house is its efficient cause; it is built to afford protection from
the weather, and this is its final cause; it is made of bricks and mortar, its material
cause; and it is a building, and a building of a particular kind, which. is its formal
cause.” (Modern Thomistic Philosophy, II, p. 234.) 2 §. Theol., I, ii, 3
HE WHO IS
C:for the most part. Therefore it follows
whose providence the world is governed. And
that there is someone by
this we call God.”’!
In this form the argument evades the objection we have just
mentioned, but it has its own particular difficulties. For, in the
first place, is a governor of the world’s processes necessarily God?
It is not immediately evident that to direct the world and to be
the source of its being are necessarily properties of the same being.
And, secondly «does the ar ( roblem of evil? What
sort of governor is the world seen to have? We spall: come back
to both these points later.
1 I, xiii. Cf. John Dam., De Fide Orth., i; 3.
CHAPTER VI
THE TRADITIONAL APPROACH
(3) Irs StcNiFIcCANCE AND VALIDITY
EFORE we gather up the results of this discussion of the
Quinque Viae in order to see exactly where it has led us, it
may be well to see what objections_ca can_ be brought.against
them as a whole....Kant, it is “well known, brought against the
» Third Way in particularvalaz/the charge that it made an implicit and
illicit appeal to the Ontological Argument in identifying necessary
ing with ens realissimum. Professor W. R. Sorley-has-pointed out
that Kant’s criticisms were oo directed against theology as
such; “We see,” he writes, “that they aré directed not simply
Pret ythe old forms of argument, but against any possible argu-
ments for a knowledge of the ultimate nature, or of the whole, of
things.” Much discussion has ranged round them, and it need
not be reproduced here. It will perhaps be more profitable to
consider the criticisms recently made by a present-day philosopher,
Professor C. D. Broad, who has stated the case against all forms of
argument from the existence of the world to the existence of God
with consummate clearness. And this task will be all the more
1 Moral Values and the Idea of God, p. 299. A very thorough refutation of the Kanan
objectionsis given by Professor A. EE. Taylor in E-R:E., XII, p. 276 f., s.v. ‘“Theism”
cf., for a more scholastic discussion, H. S. Box, The World and God, ch. xxi, or the very
long treatment by Garrigou-Lagrange in Dieu, E.T., I, p. 6r f. Prebendary R.
Hanson writes as follows: ‘‘Since the time of Kant they [se. the scholastic arguments
for the existence of God] have been very generally held to be invalid, to which it may
be sufficient to reply that on the Kantian theory of knowledge they certainly are, but
not necessarily so on a different theory of knowledge, and not at all so on the Scholastic
and Aristotelian theory of knowledge.” (‘Dogma in Medieval Scholasticism” in
Dogma in History and Thought, p. 103.) Cf. Taylor: “It is one question whether Kant
has proved that the demonstration of theism is impossible on the assumption that the
special doctrine of his Critique as to the limits of human knowledge is true, but quite
another question whether that doctrine is true, and consequently whether Kant has
proved the fallaciousness of natural theology unconditionally.” (art. cit., p. 276.)
Again, Dr. R. L. Patterson, while very critical of St. Thomas in many respects, refuses
to admit the validity of the Kantian objections; he points out that St. Thomas did not
merely assume that necessary being was ens realissimum, but argued at length to prove
this. (Conception of God in the Philosophy of Aquinas, p. 96 f.)
/ It is not always realized that it was only in late middle-age that Kant developed his
hostility to natural theology and that in his earlier works he argued vigorously in
support of it.
- 57
58 HE WHO IS
worth while, because in performing it we shall elucidate more
thoroughly the nature of the argument itself.
_.__Broad.includes under the title ofthe Cosmological .Arguihent all
types of reasoning which assert that because there exist particular
things, persons and events possessing certain characteristics there
must be some necessary being which is the cause of their existence
but requires no cause for its own. His discussion thus covers the -
first four, if not all, of the Thomist ‘‘ Ways.” He agrees that the
coming into existence of certain things and persons and the occur-
rence of certain events are felt by us to need explanation. ‘The
first move,”’ he says,
“is to try toexplain it by reference to previously existing things
or persons... and by reference to earlier events. ... Now this
kind of explanation is, in one respect, never completely satis-
factory.... This is for two Peasons. The firsts that such
explanations always involve a reference to general|laws as well as
to particular things, persons, aand. events. Now the general laws
aré themselves just brute facts, with no trace of self-evidence or
intrinsic necessity about them. The second) and more obvious
reason is the following. The earlier things, persons, and events,
to which you are referred by explanation in terms of ordinary
causation, stand in precisely the same need of explanation as the
thing or person or event which you set out to explain.”
“Tt is alleged,’ Broad continues,
“that we can conclude, from the negative facts already stated,
that there must be a substance which is neither a part of nature
nor nature as a collective whole. And we can conclude that
there is another kind of dependence, which is not the ordinary
dependence of a later state of affairs on an earlier one in accord-
ance with de
facto rules of sequence. The existence of this non-
natural substance must be intrinsically necessary. And the
/existence of all natural events and substances must be dependent
upon the existence of this non-natural substance by this non-
“natural kind of dependence.”
This is, of course, simply the argument of Professor Taylor which
was quoted a few pages back. But here is Broad’s criticism of it.
“The human intellect,” he says,
/‘is completely satisfied with a proposition when either (a) the
/proposition is seen to be intrinsically necessary by direct inspec-
\ tion of its terms, or (d) it is seen to follow by steps, each of which is
THE TRADITIONAL APPROACH 59
seen to be intrinsically necessary, from premisses which are all seen
to be intrinsically necessary. . . . Now it is logically possible that
complete intellectual satisfaction should be obtained about
- natural events and substances if and only if the following
ae conditions were fulfilled. (a) If there were one or more existential
propositions which are intrinsically necessary... . And-(d)yif all
other true existential propositions followed with strict logical
necessity from these, combined, perhaps, with certain intrinsic-
ally necessary universal premisses.... And I admit that, if the
universe is such that this kind of intellectual satisfaction is
theoretically obtainable about nature, then its structure must be
very much as philosophic Theism says that it is.”’
However, there is, according to Broad, a suppressed assumption
in the argument, namely that the universe is such as to give this
kind of intellectual satisfaction. And Broad sees no reason why it
should be.
For, he says, “‘whenever we have this kind of completely satis-
factory insight we are dealing with the formal relations of abstract
entities, such as numbers or propositions, and not with the
existence or the non-formal properties of particulars. There is
no reason whatever to think that this kind of rational insight is
possible in the latter case.”
But this is not the only objection. It has been shown, Broad urges,
that for the argument to hold there must be some intrinsically
necessary existential propositions. But such a proposition would
be of the form, ‘‘ There must be something which has the charac-
teristics x, y, z, etc.,” where this set of characteristics constitutes
the definition or description of a certain possible object. However,
necessary propositions are always of the form. “If anything had 4
the attribute x, it would necessarily have the attribute y,” or “Jf |
p were true, then q would be true,” that is, they are always |
conditional. Hence the conclusion of the argument is not only
unproven but false.
/ Furthermore, even supposing there were an existent or existents_
( whose existence was intrinsically necessary, this would not make) ~}-
nature intelligible in the sense required. on
“The difficulty is as follows. Anything whose existence was a
necessary consequence of its nature would be a timeless existent.
If a certain set of attributes is such that it must belong to some
thing, it is nonsensical to talk of its beginning to belong to some-
60 HE WHO IS
thing at any date, however far back in the past.... Now
nature is composed of things and persons and processes which
begin at certain dates, last for so long, and then cease. But how
could a temporal fact, such as the fact that there began to be a
person having the characteristics of Julius Caesar at a certain
date, follow logically from facts all of which are non-temporal?
Surely it is perfectly obvious that the necessary consequences of
facts which are necessary are themselves necessary, and that the
necessary consequences of facts which have no reference to any
particular time can themselves. have no reference to any
particular time.’”
Let us consider Broad’s objections in succession. The first was
that we have no reason to suppose that the universe is such as to
give a certain kind of intellectual satisfaction, a kind which, we
may observe, he says is reached in pure mathematics and hardly
anywhere else. It is the satisfaction obtained -when a particular
existential proposition is seen to be the necessary logical conse-
quence of existential propositions which are themselves intrinsi-
cally necessary. That is to say, we should obtain this kind of
satisfaction if the proposition, “‘A world having the characteristics
of the world of our experience. exists,’”..was-a-necessary logical
consequence of the proposition, “‘God exists,” supposing that this
latter proposition was intrinsically necessary. And Broad rightly
says that no such necessary logical implication of the one proposi-
tion by the other obtains.
Now we may certainly concede his point. Indeed we may go
further and say that if such a logical implication did hold it would
be exceedingly embarrassing for the Christian, for it would mean
that the existence of the world followed necessarily from the
existence of God ; that, to use the terminology of creation, God was
bound to create a world and, indeed, precisely this world and no
other. A Christian would receive very little intellectual satis-
faction from a demonstration that the existence of the world
followed from the existence of God in the same way that the
proposition that the angles at the base of an isosceles triangle are
equal follows from the axioms and postulates of Euclidean
geometry.” St. Thomas Aquinas, in fact, devotes three chapters
of the Contra Gentiles to arguing that God does not necessarily will
1 F.T.S., Jan. 1939, p. 25 f.
2 It is usual to take this instance as an example of an indisputable logical deduction,
but anyone who has studied the philosophy of mathematics knows how questionable
even this assumption is. Cf. Gonseth, Fondements des Mathématiques, ch: i, ii.
THE TRADITIONAL APPROACH 61
anything other than himself.t_ The_Christian theist gets his satis- \
. faction—if that is the Word to use—not by convincing |himself that -
if there is
is a God there must be aworld, but by convincing himself [Xe
that since there is aworld there mustbe a a God The world. mn 4
ceeds from God, according to Christian theism, not by a logical
necessity,_| jut
by anunnecessitated act of creative will. This may
not give us the same kind of intellectual satisfaction as is given
by a mathematical argument, but Broad has nowhere demons-
trated that no other kind of intellectual satisfaction ispossible. His >
mistake is to have interpreted as an argument in deductive logic
what is in fact an argument in metaphysics.
ok similar reply can be made to his second objection, namely
\that necessary propositions are .always.conditional. In logic, of
course, they always are, because logic is always concerned with
the relations between propositions and never with being as such.
Metaphysics, in contrast, is concerned with the inner constitutive
nature of being. If a necessary being exists, it does so not in
obedience to a logical demonstration, but because its very nature
is such as to maintain it in existence, because, if we may venture
to use the expression, it keeps going under its own ontological
steam. And the argument that if beings exist that are not self-
sufficient they derive their being from a being that 7s self-sufficient
is not merely the deduction of one proposition from another by
the rules of logic; it results from an understanding of the very
nature of non-self-sufficient being. We shall return to this point
when we consider precisely what it is that the arguments profess
to do.
The third ees that temporal facts cannot follow logically
from non-temporal facts, becomes largely irrelevant once .we-have
recognized. that the relation between God and the world-is-not
merely one of logical dependence. But even so, there is a sheer
confusion of language. For facts in the sense of true propositions
(which is the sense adopted by Broad when he speaks of “‘the fact
that there began to be a person having the characteristics of
Julius Caesar at a certain date”) can hardly be described as
temporal at all, though the persons or events which they are about
may be temporal.? The fact that Julius Caesar was born on a
certain date is true in the twentieth century and was true ten
1, Ixxxi-lxxxiii. Cf. S. Theol., I, xix, 3.
2 It is surprising to find Broad falling into this ambiguity, for elsewhere he has
written very convincingly against it. Cf. Examination of McTaggart’s Philosophy,
ch, xxxv.
62 HE WHO IS
thousand years before the great Roman general was born. What
Broad presumably means is that a fact involving temporal beings
or events cannot follow_as a mere logical consequence from a fact
involving only non-temporal ones,,_and this wema sadily admit.
But, as we have said, creation is not an operation of logic but of
will. The Christian sao: has consistent maintained. that
~ création i
whole temporal eroatelRader is maintained it
in1 existence, -and that
the creation-and conservation-of the universe is.one. timeless act.
The whole of the eleventh book of the Confessions of St. Augustine
is really an exposition and a development of the theme that the
timeless and eternal God has posited the world not in tempore but
cum tempore, by bestowing upon it its whole being. St. ‘Thomas
echoes this repeatedly. “‘The duration of God which is eternity
has no parts but is utterly simple.... Wherefore there is no
comparison between the beginning of the whole creature and any
various signate parts of an already existing measure .. . so that
there need be a reason in the agent why he should have produced °
the creature at this particular point of that duration.... But
God brought into being both the creature and time together”?
““Newness of movement is consequent upon the ordinance of the
eternal will to the effect that movement be not always.’? ‘ Crea-
tion in the creature is only a certain relation to the Creator as
to the principle of its -being.”’® "The preservation of things by
God is a continuation of that action whereby. he_gives--existence,
which action is without either motion or time.’ ‘“God’s will to
‘ create the world was eternal ; not that the world should exist from
eternity, but that it should be made when he actually did make
“ge 99
re°°5
We conclude, therefore, that Professor Broad has failed to make
good his charges against the validity of arguments from the
existence of the world to the existence of God. We may, therefore,
go on to consider what, when we look back upon them, we see that
the arguments have actually told us.
In looking back upon the Five Ways of St. Thomas, we must
4
first Observe that each of them takes its ‘starting-point in finite
being, and that, at any rate as regards the first-three of them, any
finite being will do. In contrast with the favourite modern argu-
1 $.¢.G., II, xxxv. 2 9.c.G., II, xxxvi. 3S. Theol., I, xlv, 3¢.
4 $. Theol., I, civ, 1 ad 4. 5 De Pot., I, iii, 17 ad 9.
THE TRADITIONAL APPROACH 63
ment which, in a rather different sense from that in which the
name has been given to St. Thomas’s Fifth Way, is usually known
as the teleological, they do not ask us to consider the boundless
depths of space or the unfathomable ages of past time, they do not ©
argue from any characteristics of the universe as a whole or even
of that part of it which is open to our investigation, but from any
finite being, of whatever type and however humble it may be. It
is not from the properties of this, that or the other finite being that
the argument starts, but from the very nature of finite being itself.
It calls attention to some aspect finitude
of and argues to the
existence of a being with some corresponding aspect of infinitude
as its ground. It is entirely free from the transatlantic snobbery of
size; all finite beings are so poor in comparison with God that the
least of them is as adequate as the greatest to lead us to him. “It
suffices for things to exist for God to become inevitable,” writes
M. Maritain. “‘Accord to a point of moss, to the smallest ant,
the value of their ontological reality, and we cannot escape any
longer from the terrifying hands which made us all.”
os “Two men looked out through prison bars;
The one saw mud, the other stars.”’
But for St. Thomas the mud will form just as firm a basis for belief
in God as the stars will, for, in the words of a saint of later time,
“All created things are but the crumbs which fall from the table
of God.’”?
We may remark here that, while as we have seen, to St. Thomas
the Argument from Motion was of primary importarice, many
later theologians have laid their main emphasis upon the Argu-
ments from Causality and Contingency. Thus in the seventeenth
century Dr. Thomas Jackson argues first from the maxim, “‘ What-
soever hath limits or bounds of being hath some distinct cause or
author of being”’ in order to prove that the existence of the world
implies the existence of a First Cause, and then from the maxim,
“\“ Whatsoever hath no cause of being can have no limits or bounds
of being,” in order to prove that this First Cause is infinite. He
~distinctly states that he is concerned with natural and not temporal
precedence, and argues:
1 The Degrees of Knowledge, E.T., p. 132.
2 St. John of the Cross, Ascent of Mount Carmet, {, vi, 2. Cf. Robert Sanderson:
“From the goodness of the least creature guess we at the excellent goodness of the
great Creator. Ex pede Herculem.” (Sermon at Paul’s Cross, 21st November 1624,
quoted in Anglicanism, p. 227.)
64 HE WHO IS
‘Whether we conceive effects and causes distinctly as they are
in nature, or in gross, so long as we acknowledge them (this or
that way conceived) to be finite and limited, we must acknow-
ledge some cause of their limitation, which (as we suppose) cannot
be distinct from the cause of their being..
“The greatest fulnessof finite_ existence ‘conceivable cannot
reach beyond all “possibility. of non-existence, nor can possibility
of non-existence and perpetual actual existence be indissolubly
wedded in any finite nature, save only by his infinite power who
essentially is, or whose essence is to exist, or to be the inexhaustible
fountain of all being. The necessary supposal or acknowledgment
of such an infinite or essentially existent power cannot more
strongly or more perspicuously be inferred, than by the reduc-
tion of known effects unto their causes, and of these causative
entities (whose number and ranks are finite) into one prime
essence, whence all of them are derived, itself being underivable
from any cause or essence conceivable. In that this prime
Essence hath no cause of being, it can have no beginning of being:
nd yet is beginning of being the first and prime limit of being,
without whose precedence other bounds or limits of being cannot
\ follow.’”2
And a century later Bishop Beveridge in his Discourse on the Thirty-
nine Articles states the regress-argument for a First Cause in the
classical form,* whilein his Thesaurus Theologicus he outlines his
first argument for the existence of God as follows: —
6é
1. From the order of causes; for of every effect there must
be a cause, till we come to the first and Universal cause of all
things.
“Everything that is, was either made or not made; if -
made, it must be made by something that was not made.”’
and then adds in confirmation arguments from conscience,
miracles; prophecy and universal consent.? In his Private Thoughts
he gives the argument in a peculiar quasi-Cartesian form:
“The other articles of my faith I think to be true because they
are so; this is true, because I think it so: for if there was no God,
and so this ies not true, I could not be, and so not think it
' true Butin that I think, Iam sure I am; mee in that Tam; Tam
sure there-is a God; for ifthere was no eae how. came I to be?’
1 Works, V, p. 12 f. First published between 1613 and 1637.
2 Works, VI, p. 5. First published posthumously in 1716.
3 Works, TX, p. 49. No details of the arguments are given. The Thesaurus was first
published in 1711, posthumously.
; THE TRADITIONAL APPROACH 65
He adds that
“the like may be said of all other created beings in the world.
For there is no natural cause can give being to any thing, unless it
has that being it gives in itself; for it is a received maxim in
philosophy, that ‘nothing can give what it has not.’ And so,
however the bodies of men, or brutes, or plants, may now in the
ordinary course of nature be produced by generation, yet there
must needs be some one Supreme Almighty Being in the world,
that has the Being of all other beings in itself; who first created
these several species, and endued them with this eens bss power
to propagate their kind.’”!
The arguments, as we have seen, depend simply upon the brute
fact of the existence of being as eine: The particular properties
of the beings under consideration are entirely irrelevant. As it has
somewhere been said, reduced to their bare essentials they are
8
equivalent to the unadorned assertion; “Ifanything exists, God
must_exist; if God: does not exist, nothing can exist.” They do,
however, involve the recognition of the existence of finite being.
This might not seem a very startling assumption to make, in view
of the fact that we plainly perceive beings that exist, a fact that to
St. Thomas would be too obvious for question. Later philo-
sophers have, however, so subtilized and idealized being away
that the point needs emphasis. It is a fundamental axiom of the
Thomist theory of knowledge that the human intellect is by its
very nature capable of apprehending beings, whatever may be at
times its misunderstanding of their nature. Against all sensation-
alist and conceptualist theories it insists that the impressions pro-
duced upon the senses and the concepts formed in the mind are not
the objects of perception but are the media through which external
objects are perceived; not the quod but the quo of the cognitive
act. And against all phenomenalist theories it makes the same
assertion about the qualities which we observe. Extra-mental
reality is |the direct object of human perception; we know from
experience _‘that beings exist. This is not something that can be
argued about; a man either admits it or he does not. But even if
he does not ans it in words he must behave as if he believed it,
‘if he is to go on living at all.?
“=+-Works, VIII, p. 140.
2 A note may not be out of place on the more recent.type-.of scepticism taught by the
logical positivists, who, following in the steps of Wittgenstein, deny the relevance of all
metaphysical arguments because, as they hold, all ‘Tnetaphysical statements are
meaningless and therefore cannot apply” to anything: aiyats)eee is ai to two thain
66 3 HE WHO Is
But even if this realist theory of perception is rejected, the
arguments do not lose all their weight, for it is in practice impossible
to deny that something exists, unless you have attained a kind of
philosophical Nirvana. If, with Mr. F. H. Bradley, you reduce
abjections. (1) If it is true, no syntactical proposition can be significant, and hence its
basic principle that ‘‘the meaning of a proposition is the method of its verification,”
which is a syntactical proposition, is nonsense. Hence, _as_ Weinberg says, “‘logical
positivism cannot eliminate metaphysics. without destroying itself.” (Examination of
Logical Positivism, p. 199.) ¢(2))It involves itself in’a peculiar type ofsolipsism (“lin-
guistic solipsism”’ or “‘solipsism without.a subject” [Weinberg, op. cit., ITI, vii, and
p. 68]), in which communication of the sense of propositions is impossible. There is
also a very doubtful postulation of “empirical atomic facts” as the ultimate and exclu-
sive referents of discourse. Carnap’s logical syntax of language and the “‘radical
physicalism”? of Carnap and Neurath try to meet these objections, but apparently
with doubtful success; they are briefly described in Carnap’s two small books,
Philosophy and Logical Syntax and The Unity of Science. Weinberg sees in Carnap the
same linguistic solipsism that Carnap sees in Wittgenstein, and so produces his own
theory of language, which he claims is less open to question. We may also refer to the
very telling criticisms made by Mr. Bertrand Russell in Meaning and Truth, ch. x, xxii.
It is interesting to note in the whole movement a transition from an earlier phase in
which the rejection of metaphysics as meaningless is necessitated by the logical theory
that has been adopted, to a later stage in which, the former logical theory having beer’
largely abandoned, the meaninglessness of metaphysics is postulated as a dogma and a
new logical and linguistic theory is constructed to conform to this. One is thus led to
suspect that perhaps after all metaphysics has some validity. Weinberg himself rejects
metaphysics on the ground that “‘it is plain that the only specifiable referent of a non-
logical statement is the non-discursive realm of empirical reality”? (op. cit., p. 290).
But he apparently identifies metaphysics with the attempt to extract existential pro-
positions from purely logical] or ideal data. This would provide a valid argument
against the ontological arguments of St. Anselm and Descartes or the Hegelian
dialectic ;indeed, it is the essence of the objection made by St. Thomas to St. Anselm.
It does not, however, destroy the Thomist position, which takes its starting-point in
concrete empirical beings. The Thomist position does, of course, assume the doctrine
of analogy, which Wittgenstein’s statement of the meaning of a proposition denies;
but then, as has been remarked, it is not clear that Wittgenstein’s approach is valid.
Mr. A. J. Ayer has tried in.his recent work, Language, Truth and Logic, to save the
situation.by a considerably mitigated positivistic doctrine. He distinguishes practical
verifiability from verifability in principle, since we understand many propositions
which we cannot verify, e.g. ““There are mountains on the far side of the moon.”
Again, he distinguishes the “‘strong’’ sense of verifiability (when the truth could be
conclusively established by experience) from the “weak” sense-(when experience
could only lead to a probability). He points out that the strong sense would rule out
as meaningless such propositions as “Arsenic is poisonous” or “All men are mortal”
(which such positivists as Schlick indeed describe as merely “important nonsense”).
All that he demands for a proposition to have meaning is that an affirmative answer
can be given to the question, “Would any observations be relevant to the determina-
tion of its truth or falsehood?”’ Having said this, he then goes on to argue that meta-
physics, ethics and theology are all mere pseudo-sciences, since, as he alleges, the
statements which they make are, even in the light of this mitigated doctrine, meaning-
less.
Three replies seem to_be called for. (1)If ““meaningfulness” means ‘‘verifiability,”
then the strongest form of the positivistic doctrineisnecessary. If, on the other hand, a
weaker form ofthe doctrine is adopted, then ‘“‘meaningfulnéss” and~“‘verifiability”’
are distinct, and the very basis of logical positivism is undermined. “(2) In any case,
whether significant statements are always verifiable or not, “méaningfulness”” does
not mean “‘verifiability” but “intelligibility.” Metaphysical statements may ““mean”’
something even if we cannot verify them or even if they are inherently. unverifiable,
for we may know their meaning not by verifiability but, for example, by analogy,
THE TRADITIONAL APPROACH 67
everything that the ordinary person assumes to be real to the
status of a nce, the question arises—appearance
of what?
If; with Kant, youmaintain that the di is unknowable,
atleast you know that it exists. If,with Sir WillSpend} youdefine
objects_as_complexes
of persisting opportunities of experictice, it
may well be asked—experiences
of what? For it will not get us
any further to say that we have opportunities of experiencing
complexes of opportunities of experience.! If you descend to the
depths of solipsism—that vice which, as Professor Broad has
remarked, is more often imputed than committed?—you presum-
ably believe in your own existence, unless indeed you have
managed to attain the position of the character in Chekhov’s
Three Sisters who toyed with the hypothesis that perhaps, after all,
we don’t really exist but only think we do. [f,again with Kant,
you maintain that the category of causality applies only within the
sphere-of phenomena, the answer is that the sphere of phenomena
is precisely the sphere in which causality—not in the secondary, _
Fc a watts j F 4
\ (3) If moral and aesthetic judgments are rejected as mere expressions of feeling without
any Meaning, it is difficult to see why judgments about the external world should not
be rejected in the same way. It may be added that, in spite of his seventh chapter,
Mr. Ayer does not succeed, as Professor Stebbing showed in her review in Mind of
July 1936, in avoiding linguistic solipsism, nor does he seem to do so in the discussion
of “the Egocentric Predicament”? in ch. iii of his later work, The Foundations of Empirical
Knowledge.
The outcome would seem to be that, when it discards whole tracts of human thought
as meaningless, logica] positivism is merely demonstrating the limitations it has placed
itself under by its initial assumption that meaningfulness is identical with verifiability.
If, as Mr. Ayer contends in opposition to the whole tradition of the philosophia perennis,
philosophy is purely linguistic in character and so is concerned with words and not
with things (Language, Truth and Logic, p. 62), the question of the relation between the
realm of logic and the realm of concrete reality, between words and things, remains -
unanswered, for there is no other science that can answer it. And it is interesting to
note that, at the end of his recent Inquiry into Meaning and Truth, Mr. Bertrand Russell
concludes “‘that complete metaphysical agnosticism is not compatible with the main-
tenance of linguistic propositions” (p. 347). He previously expresses the opinions
that “ ‘true’ is a wider concept than ‘verifiable,’ and, in fact, cannot be defined in
terms of verifiability” (p. 227), and that ‘“‘there is no reason why ‘truth’ should not
be a wider conception than ‘knowledge’ ” (p. 246).
It is no doubt too early to pass a final judgment upon the various types and off-
shoots of logical positivism. But, for a short discussion which is both sympathetic and
discriminating, see the article by Mr. D. M. MacKinnon under the title ““And the
son of man...” in Christendom for September and December 1938. The philosophia
perennis would at least agree with the logical positivists in rejecting a purely deductive
metaphysics having no basis in concrete fact, though it may be doubted whether they
have realized this.
1 See ‘‘The Eucharist” in Essays Catholic and Critical and cf. Belief and Practice,
lect. xi; also the discussion between Sir W. Spens and Canon Quick in Theology,
January to March and May 1939, following the publication of the latter’s book, The
Christian Sacraments.
2 Examination of McTaggart’s Philosophy, II, p. 259. Cf. Mr. B. Russell’s amusing
story of the correspondent who said that solipsism was so reasonable a philosophy that
he could not understand why so few people held it. (Outline of Philosophy, p. 302.)
F
68 HE WHO IS”
and for our _purposes -irrelevant, sense that the word bears in
physics, as merely indicating t
the occurrence ofregular sequences,
but in the proper; and philosophical ssense of,efficient causality—
can be«dispensed with.! Where it cannot be dispensed with is in
“the: sphere of being. As soon as we have recognized that something
really exists, then either it is necessary being and the argument is
complete before it has begun, or it is contingent being and so
necessarily implies the existence of necessary being.
A rather different objection might be brought against our
position. We agreed in an earlier chapter with Professor Broad,
when he found fault with the Ontological Argument on the ground
that it professes to make a categorical comparison between a non-
existent and an existent in respect of the presence or absence of
existence, whereas no comparison can be made between a non-
existent term and anything else except on the impossible hypo-
thesis that it exists. Now, it might be said, does not our present
argument contain just this flaw? In arguing that contingent being
points to necessary being as the ground of its existence, are we not
in effect comparing a contingent being that exists with a being
that is precisely similar in every respect save that of existence and
then postulating necessary being as the cause that confers existence
in the one case and not in the other? A little consideration will
show that, plausible as this accusation may appear, it is not true.
According to the classical arguments God is not needed simply as
the power which can confer existence on essences which dwell
independently of him in an ideal.world, as in Professor Eddington’ S
famous example of the uncocked snooks waiting in patient non-
existence for somebody to cock them.* Such ultra-Platonism is
entirely foreign to their outlook. They postulate God as
1s.the
essence-realized-in- -existence, of the ‘finite thing. They do not
compare the existent essence with a non-existent essence, because
they do not compare it with anything at all. They direct our
thought, not to existence illicitly considered as one of the thing’s
1 At least a passing reference must be made to the vigorous defence of the realist
doctrine of perception undertaken by M. Gilson in his Réalisme ThomisteetCritique dela
Connaissance, and to the criticisms which he passes there on such attempts as those of
Mer. L. Noél, Fr. Picard, Fr. Roland-Gosselin and Fr. Maréchal to combine the
essential features of Cartesian and post-Cartesian thought with those of Thomism.
Chapters vii and viii of this book contain a very clear statement and. defence of the
realist position.
2 The Philosophy of Physical Science, p. 214.
THE TRADITIONAL APPROACH 69
Po
properties, but to the contingency which is radically inherent in
its very being, in all that it is and in all that it has. They are
concerned with the thing itself in its concrete actuality; the
universal is all the time in re.
_The arguments are thus entirely general in their scope, and
require as their starting-point nothing more than the real existence
of something, no matter what. But, for this very reason, they can
tell us very little about the nature of the ultimate. being at which
they arrive. Thus St. Thomaswrites
yrites that:
“from effects not proportionate to the cause no perfect knowledge
of that cause can be obtained. Yet from every effect the existence
of the cause can be clearly demonstrated, and so we can demons-
trate the existence of God from his effects ;though from them we
cannot perfectly know God as he is in his essence.”
And again:
“Our mind cannot be led by sense so far as to see the essence of
God; because the sensible effects of God do not equal the power of
_ God as their cause. Hence from the knowledge of sensible things
the whole power of God cannot be known; nor therefore can his
essence be seen But because they are his effects and depend
on their cause, we can be led from them so far as to know of God
whether he exists, and to know of him what must necessarily belong
to him, as the{ first cause of all things, exceeding allthings caused
by Ga “na
aed
We cannot, therefore, discover by argument either why God is or <-——
what he is, but only whether he is and what he is not. Our argu-
. ments are demonstrationes quia, not propter quid. They answer the
question an sit, not quid est. This might seem to deprive them of
all value. What, it might be said, is the use of knowing that there
is a God, if we know nothing about him except that he is the cause
of finite beings? Is not our knowledge of him as scanty as would
have been Robinson Crusoe’s of Man Friday if when he saw the
footprint on the sand he had had no previous knowledge what-
ever of human beings or their feet? ‘This is a question that can
only be answered by experiment. And it is, in fact, surprising
how many attributes of God St. Thomas sees to be implied in the
fact that he is the cause of finite being. The reason—so far as we
can expect a reason—would seem to lie in the amazingly full and
1 §. Theol., I, ii, 2 ad 3. 2 §. Theol., I, xii, 120.
70 HE WHO IS
fertile content of the idea of being itself.1 Little, indeed, can be
deduced about a being from the mere knowledge that it must be
such as to be able to leave an impression of a certain form; but
much can be deduced about it from the knowledge that it must
be such as to be able to impart not only form but being. And it is
almost amusing to see how much St. Thomas tells us about the
First Being after he has informed us that we have really proved
nothing about him except that he exists. Even so, we can acquire
by the use of reason no information as to the inner constitution of
his being. We know that God is supremely intelligent, but not
that this intelligence is the eternal generation of a Son; we know
that he is good, but not that he raises us to a share in his own
nature. All such truths as this come to us by revelation alone.
We saw that each of the Five Ways ends with the words, “And
this everyone understands to be God”’ or their equivalent. A very
natural reaction to this assertion would be to say, ‘But why?’’
St. Thomas has not at this stage very clearly even committed
himself to a definition of what he means by God, though he has
admitted that St. Anselm’s quo majus nihil is a true description of
God if he exists? and has just quoted the Mosaic text, Ego sum qut
sum2 The Five Ways have each led us to admit the existence ofa
being (or, we might wonder, should it be “beings” ?) with a deter-
minate character: First Mover, First Efficient Cause, First Neces-
tee
sary Being, First and Supreme Perfection, Ultimate Final Cause.
But must.all thesebe the same being? Is it, for example, certain
that the First Mover is also the First Efficient Cause? Might not
the universe be governed by a Supreme Council of Five?
One way of meeting this objection is to try to show that each
of the five attributes can only belong to sélf-existent-being. Dr.
“Phillips summarizes this reply in‘s book from which we have
already quoted. He argues that (1) ) the First Mover is always in
act and contains no element of potentiality, hence it is not merely
capable of existing, and so exists essentially;
(2) the First Cause,
Pesach
1 Cf. Maritain: “The scire an sit or quia est (knowledge by the record or the perspec-
tive of fact) is in no way limited to knowledge of an inductive type, for (in opposition
to the scire quid est or propter quid est, which is knowledge in the record of or the perspec-
tive of reason of being) this expression includes all knowledge which does not arrive
at grasping the essence itself in the totality of its intelligible constituents.”’ (Degrees of
Knowledge, E.T., p. 42, n. 1.) He remarks that ‘“‘all knowledge which does not attain
to the essence in itself belongs to scire quia est.” (Ibid., p. 283.) Dr. R. L. Patterson has
indeed gone so far as to argue from this that the distinction between knowledge quia
Deus est and quid est is invalid. (Cone. ofGod in Phil. of Aquinas, p. 257.)
28. Theol., I, ii, 1 ad 2. 3S. Theol., I, ii, 3, sed contra.
‘ THE TRADITIONAL APPROACH 71
being uncaused, cannot receive existence from another, and hence
must itself be existence a3) Hecessery Being is incapable of not
existing, and hence must have existence as an essential predicate;
Supreme Being cannot be composite, and_so cannot have
a share of existence, but must be existence; hough the
Supreme Intelligence has not been strictly demonstrated to be
infinite, there is no reason to disassociate this argument from the
others, and furthermore such an intelligence must be Pure Act, for
if it were essentially related to an object of intelligence, such a
relation would, in virtue of the Teleological Argument itself, have
to be produced by a higher intelligence.
It is doubtful whether such arguments as those just quoted are
really conclusive, however probable they may seem. It might be
suggested, for example, that “‘self-existent’’ does not mean pre-
cisely the same in each case; that in the first it merely means
“receiving motion from no other being,” in the second “being
efficiently caused by no other being,” and so on. Haveall the
five First Beings been shown to be self-existent purely and simply,
or only in five closely similar but-not necessarily.identical.senses?
No doubt this question could be argued interminably, but it is
noteworthy that St. Thomas himself does not follow this line. He
simply says, “This all men speak ofas_God,” and he seéms much
more interested to deduce the attributes of God from the Argu-
ment from Motion, which, as we have seen, is given the place of
priority, than to demonstrate that the Five Ways all lead to the
same Supreme Being. In the Compendium Theologiae, having
briefly stated the Argument from Motion, he develops 1in turn the
attributes of:immobility, eternity, necessity, aseity, unity, and so
on.2. He followsa similar course in the Contra Gentiles, atid; while
“— go
it is true that in the Summa Theologica he expounds the Five Ways
we
en bloc, it is also true that he hastens‘to establish, as the core of his
thesis, that God is his own essence and that therefore in him
essence and existence arethe same. But perhaps we may see in
the very phrase, “‘ This all men speak of as God,” the clue that we
are seeking. For, as we saw earlier, the Five Ways are, if Fr.
Garrigou-Lagrange is correct, simply five scientifically stated
analytical formulations of different aspects of one fundamental
inclusive argument addressed precisely to the “‘all men”’ to whom
St. Thomas appeals. They are, that is, not really independent of
one another, but are related as different aspects of one whole.
1 Modern Thomistic Philosophy, II, p. 292. 2 I, iil, seq.
qo HE WHO IS
But perhaps we shall find even a fuller answer when we go on to
examine more thoroughly the evidential value of the Five Ways.
We may begin this examination by observing that, in effect,
each of the ways is in the form of a hypothetical constructive
syllogism (modus ponens). It proceeds as follows:
If there exists a being that is finite in such-and-such a respect,
then there exists a being that is infinite in such-and-such a
corresponding respect.
But there does exist such a finite being.
Therefore, there exists the corresponding infinite being.
Then there is the appeal to universal agreement : this infinite being
must be what we mean by God.
For instance, the third way argues thus:
If there exists a contingent being, there must exist a Necessary
Being.
But there does exist a contingent being (e.g. Cleopatra’s Needle,
or Dr. Goebbels, or the oxygen in the atmosphere).
Therefore, there exists a Necessary Being.
And “this all men speak of as God.’”?
But, it may be asked, how do we know that the major or
hypothetical premiss is true? Presumably because we have come
across contingent beings, such as Cleopatra’s Needle, Dr. Goebbels
or the oxygen in the atmosphere, and have seen from a thoughtful
onsideration of them that they do not provide the reason for their
own existence. If, per impossibile, I have never met with a con-
tingent being—and the hypothesis is strictly impossible, since it
supposes that I had never met with myself—or if, as is more likely,
I had never seriously thought about one—I shall certainly never
have been led to the assertion that, if a contingent being exists, a
/ Necessary Being must exist as well. This is a good Thomist argu-
ment, for one of the basic principles of Thomist philosophy is that
our abstract knowledge is always derived ultimately from acts of
sensible experience; nihil in intellectu quod non prius in sensu. In other
words, although we may be convinced of the truth of the major
premiss as a necessity ofreason, and although we may be convinced
that, from a purely formal standpoint, the syllogism is a valid
logical inference, in the concrete our postulation of both the major
1 J have here stated the argument in its simplest form, not in the two steps as in the
Summa Theologica.
r THE TRADITIONAL APPROACH 73
and the minor premiss has been derived from the same basis in our
experience, namely,our acquaintance with contingent beings. In
practice, it is impossible for us to have been led to accept the
major premiss without having been led to accept the minor as
well, and, unless we are quite extraordinarily devoid of powers of
inference, without having been led to accept the conclusion too.
That is to say, in practice the argument is either accepted or
rejected as a whole according as we have or have not come to
know the things of this world as being what they really are. And
this contention is strengthened by the fact that, as we have seen,
in the hands of St. Thomas the argument does not simply assert
that the proposition “Necessary Being exists”’ is a logical conse-
quence of the proposition “Contingent being exists,’ but main-
tains that contingent being derives its own existence from Neces-
ry Being; in other words, that we are not concerned just.with~.
ekerelations between. propositions,_but..with metaphysical-or™”
“entologieal relations between existent beings.
What is necessary, in short, if we are to pass from a belief in the
existence of finite beings to a belief in the existence of God is not
so much that we should thoroughly instruct ourselves in the laws
d procedures of formal logic as that we should thoroughly
Caequain ourselves with finite beings and learn to know them as
y really are.
This is not it must be stressed, a lapse into ontologism; it is not
here maintained that we can, by the use of our natural powers
directly apprehend God, that God is the terminal object of all or
any of our acts of perception, or that finite beings are only the
instruments or media through which we perceive him. On the
contrary, our acts of perception terminate in the finite beings,
which are the quod and not the quo of the acts; and it is only too
possible for us to apprehend them in their bare particularity and
to derive from them no notion of a Creator. Since, however, the
human mind, just because it 7s a mind, is essentially adapted for
the understanding of being as such (however much that under-
standing may be limited by the fact that the mind is united to a
body and experiences external beings through the bodily senses,
and however much it may be obscured by the mind’s own frailty
and opacity), it remains true that the mind is not only able to
apprehend the mere existence of finite beings and*their-external
properties, butis also capable in some dégree-of-compreheénding
them,of entering into their inner essence and making them its
74, HE WHO IS
own, of recognizing not only their finitude but also whereon that
fini “Ifperceive
We finite beings as they actually are, we
shall perceive them as the creatures of God. And if we do so
perceive them sub ratione creaturarum, we shall in perceiving them
recognize the existence of the God whom we cannot perceive. We
cannot, at any rate in this life, know God under the aspect of his
deity, but we can know him under the aspect of his Creatorship,
in recognizing his creatures for what they are.
Mr,_C..E. Faithfull-has said that “it would seem that St. Thomas
can only escape the charge of ontologism in some form or other if
we may, with some Neo-scholastics, see implied in all his system
the activity of intuitive as well as of discursive reason.” He nr
recognizes that St. Thomas denies to man the type of intuition that
the angels possess; that is to say, a pure intuition of the essence
of beings unmixed with any discursive mental act or any inter-
vention of the senses ; but he points out that the intuitive powers of
/the human mind function only in the closest co-operation with the
|discursive reason, ““T herecognition-of
contingent being, and the —
argument from the contingent to.the necessary,”’-he-says,-<<illus-
trates the force with which the principle of causation can be
recognized apparently spontaneously. In view of the modern
awareness of the hidden activity of the discursive reason that can
be involved in such an experience, and that unconsciously but
unremittingly tests it by the light of other experience, it could
hardly be maintained that St. Thomas would be forced by his
rejection of intuition to deny the identity of this approach to God
with his own. ... Granted the possibility of the illumination of
experience by some form of intuition, there is no great difficulty of
allowing that the existence of various degrees of being may be
perceived as a fact, and that the existence of the standard, the
absolute, may be not a presupposition nor yet a corollary, but
perceived in the same intuitive act.’! It is, we must observe, not
1 “The Proofs of the Existence of God in the Philosophy of St. Thomas Aquinas” in
Laudate, September 1934, p. 160 f.
How real an appeal to intuition is hidden behind the axiomatic and deductive
structure of even the most abstract of sciences, namely pure mathematics, is seen from
such a work as Dr. F, Gonseth’s Fondements des Mathématiques. “It is not only,” he
writes, ““when he applies rigorously and impeccably the rules of logic that the mathe-
matician can avoid contradiction, but when in addition, by an unconscious and pro-
found divination, he applies them knowingly. ... Thought, when it tries to find its
support and its foundation in itself, can ultimately meet with nothing but a vacuum.
. . . In its essence mathematics is only a collection of schematic views and processes of
our mind, a conscious replica of the unconscious activity which creates in us an image
of the world and a collection of norms according to which we act and react” (pp. 239,
240).
\
i” THE TRADITIONAL APPROACH 75
God himself that is perceived, but the fact of his existence; or,
since the word “‘perceive”’ is usually restricted to our knowledge
uae of facts but of beings, it would “pepe be better to say not
“perceived” but ‘apprehended. se
If this account is true, it will follow that in knowledge of God
to which the consideration of finite beings leads us is not merely
probable but certain. Provided that we put ourselves in the right —
frame of mind for seeing things as they really are—and this, of | |
course, in practice involves a real effort of moral and intellectual
integrity—we can grasp the fact of God’s existence as the ground
of the existence of the beings under our consideration with just as
much certainty as wewe| perceive the beings themselves.!1 If we were
merely arguing discursively from the fact of their existence our
argument might indeed be only probable, and we should have to
agree with many modern writers that the exercise of reason can
lead _us to no more than a probability that God exists; then, like
scal,in his famous “wager argument,” we should have to stake
Oo s upon that probability.? The various arguments ‘for God’s
existence would then merely reinforce one another by increasing
the probability, but we should never attain certainty, and our
recognition of God would be a pure leap in the dark—an act,
which, for some reason, is held by many people at the present day
_to be highly meritorious. This is not, however, the position that is
being urged here; and we agree with St. Thomas that the various
arguments do not increase the probability that God exists, but
exhibit him ‘to us under a number of different aspects, as First
Mover, First Efficient Cause and so on. And here we must observe
that in this particular respect the parallel that was drawn in the
third chapter between our conviction of the existence of God and
a man’s conviction of his wife’s fidelity breaks down. For this
latter conviction, based as it is upon a vast number of indications
most of which could only with extreme difficulty be put into words,
can never lead to complete certainty; it can at most lead only to
that moral certitude which is all that we demand for the practical
ordering of our lives; even those husbands who were most sure of
their wives’ faithfulness have sometimes been deceived. And pre-
sumably even the most devoted and happily mated of husbands
would, if he were a convinced believer in God, be ready to admit
1 See Appended Note on “Certainty and Certitude,” p. 93, infra.
2 The “wager argument” is, of course, strictly ad hominem and does not represent
adequately the basis of Pascal’s own belief. It is stated in the Thoughts, E.T., No. 233.
76 HE WHO IS
that his certainty of God was of a higher order than his confidence
in his wife. For what is being maintained hére is that our convic-
tion of God’s existence is not merely of a high degree of probability
or even of purely moral certitude; it is objectively certain. And,
as has just been indicated, the purpose of the multiplication of
proofs is not to increase our assurance of God’s existence but the
scope of our knowledge about him. .
We may admit that, to the modern mind, it will appear that in
thus maintaining that the existence of God can be known with
certainty we are depriving belief in him of all merit. What credit,
it will be asked, can there be in recognizing a fact that is indis-
putable? We do not admire a man for admitting that seven nines
are sixty-three. This objection might be countered by inquiring
what merit there can be in acting as if things were certain when
they are only probable. Is not this, we might say, simply intel-
lectually dishonest? Is it not acting a lie? Admittedly we have to
act either as if God existed or as if he did not, but, while it may
be more prudent to base our lives on the more probable alterna-
tive, it is difficult to see that such a course is more creditable. And
if we do not admire a man for admitting that seven nines are
sixty-three, nor do we admire him for betting upon the crew
which he thinks is more likely to win the Boat Race. Such a reply
would, however, be merely ad hominem, and having made it we
may, like St. Augustine in another connection; disclaim it with the
ervation, Aliud est videre, aliud est ridere. And, in.fact, there is at
least one respect in which some credit is due to a man for recog-
nizing that God exists, even if his existence is absolutely certain.
We must_notice first that the act of accepting any intellectual
proposition involves.the.operation-of the-willkEts-subject.is
not
the intellect in. isolation,..but-theewhole*man; and the will is one
of his faculties just as much asthe intellect is. I have to make an
act of will in order to admit that seven nines are sixty-three just
as I have to make an act of will in order to speak courteously to
an unwelcome visitor. It is, of course, true that in most cases to
admit that seven nines are sixty-three makes no moral demands on
me; therefore the admission is in no way meritorious and I make
it spontaneously. There are certain intellectual propositions, on
the other hand, whose acceptance makes a real moral demand. I
1 This apprehension of God’s existence seems to be in essence identical with the
knowledge of divine things by intellectual connaturality which M. Maritain describes
in the third chapter of Quatre Essais sur l’Esprit (p. 134), a natural contemplation which
he distinguishes plainly from ‘mystical experience of the natural order.
‘
e THE TRADITIONAL APPROACH 7 is
may refuse to admit—even to myself—that I owe a certain debt,
not because there is any doubt about it, but because I am
avaricious. I may refuse to accept a clear demonstration that I
am suffering from an incurable disease because I am lacking in
courage. And I may refuse to admit that finite being is really
what it is because I am lacking in the humility necessary to accept
my own finitude. For, as we have seen, the arguments for the
existence of God start from the recognition of finite being, and this
involves the recognition that we are finite ourselves. It must be
observed that what is in question at the moment is not the previous
acts of diligence necessary if we are to embark upon the considera-
tion of the arguments, or the subsequent acts of obedience to God
that are necessary if we are to put our belief into practice after we
have acquired it. Those acts are indeed meritorious, but they do
not impart merit to the acceptance of the proposition itself. The
point is that, however clear the truth of the proposition “God
exists’? may be—and it has been asserted that our recognition of
it can be so immediate as perhaps to deserve the name of intuition
rather than of argument—unless we have the virtue of humility we
shall simply be unable to see the data as they are and so we shall
also be unable to see God’s existence as implied in them. There
is thus a thfefold orala tivity involved : diligence. in investigat-
ing t question, humility jn_recognizing-the-data, andCourage
in acting upon tlthe conviction when acquired. But it is the sécond
of these that is involved iin_the actual intellectual.acceptance of
God? s existence, and, if it is lacking, we shall simply hide God’s |
_evidence from ourselves by putting up a kind of intellectual smoke-/
\screen. aes
One other condition is also necessary. If we are to recognize
finite being as what it reallyis, it is not sufficient merely to recog-
nize it as finite; we must_ dle recognize it as,being. “Yhis is, of
course, where idealisticphilosophies of the Bradleian type are so
1 We may recall the peculiar semi-Cartesian argument for the existence of God stated
by Bishop Beveridge iin his Private Thoughts, which is based upon his own existence as a
thinking being. “‘The other articles of my faith,” he wrote, “I think to be true
because they are so; this [sc. that God exists] is true, because I think it so: for if there
was no God, and so ’ this article not true, I could not be, and so not think it true. But
in that I think, I am sure that I am; and in that I am, I am sure there is a God;
for if there was no God, how came I to be?” (Works, VII, p. 140.) ‘This particular
form of statement is of hardly more than historical interest, but it exemplifies the point
just under discussion, namely the finitude of the investigator himself. It is followed by
a general statement of the argument from causality:“And the like may be said of all
other created beings in the world. For there is no natural cause can give being to any
thing, unless it has that being it gives in itself,” etc. (p. 141).
d \
78 ; HE WHO IS
unsatisfactory. In his famous work, Appearance and Reality, Mr.
F. H. Bradley claimed to have reduced to the status of appearances
everything that we normally recognize as being real: qualities,
relations, time, space, things and selves fall beneath his blows in a
remarkably small number of pages. Beneath-all his arguments,
_vhowever, th
there lies one recurrent fallacious assum:
a that_anythingg that can be shown not to be infinite being is not
“being at all,but. only appearance. “And as it is fairly easy to show
“that none of the entities mentioned above is infinite being, it is
fairly easy to draw his conclusion. The consequence of his denial
of the analogy of being is that, instead of recognizing finite beings
as the creatures of God, he treats them as appearances of the
_-Absolute. But Christian theism insists that finite beings can be
ra directly _ apprehended_as_beings that aré~finite;-and™that~ this
recognition involves the immediate postulation of a Beng thatis
_ infinite as their-ground.
If the nature of the classical arguments for the existence of God
is really as we have described it, one further point becomes clear.
That is, that the Five Ways do in fact lead to the same infinite
Being and not t to five different beings infinite in five different
respects... It also solves such difficulties as that which may be
raised in connection with the Third and Fifth Ways, that it is not
clear that there is only one Necessary Being and only one ultimate
Final Cause. For though, considered simply as syllogistic argu-
‘ments, the different Ways terminate in five different aspects of
infinite being and thus do not demonstrate their unity in one
Infinite Being, considered as intuitions (in the special sense that
we have given to this word) they all terminate in the same Being.
The different arguments are not obtained by syllogizing five
different kinds of act of inspection of finite beings; they are all
obtained from one kind of act, and indeed can all be obtained
from one single act. If we see any finite being as it really is—one
point of moss or the smallest ant, to use Maritain’s illustration—
we shall see that God is implicated in it as First Mover, First
Efficient Cause and all the rest. We penetrate the nature not of
one particular being, but of being itself, in all its complexity and
fecundity, in all its analogical character and inner dynamism.}
} As Fr. Rousselot says, “ Far from characterizing intelligence as the faculty of abstrac-
tion, we must, on the contrary, designate it the faculty of.complete “intussusception’.”
(The Intellectualism of St. Thomas, E.T., p. 32.) He is, of course, at great pains to make
it clear that the human soul is the lowest of all intelligences.
” THE TRADITIONAL APPROACH 79
For, to quote Maritain again, “one cherry between one’s lips
holds more mystery than the whole idealistic metaphysic.”’+
When this has been said, however, it must not be concluded that
the statement of the Ways in the form of syllogistic arguments is
unnecessary or mistaken. The precise formulation which St.
Thomas gave to them was of course adapted to the intellectual
climate of his time; it was because philosophers (including him-
self) were interested in Aristotle and the syllogism that he put his
arguments in the precise Aristotelian form. But some syllogistic
or quasi-syllogistic form was necessary if they are to be stated as
arguments at all, and even in the formation of an act of intuition
such as we have described argument plays a very real part. The
relation between argument and intuition may perhaps be made
clearer by a simple parallel.
Anyone who has studied mathematics knows the difference
between merely being convinced by a long chain of reasoning and
getting areal and intimate grasp of amathematical theorem. Inthe
former case, if each step is seen to be valid, the student feels bound
to admit the truth of the conclusion, but his acceptance of it
remains disturbed and dissatisfied; he sees that the conclusion is
true, but does not see why it is true; he does not see how premisses
and conclusion are related as parts of a whole possessing a definite
—one might even say, an aesthetic—form. From time to time,
however, and very often after long and painful consideration of
mere chains of reasoning, the form suddenly becomes evident in a
flash; the theorem has been “got hold of”’ at last, it is, as it were,
seen ‘‘in the solid,” and the student feels that he is now not just
assenting to an external fact but that he has penetrated to the
nature of the object and. made it part of himself. He sees the
conclusion as involved in the premisses, and not only as derived
from them by a chain of discursive reasoning. And the reason
why some mathematical methods are disliked as being “‘messy,”’
while others are described by such adjectives as “‘elegant’’ and
“beautiful,” is primarily because the latter, but not the former,
stimulate precisely this kind of intuitive grasp of theorems as a
whole. A fairly close parallel may be drawn between this and the
case that we have been considering, though we must not forget the
difference in the subject-matter of the two sciences, mathematics ee
Os
being concerned with entities of reason derived from concrete
beings (entia rationis cum fundamento in re) and metaphysics with
1 The Degrees of Knowledge, E.T., p. 411.
80 HE WHO Is
concrete beings (entia realia) as such. ~The_arguments for the
existence of God are not fallacious, and to anyone who under-
standswhat they are about and is capable of following them they
can carry complete conviction. Nor are they unnecessary, for —
without them—or at least without some equivalent consideration
such as Garrigou-Lagrange’s one general proof—our belief will
not be explicitly rational. But their real value is in stimulating
the mind to examine finite beings with such attention and under-
standing that it grasps them in their true ontological nature as
dependent upon God, and so grasps God’s existence as their
Creator.
This may provide the clue to the fact that at the present day
people in general seem to find it extraordinarily difficult to see the
necessity of the existence of God. Even if they can be got to listen
to an exposition of the Five Ways and to follow the steps of the
argument, they are frequently quite unmoved. This is, it may be
suggested, mainly due to the fact that, under modern conditions
of life, people very rarely give themselves the leisure and the quiet
necessary for the straightforward consideration of finite being.
They never really sit down and look at anything. This diagnosis
is borne out by the common experience of people making their
first retreat, that after the first day or so natural objects seem to
acquire a peculiar character of transparency and vitality, so that
they appear as only very thinly veiling the creative activity of God.
But what is possible after a period of repose and silence in the
atmosphere of a retreat-house is very difficult to achieve if one
spends one’s working life in a stockbroker’s office or a steel
foundry and one’s leisure time at the pictures or dancing to the
accompaniment of a saxophone-band. It is more easily achieved
in the country or on the ocean, and it is well-known that country-
folk and sailors have a sense of God as immanent in nature which
town-dwellers rarely possess. The following story, related by
Mr. Arnold Lunn, is very much to the point. After remarking
that “materialism is a disease of the great cities and does not
flourish among men who are in close touch with Nature,” he
writes: ‘‘I remember a night on an East Coast Convoy, when we
were expecting an E-boat attack. I asked the Captain, as he
peered out into the star-reflecting waters of a calm but sinister
sea, whether he had ever met a sailor who was an atheist. ‘No,’
said the Captain, ‘not one. Sailors have time to think.’ ””?
1 Blackfriars, September 1942, p. 368.
w THE TRADITIONAL APPROACH 81
We may add that, under modern conditions of life, our attention
is almost always concentrated upon the form that man has
imposed upon matter rather than upon the nature that God has
given it; we are thus led rather to see what man has done with
it than what God is doing with it. We do not consider copper and
aluminium and indiarubber and ebonite in their nature as physical
matter given them by God, but in their function as components
of a wireless-set given them by man. It is not therefore surprising’
Afwe find it difficult to see the Creator at work in his creatures; j
/ for we seldom or never look honestly and calmly at the creatures’
\_ themselves.
“One of the essential prerequisites, therefore, for an acceptance
of Christian theism is a contemplative and reverent attitude to
finite beings. It has been pointed out by several recent writers
that modern man’s lack of reverence for Nature has led to some
of the great economic evils of the day, the ruthless exploitation and
exhaustion of the soil for financial gain, the heedless squandering
of subterranean deposits, the almost irreparable destruction of
forests and so on; it now appears that the same failing has largely
been responsible for the loss of his belief in God. For it is hardly
likely that we shall see things as the creatures of God if our
primary attitude to them is as things for us to do something with.
We have already seen that a certain moral integrity is needed for
understanding the arguments for Christian theism; now we have
seen reason to believe that a readiness simply to sit down and look
-at things in order to penetrate into their nature is no less indis-
pensable.
There is one final point that calls for notice. It may seem that,
when all is said and done, we are hardly within sight of anything
of real value for religion. What, after all, is Self-existent Being in
comparison with the God and Father of our Lord Jesus Christ?
Is not Barth right in maintaining that the God whom natural
reason discloses to us is so faint a reflection of the God of revelation
that the natural reason, so far from providing a basis upon which
revelation can build, merely misleads and deceives us? What can
be deduced about God from natural theology will be discussed
later on. We may, however, admit at once that natural theology
is woefully insufficient to supply the religious needs of man; this
is one of the reasons why it is supplemented by revelation. But
when we look back, in the light of revelation, upon God as reason
has shown him to us, we are able to see that ipsum esse subsistens is
HE WHOIS
é — with all the fullness of Christian truth. oS
..which_is_apprehended by faith, surpasses reason, but doés-not
destroy. its.deliverances. ~
Andasrevelation surpasses-reason, sO
does the experimental knowledge of God that is granted to the
mystics surpass revelation. With pinasentences from M. Jacques
Maritain we will conclude.this.chapt
“In saying ‘Sabsistent t Being iitsell or ‘In him there is no real
distinction between essence andJ,exivtented the metaphysician
unseeingly describes that sacred abyss before which the angels
fall trembling with love and terror.’’! ‘‘Ifan ignorant shepherdess
can be raised to such [mystical] wisdom, it is true that she is
ignorant of metaphysics and theology, not that she is an ignorant;
she has faith, and by faith she grasps in their divine source those
truths which theologians disclose in the sweat of their brows.’”?
Of the Christian mystic it is true as of St. Augustine that “‘he
_ wishes to be united in the profoundest depths of the heart with him
_ who dwells there as in a temple and in whom alone the heart can
find rest, not the God of the philosophers and the wise men, who
may be attained without faith, not even the God of the theologians,
who may be attained without charity, but the God of the saints,
(the Life of our life, who gives himself to us through grace and in
\ love.’”3
>
1 Degrees of Knowledge, E.T., p. 284. 2 Thid., p. 294.
3 Degrees of Knowledge, E.T., p. 366. Cf. the words of a writer whose general attitude
is widely different from that of Maritain: “‘The experience of the saints gives to us a
deeper understanding |of human personality than the whole of metaphysics and
theology put together.” (Berdyaev, Freedom and the Spirit, E.T., p. 39.)
CHAPTER VII
INTELLECT AND INTUITION
HE argument of the preceding pages has involved the
assumption that our knowledge of the external world is
essentially what scholasticism would call adaequatio
adae _intellectus
R ad rem, that the human mind, by itsvery constitution, is capable
: of penétrating beneath. the phenomenal surface of finite beings
_ and of grasping them, however imperfectly and partially, in their
. ontological nature and so apprehending them in their dependence
upon an infinite Being which, as St. Thomas says, all men call
~ God,—A good deat has been” iG by the way in justification of this
point of view; but it is so foreign to the habits of thought of most
modern philosophers that it will be worth while spending a little
more time in trying to make its truth more readily acceptable.
The prevailing trend in epistemology in recent years has been
to deny both that extramental beings conceived as concrete sub-
stances or essences exist, and in consequence that the human mind
is constitutionally capable of apprehending them. The funda-
mental elements of our experience are taken to be such mere
particulars of sensory awareness as red patches, loud noises, hot
feelings on the skin and the like, and the mind is alleged to
construct from them, in one way or another, what it mistakenly
believes to be external beings. The particulars may be described
in many ways—as sense-data, sensa, sensibilia, or even oppor-
tunities of experience—and the operation of the mind may be
conceived either as the building up these particulars into artificial
complexes on the ground of accidental similarities, or as the
addition to them of characteristics which they did not possess
before, or in some other way. Dr. Tennant, for‘example, believes
that our experience is inevitably contaminated by an interpreta-
tion which we impose on it in the very act of experiencing; his
distinction between the “psychological”? and the ‘‘psychic,”
though superficially it resembles the scholastic distinction between
the quod and the quo of experience, is used in such a way as to deny
that we can have any undistorted knowledge of the quod.1 Mr.
Bertrand Russell believes that ‘‘events’’ are strung together into
1 Phil. Theol., 1, p. 255. See p. 165, infra,
G 83
84 HE WHO IS :
objects by the action of the person who perceives them, but as this
person is believed to be nothing more than a string of events him-
self it is rather difficult to make out who does the tying.1 Both of
them would agree that the only object of which we have direct
knowledge is the particular sensum, and both would assert that
the mental element in perception inevitably introduces error,
which subsequent reflection may or may not be able to correct.
All such views as these are in radical opposition to the Thomist
doctrine that, in the very act of apprehension, the reception of the
sensible species by the sense is accompanied by the abstraction of
its intelligible content by the intellect, which, passing through the
species, obtains a direct, though mediated, knowledge of the thing
itself, so that any error which may be involved is due, not to some
kind of demonic twist in the nature of intellect as such but to an
accidental disordering of it in the particular case. To say this is
not to assert that the human mind is capable of abstracting the
fullness of the essence and knowing it in all its ontological richness
as God, or even the angels, do. On_the contrary, scholastic
philosophy has always insisted that the human mind is the lowest
of all intelligences,
and indeed not worthy
to be called
an intel-
ligencé; but rather-a mere intellect; Fr: Rousselot has shown most
clearly how the human mind substitutes knowledge by concepts,
science, systems and symbols for knowledge by the pure idea.?
What we are concerned to maintain is that, with whatever limita-
tions, it is the inherent function of the human intellect, as the very
/derivation of “intellect” —from intus legere—implies, not to read
' into its object qualities that are not really there, as all the Kantians
\.and quasi-Kantians teach, but to read that which is within them,
to extract from them, however partially and imperfectly, their
intelligible content. The truth is not that the human mind is an
instrument which in its normal functioning misintérprets the
universe in such a way that only philosophers can disentangle the
1 Cf. The Analysis of Mind, especially lect. xv. “‘Mind and matter alike are logical
constructions ; the particulars out of which they are constructed, or from which they
are inferred, have various relations, some of which are studied by physics, others by
psychology. Broadly speaking, physics group particulars by their active places,
psychology by their passive places’’ (p. 307). “‘The causal laws of physics . . . differs
vfrom those of psychology only by the fact that they connect a particularwith other }
' appearances in the same piece of matter, rather than with other appearances in the_/
‘same perspective ”’ (p. 301).
2-The Intellectualism of St. Thomas, E.T., Part II. Rousselot tends to a rather extreme
view of the limitations of the human intellect through the union of the soul with the
body, and it would be inaccurate to claim his support for all the details of the position
maintained here, But he is as insistent as anyone upon the fact that the intellect is
essentially the faculty of apprehending being.
” INTELLECT AND INTUITION 85
Jal objects of its perception from the elaborate illusions which
it has worked into them and discover their true nature; rather it
is that the mind, while it is, like all human faculties, including
. the genius of the philosophers, subject to error, is nevertheless
capable of apprehending truth and normally does so.
The interpretation which we have given to the Five Ways of
St. Thomas involves that just such a power as this is native to the
human mind, though like other powers it may need training and
developing. In rising to a conviction of God’s existence from
finite beings, we do not, it was asserted, merely perceive the
existence of these beings by the senses and then, by a subsequent
process of purely logical deduction, arrive at an intellectual
acceptance of the proposition “God exists.”» On the contrary, if
our mind is in a healthy and vigorous state and is able freely to.
fulfil its proper function of apprehending finite beings as they
really are, it will, in the very act by which it apprehends them, be
capable of penetrating to the ontological depths of their nature
so as to know them as the creatures of God.
This doctrine, that the proper function of the intellect is not
merely discursive ratiocination about the deliverances of the
senses but the penetration of the inner essence of things, forms the
main theme of one of the most remarkable books on the philosophy
of religion of recent years, Fr. M. C. D’Arcy’s Nature of Belief. It
is, in principle, a restatement and an expansion of Cardinal
Newman’s teaching about the “ Illative Sense,” detached from the
unfortunate philosophical background against which Newman
1 A very thorough answer to the problem of error isgiven by M. Gilson in the course
of ch, vii and ch. viii of his Réalisme Thomiste et Critique de la Connaissance.“The-main
points of hisargument, which is strictly Thomist, are as follows ».(1)Gonerete-existence
is always of singulars; universals occur only in them (and, intentionally, in the intellect
which abstracts from them). (%) The apprehension of singulars is the function of
sensation, the apprehension of universals is that of intelléction; since it is the human
being asa whole who has both intellect and senses, he thus knows the universals as they
occur in the particulars. (3).-Jf.sensation..is.deranged, or temporarily latent, the
criterion by which existence is judged no longer operates....““The problem of the
judgment of existence . . . is linked up with the analogous problem of the apprehension
of singulars. It must be so in a doctrine for which only singulars exist.”’ (op. cit.,
p. 210.) “Since it is sensation that attests existences, we need no other existential
index than the certitude with which sensation is accompanied. ‘That a person who
does not possess the criterion should be exposed to error is only natural. So long as he
is asleep a dreamer cannot know that he is dreaming, but he knows that he has been
dreaming as soon as he wakes up. It would indeed be tragic,”’ he adds, “‘if philosophers
philosophized in a dream; the realist at least tries to avoid that.” (op. cit., p. 204.)
Cf. the rather different discussion byRousselot in The Intellectualism of St. Thomas,
E.T., p. 73 f.; he writes: “The root-cause of all human error lies in the twofold
multiplicity that characterizes human knowledge: St. ‘Thomas says so explicitly, The
cause Of erroris to be sought either in sense-knowledge, or else in that multiplicity
which-is-implied-in‘diseursive reasoning.” (op. cit., p. 81.)
86 HE WHO Is
wrote. Newman’s starting-point was the undeniable fact that
many of the beliefs about which we are most certain—such as, for
example, that Great Britain! is an island—have never been
obtained by rational demonstration, and would moreover be
practically impossible to demonstrate. Anyone who has argued
with a supporter of the flat-earth theory will have realized how
little we depend on clear logical demonstration for some of our
most firmly held convictions. The fact is that the human mind
is so constituted that it is able to extract from the whole body of
fact with which it is confronted certain particular truths by a way
quite different from those of scientific investigation or mathe-
matical deduction; “given sufficient detail the mind has the power
of interpreting what that detail signifies.”? This may involve
deriving from a number of instances of a particular truth the
truth itself as a universal property, in spite of the fact that
syllogistic reasoning could not attain it ;this would be the case, for
example, if, from the spectacle of a number of corpses we were led
to assert that all men were mortal, not because the dead men we
had seen actually comprised the whole human race, but because
the contemplation of these spectacles of mortality had given us an
understanding such as we had never had before of the inherent
corruptibility of man and had made us see how frail and uncertain
our own condition was. On the other hand, our certainty may be
derived not from a number of instances of what we afterwards
come to accept as a universal law, but from a multitude of varied,
and possibly bewildering, facts, as in the case already mentioned
of our conviction that Great Britain is an island. In the former
case we have what traditional logic calls an “induction,” in the
latter case what has been called in recent years a ‘“‘unity of indirect
reference.” But in either case we have the human mind going
beyond what it knows by formal logical argumentation and
exercising a kind of intuition.* This operation is given by
Fr. D’Arcy the name of interpretation, and the main contention of
his ‘book is that, like other operations of the mind, it is a reliable,
though, again like other operations, not an infallible, instrument
of knowledge. ‘“‘Just as the object of sense presents itself to us as
1 D’Arcy actually says “England”?! 2 Nature of Belief, p. 180.
3 The argument is in no way weakened if, with the modern logicians, we interpret
universal propositions as not categorical but hypothetical. If “‘All men are mortal”
really means “If x is a man, x is mortal,” it can only be known to be true either by
enumeration, which we have seen to be impossible, or because it is already known that
6e
humanity entails mortality,” and this involves an intuition of humanity as an
- * . . . . “Je .
essence.
» INTELLECT AND INTUITION 87
one whole,” he writes, commenting on Newman, “‘so too we grasp
the full tale of the premises and the conclusion per modum unius,
‘by a sort of instinctive perception of the legitimate conclusion
in and through the premises, not by a formal juxtaposition of
propositions.’ 1 ‘‘There comes a point when the truth shines
out, when the fact or object or meaning is manifested in its unity
through the signs.”’? “‘ There is good reason . . . for thinking that
the mind has the power of recognizing when the complexity has
reached the requisite degree for certainty,” though “‘no rule of
thumb can be given which will tell us once and for all when and
where on all possible occasions this requisite complexity is
present.”? Yet, as was said above, this power is finite; “‘nothing
. . justifies the pretension that we know exhaustively the inner
nature of reality. There are deeps within deeps,. there are
secrecies which the sense-bound mind of man cannot penetrate.’’4
Our case will be considerably strengthened if we see in more
detail the light which this view throws on the problem of induc-
tion, for that problem is one which logicians have found extremely
baffling. ‘“‘Why,” asked J. S. Mill, “‘is a single instance, in some
cases, sufficient for a complete induction, while in others myriads
of concurring instances, without a single exception known or
presumed, go such a very little way towards establishing a universal
proposition? Whoever can answer this question knows more of
the philosophy of logic than the wisest of the ancients, and has _
solved the problem of induction.”® “I am convinced,” writes
Mr. Bertrand Russell, “‘that induction must have validity of some
kind in some degree, but the problem of showing how or why it
can be valid remains unsolved. Untilit is solved, the rational man
will doubt whether his food will nourish him, and whether the
sun will rise to-morrow.’”®
It will be clear that if the ultimate Bipen of which the world
is composed are isolated and unrelated particulars there is no
justification for the principle of induction at all. If a number of
flashes in the sky, which are sufficiently similar to one another to
be all called “lightning,” are each followed by noises which are
sufficiently similar to one another to be all called “thunder,”
while these similarities are purely accidental and in no way mani-
fest any real common characteristic of the flashes or the noises,
1 Nature ofBelief, p. 137. 2 Thid., p. 187. 3 Ibid., p. 198. 4 Tbid., p. 196.
5 Logic, III, iii, 3. Quoted by H. W.B. Joseph in An Introduction to Logic, P. "400.
6 Outline ofPhilosoply, p. 14. Mr. Russell hastens to add that he is not a “rational
man,’
88 HE WHO IS
then there is no reason whatever to suppose that another flash
called lightning will be followed by another noise called thunder.
ere can be no such thing as inductionin anominalistic universe.
On the other hand, if it is possible for ‘similarities to manifest
common natures, haw can we distinguish when these cases arise,
and how can we be quite sure that, in such a case as that of
lightning being followed by thunder, there is a genuine case of
causal connection in view of which we may expect lightning to be
followed by thunder next time, and that the sequence was not a
pure coincidence or due to some other cause? There is a well-
known story of a man who, having got drunk on three successive
nights on whisky-and-water, brandy-and-water and rum-and-
water, decided by applying the method of agreement that it must
have been the water that was bad for him. And even when on
the fourth night, by applying the method of difference and
drinking whisky neat, he decided that his conclusion was wrong
and that the real intoxicant must have been some hypothetical
common constituent of whisky, brandy and rum which we may,
for convenience, designate by the name alcohol, why should we
suppose that, if drinks containing alcohol made a man drunk four
nights running, they should necessarily make him drunk on the
fifth? There is no mood of the
e syllogism that enables us to argue
as follows:
oe B, C, D contain alcohol.
(A, B, C made me drunk.
“——~ Therefore D will make me drunk.”
We could indeed draw this conclusion if we knew that ail
alcoholic drinks were intoxicants. But that Would involve our
having discovered. that D was an intoxicant, and this is the very
conclusion which we are trying to prove. It is clear that mere
enumeration of instances can never give us a logically coercive
induction.
Nevertheless, the attempt to prove that it can has been made
by more.than_ one logician. J. S. Mill argued that, given the law
cecausation, induction could ia reduced to deduction. Mr. ©
a cause; the cause cannot be C or D or E or etc., because we find
by experiment or observation that these may be present without
producing B, On the other hand, we never succeed in finding A
without its being accompanied (or followed) by B. If A and B
o INTELLECT AND INTUITION 89
are both capable of quantity, we may find further that the more
there is of A the more there is of B. By such methods we eliminate
all possible causes except A; therefore, since B must have a cause,
that cause must be A.” But, as Russell remarks, ‘‘all this is not
really induction at all; true induction only comes in in proving
the law of causation.... We are left with the problem: Does
mere number of instances afford a basis for induction?’’4
The most thorough.inyestigation of this question that has been
made is due to Mr. J. M. Keynes. He ‘‘holds that an induction
may be rendered more probable by number of instances; Hot
bécause of their mere number, but because of the probability,
if the instances are very numerous, that they will have nothing in
common except the characteristics in question. ... If we can so
choose our instances that they have nothing in common except
the qualities A and B, then we have better grounds for holding
that A is always associated with B. If our instances are very
numerous, then, even if we do not know that they have no other
common quality, it may become quite likely that this is the case.’”®
Mr. Keynes concludes that, as a consequence, without any appeal
to law or causality, it has been shown that the probability that
A and B will occur together next time increases with the number
of times that they have been observedto do so in the past. But
unfortunately it does not follow that this probability approaches
certainty as a limit or even that it becomes greater than one-
half (i.e. that A and B are more likely to occur together than
not).
To achieve this last requirement Keynes has to introduce a
postulate called the ‘“‘principle of limitation_of variety,” This, as
corrected by Nicod, demands that, for each of the objects in the
field, there shall be a finite number n which is such that there is
a finite probability that the number of independent qualities of
the object is less than n, where ‘“‘independent qualities’? means
qualities that do not cohere together in groups of invariable
connection. We might express this roughly by saying that it is
necessary that some limitation should be placed on the variety in
the universe. But it seems very difficult to see any a priori reason
why this principle should be true, and the slight indications that
1 Outline of Philosophy, p. 281. (I have taken the outline of Mr. Keynes’s investigation,
which will be found in his Treatise on Probability, from this same work.) A more recent
discussion, leading to the same conclusion, is given in Weinberg’s Examination of Logical
Positivism, ch. iv. Unfortunately this last book contains a large number of very con-
fusing misprints. | 2 Russell, Outline of Philosophy, p. 281.
90 “HE WHO Is
point that way rest on scientific data which have been derived by
the use of induction and therefore cannot be validly used to prove
that induction is itself valid.
Since, then, the attempts to set the method of induction upon a
firm basis of deductive logic have come to a dead end, in spite of
the fact that both science and life are impossible without a constant
use of the method, we may notunreasonably fall back upon
Fr. D’Arcy’s contention that the validity of
o' the inductivé méthod
depends upon an inherent power of the human mind to recognize”
the element_ofcausality innature _when i
it receives an
an adequate
/
j stimulus to do so. _he writes, “is neither by syllo-
gisticinference nor by. exhaustive enumeration, butatype ofwhat
»,Lhave named. interpre tation. The multiplication of experiments).
f
Be
& the refinementt of data, serve to eliminate the irrelevant and leave ;
“the way open for the mind to detect the essential ;once that has’
|been discovered the rest slide into their place. ae;
“It may be well to emphasize again what is the precise point for
our purposes of this discussion of induction. It is simply this: that
the validity of induction is universally accepted by modern
philosophers, since they can neither philosophize nor live without
it. Nevertheless, they find it impossible to justify it rationally on
the basis of the subjectivist and particularist theory of perception
which, in some form or another, most of them hold. On the other
hand, the theory of perception which we have been maintaining
here does provide a rational basis for induction.? This fact may
therefore be reasonably urged _as providing strong corroboration
1 Nature of Beliefs-p. 177. Whitehea ~ nificantly remarks: “‘The very batHing task
of applying reason to elicit thé’ genera “characteristics of the immediate occasion, as
set before us in direct cognition, is a necessary preliminary, if we are to justify induc-
tion ;unless indeed we are content to base it upon our vague instinct that of course it is
all right. Either there is something about the immediate occasion which affords
knowledge of the past and the-future;-or-we-are reduced to utter scepticism” as*to
memory.and.induction.” (Scvence and the Modern World, p. 61 (58).) Mr. A. J. Ayer, ©
“in expounding his mitigated logical “positivism, “tries to. sidestep the issue by main-
taining that the problem set by induction is meaningless. “It appears,” he writes,
“that there is no possible way of solving the problem of induction, as it is ordinarily
conceived. And this means that it is a fictitious problem, since all genuine problems
-are at least theoretically capable of being solved.” (Language, Truth and Logic, p. 47.)
But all that this really shows is the inadequacy of logical positivism, which denies the
“. problem instead of solving it.
2 It is not, of course, claimed that we haye solved in this chapter all the problems
connected with induction ; that is not, in any case, our business here. But it is alleged
that only on some such view as we have put forward can the problem even begin to be
fruitfully discussed. We cannot attempt in a footnote to appraise the treatment given
in Mr. W. H. V. Reade’s recent book, The Problem of Inference, but it is of interest to
note that he refuses to accept the viéw that inference is therely equivalent to the
recognition of logical implication, and that he insists that our immediate knowledge
In_perception.is not just of particulars but ofSS ae
INTELLECT AND INTUITION gi
for the brah of our theory, though it does not of course prove that
no other theory can be devised to meet this need. But we must
now return to Fr. D’Arcy.
When he passes on to consider the evidence for the existence of
God, Fr. D’Arcy’s argument is in substantial agreement with our
own, though he does not give a detailed discussion of the classical
Five Ways. He denies the validity of the argument from religious
experience in_the forms iin which
it is con amonly stated. “Its
value,’ he writes, “‘seems to me not to consist in any appeal to
feeling or interior conviction which cannot be brought before the
footlights, nor to some imaginary contact with God by intuition.
There is no need to return to the occult to escape the atheism d of
science."=.
(In passing, it must be observed that the “contact
with God by intuition’? which D’Arcy rejects is the immediate
awareness of God as the direct object of knowledge taught by
ontologism, not, as we shall see, the intuition of God’s existence
as the ground of the existence of finite beings. ) Thetruthis rather ,
that, by unity of indirect reference, God_is_ apprehended as the |:
ultimate reality upon whose creative act the world depends.
“The certainty rests on the infinite complexity ‘of the evidence
i] which is afforded by indirect evidence [query, reference?] and
given to us under the form of unity. . .... In one and the same. kind
of act itis seen that what we know has no ground or unity without _
a whole, and that we -interpret this whole not abstractly but_as a
cause Or power or Spirit. « Ss Simple people . . have a world of
discourse and a universe in which they view oan. and they have
no difficulty in identifying this unity with a God who is not the
universe but the ground of it. They interpret, that is, what they
see as the work of a spirit in an analogous way to that in which
they interpret the bodily actions of their fellows as the signs of a
human will.” The bearing of this on Garrigou-Lagrange’s
general proof, which includes the Five Ways as special cases, will
be obvious, but D’Arcy’s discussion becomes still more relevant
to our present concern as he proceeds. After remarking that the
refutation which Kant.claimed to have provided of the classical
arguments “depends upon the limitations which he himself
imposes on the extent of human knowledge,” he insists that we
must not begin with the phenomena of nature as science relates
them to us, since these are confessedly_ nothing but.appearances:
“The objects which we meet in knowledge when we are not
1 Nature of Belief, p. 259.
92 HE WHO IS
working as scientists are always sensible things, that is to say, they
are made up of a sensible and an intelligible content,” and even
‘scientists abstract from things in themselves in order to be able
to learn more about them. ... Their methods... bear out...
that we know something, that we are looking through the appear--
ances at reality.” And, he continues, ‘“‘once this is granted the
existence of God must be admitted without more ado.”’ For, in this
act of knowledge, either “we do. nothing but know the one, com-
plete,_self-sufficient_reality—and this we must~call” ‘God—or, if
we know the dependent, we are bound to assert the existence of
its source and ground,” and this latter again is God. Since,
therefore, we do not know completeness in every act of know-
ledge, what we know_is_the contingent, and the contingent as
~ “a suppliant for its existence and its meaning on what is not
—teself
Fr. D’Arcy’s conclusion, therefore, is that the alleged “Argu-_
Se hee in so far_as
itis valid, is simply
what the argument from contingency]
"y.becomes in a world in which
thought has become more and more subjective, in which men have
focused their attention less and less upon outside things and more
and more upon themselves. ‘“‘‘The modern,” he says, “has made
a discovery that in his own experience he cannot escape God,”
. but “the experience which modern writers on religion uphold is
nat a direct awareness of God but. the.consciousness oftheir own
state as creatures.” As we said on an earlier page, “the argu-
ments for the existence of God start from the recognition of finite
being, and this involves the recognition that we are finite our-
/selves.”? And we may agree with Fr. D’Arcy that the argument
' from religious experience may well turn out to be ‘“‘nothing but
the old argument from contingency looked at from inside instead
of from outside.’
:
It would, of course, be unfair to claim from another writer more
agreement with one’s own argument than can be certainly sus-
tained. In particular, we have not laid quite as much stress as he
has upon religious experience in.elation to the unity of indirect
reference. But on our two main _Points we_ have his support:
tiamely, the inherent power of the human-mind-to ‘penetrate.into
the heart of beings‘and obtain a trye, cven though a limited, grasp
of their ontological status, and’ ‘the £onsequent Possibility of an
immediate recognition thatthat’status démands as its ground the
1 Nature of Belief, pp. 258-62. 2 See p. 77 supra 3 Nature of Belief, p. 263.
CERTAINTY AND CERTITUDE 93
existence of that infinite Being whom, as the Angelic Doctor tells
us, all men are agreed in calling God.}
tee ea ee
ADDITIONAL NoTE A To CHAPTER:
VII
CERTAINTY AND CERTITUDE
It may appear that insufficient care has been taken to distin-
guish between objective certainty and subjective certitude, and
that the statement that the €xistence of God can be known with
certainty from the consideration of finite beings is in effect
annulled by the admission that the human intellect, like every
other human faculty, is not infallible. It may be replied that, while
the great majority of the beliefs upon which we rely for our daily
living are ultimately based merely upon that ‘probability which
is the guide of life,” our apprehension of the existence and status
of finite beings is of a different order altogether. The more we
examine the former beliefs the more clearly they are seen to be
inferential, while, as was pointed out on page 80 above, the more
carefully we contemplate the beings which surround us the more
their dual character as real beings and yet finite ones manifests
itself. If it is objected that dreaming or over-indulgence in
alcoholic stimulants may result in the presentation of spurious
sense-data which prima facie seem to have as much objectivity as
the objects of our normal experience, it may be replied that such
data do not, by the nature of the case, lend themselves to con-
templative inspection, nor is the person who experiences them in
a condition in which he can achieve it. Furthermore, such
1 Such a presentation of the rational approach to God as has been given above may
go some way towards meeting the criticisms made by Dr,John-Baillie in ch. iii of his
work, Our Knowledge of God. If we draw together Abbot Chapman’s insistence on the
essentially abstract nature of natural theology (p. 26 supra), Fr. Garrigou-Lagrange’s
“general proof which is necessary for every man” (p. 37 supra), and Dr. Baillie’s
doctrine of a ‘‘mediated immediacy” (op. cit., sec. 16), we may conceivably find that
the difference is not as great as appears at first sight. We may, however, suggest that
Dr. Baillie’s upbringing in a devoutly Christian family may have led him to assume an
immediacy of God to the mind of man that others would dispute. In any case, it must
be remembered that, as St. Thomas himself remarks, “faith is a kind of knowledge”’
(S. Theol., I, xii, 13 ’ad 3), but this does not make natural theology illicit. And the
distinction between the ordo esséndi and’ the-ordo cognoscendi must never be forgotten
(p. 1 supra). Cf. the short but penetrating review of Baillie by Fr. Ivo Thomas, O.P.,
in Blackfriars, February 1940, p. 123.
94 HE WHO Is
sensory presentations either have as their content imaginary
figures which have had real counterparts in the past experience
of the experient (as, e.g., rats and snakes) or else are constructed
from elements in various objects of past experience (as are, e.g.,
unicorns and dragons) ; there is thus always an ultimate appeal
to real beings.!_ The appeal is made in the text not merely to crude
awareness of the world which is our environment, but to careful
attention to its nature. We may agree in principle with Dr.
|
\
Tennant that the deliverances of “psychic”? immediacy need to
be checked by “‘ psychological”’ reflection and discrimination,? but
\i
we must insist that this discrimination must be applied not merely
| to the mechanism of perception, which (through its expression in
\ terms of the functioning of the sense-organs, nerves, etc.) itself
involves an assumption of the real existence of finite beings, but
‘also, and primarily, to the object of perception, in order that it
‘may be plainly understood what is the ontological status of the
_objects perceived.
1 Cf. p. 85, n. 1, supra. 2 See p. 162 infra.
CHAPTER VIII
GOD AND THE WORLD: ANALOGIA ENTIS
HE foundation of the argument for the existence of God
which has been built up in the preceding chapters is the
existence of finite beings which demand the existence of an
infinite Being as their ground. Indeed, the argument has been
little more than an elaboration of the Pauline thesis that “the_
; invisible things of him since the creation of the world <are clearly
seen, being perceived through ‘the ‘things ‘that are made” ;? to
adapt thé title*of a famous book by Cardinal Belicrnmne, we
have made an ascent to God by a ladder of created things. We
have proceeded from that with which we are immediately
acquainted to that which is less familiar ;we have started from the
evident fact of the existence of the world of which we ourselves
are part, and have ended with the God who infinitely transcends
it.
Once, however, we have been led to affirm God’s existence, our
whole perspective changes, and wé see that it is not God’s existence —
that requires explanation but the existence of anything else. “This
proposition ‘God _exists’,” says St. Thomas Aquinas, ‘‘of itself is
self-evident,” but “because we do not know the essence of God,
the proposition is not self-evident to us; but needs to be demons-
trated by things that are more known to us, though less known in
their nature—namely, by effects.”? The real miracle is not that ©
God exists but that the world does. “God—the self-existent, —
onperfect, changeless
¢ ‘Being, thePure Actiin‘whom allthat supremely ,
but be; but that he in whom aan is thaldae should confer ,
existence on us—that is the wonder which may well stagger our~
minds;
In the words of a modern theologian, “‘the relation between
God and his creatures is a wholly one-sided relation, in that while
the creation depends absolutely upon God, God in no sense
depends.upon.his.creation. God would be neither more nor “Test
perfect. ifthe creation dissolved into utter nothingness. The <
1 Rom. i, 20. 2 $. Theol., 1, ii, 1¢.
95
96 HE WHO IS
absolute perfection of perfect being would still exist.” This
doctrine, continues Prebendary Hanson, ‘‘is dualistic, it is
miraculous, it presents great difficulty to the human reason, and
still more to human conceit, for though the modern man gets
on very well without God and can even ‘make it his boast, he
finds the conception ‘of a God who can get on very well with-
out him highly offensive.”! As Gilson says, ““God added
nothing to himself by the creation of the world, nor would any-
thing be taken away from him by its annihilation—events which
would be of capital importance for the created things concerned,
but null for Being who would be in no wise concerned qua
being.’’?
In view of the widespread tendency even among theologians
to-day to be satisfied with a doctrine of God as in one way or
another conditioned by or dependent on his creation, it is impor-
tant to stress the absolute necessity of the conception of the entire
oe
that ihe‘only hope ot
ofexplaining the existence of finite beings at all
is to postulate the existence of a Being who is self-existent. A
first cause who was himself in even the very least degree involved
in the mutability, contingency or insufficiency of the universe
would provide no more in the way of an explanation of the
existence of the universe than it could provide itself; such a God
would providea foundation neither for himself nor for anything
else. Unless we are prepared to accept the God of classical
theism, we may as well be content to do without a God at all. If
we admit any dependence of God upon the world, the very
basis of the arguments bys which we have been led to him is
destroyed; a “first_cause’ "who is not Sseifeulecient vexplains
nothing. ~ hae. Set
_-fhe God that philosophy demands is thus nothing 1a chins the
‘€God of Exodus, the God whose name is J am. The Thomist trans-
formation of the Aristotelian deity into the God of Christian theism
is thus not a subtle and illicit importation into philosophy of con-
cepts derived from an alien sphere; it is the provision of the
Aristotelian arguments with the only conclusion that can ulti-~
mately satisfy them. If anything exists, then self-existent Being
must exist; if self-existent being does not exist, then nothing can
¢ A
y,
exist—this iis the fact of the matter. a
at “Dogma in Medieval Scholasticism” in Dogma in History and Thought p. 105.
2 Spirit of Medieval Philosophy, p. 96.
GOD AND THE WORLD: ANALOGIA ENTIS 97
This on may be made in many ways. One.way of expressing
re intstnbiny,
it is in the form of Dr. William Temple’s two quasi-mathematical
equations :1 -
a, God — the world = God,
The world—God=O; |
another way is in the dictum that the world exists not for God’s
utility but for his glory, where “glory’’ must be understood in
the theological sense of the “accidental” and not the‘“essential,”’
glory of God; yet another way is in the words of the Vatican
Council that God made the world “not in order to achieve or
increase his own happiness but in order to manifest his perfec-
tion.”’? As Thomas Jackson wrote, ‘‘ The world is created, and yet
it was not necessary that it should be created. . . . It was likewise
free for the Almighty to create or not to create man or angel; but
his free purpose to create them after his own image being sup-
posed, it was not merely possible, but altogether necessary, that |
_they should be created good.”’* We must altogether rejectthe \
“4-assertion.of. Hegel that “God without the world is not God,’’! or of, 1-4
Whitehead, that “God is completed by theiindividual, fluent “{. }
ee
Satis: act ons”~of finite fact, 5 or of Tennant that “God EST a ar Fe
\ world, or a Real other, is not-.God_but an abstraction.” Since _aea Ee
e world can find its explanation only in the creative act of a-
God who is self-existent Being, God_mustbe a creatorin the sense
that he can create if he wills to do so; but precisely because he is
self-existent Being, he need not erie unless he does so will. In
Gilson’s words, ‘‘it is quite true that a Creator is an eminently
‘Christian God, but a God whose very essence is to be a creator is
not a Christian God at all. The essence of the true. Christian God__-\
is not. to create but to.be...‘ He whois’ can also create>ifhe chooses; —_
but he does not exist because he creates, nay, not even himself; he |
can create because he supremely is,’”’
The relation that obtains between the lesen God and the
world whose existence is completely dependent upon his will is thus
absolutely unique; it has therefore a special name, which is
creation, Gpeton in the scholastic definition, ‘‘the production
1 Nature, Man and God,p. 435. A similar formulation is quoted from Coleridge by
A. S. Pringle-Pattison: The Idea of|God, P- 315.
2 Const. dogm. de fide cath., cap. 1: “non ad augendam suam beatitudinem nec ad acquirendam,
sed ad manifestandam perfectionem suam,”” (Denzinger, Enchiridion, 1783.)
3 Works, V, pp. 296-7. 4 Quoted by James Ward, Essays in Philosophy,p. 287.
(a Process
3 and Reality, p.
p 492. © Phil. Theol., II, p. 168. ? God and Philosophy, p. 88,
SRR ER eI
98 ; HE WHO Is
of the whole substance of a thing in the previous absence both of
itself and of any other subject.’’! It is literally making out of nothing,
where “nothing” is not some kind of Ungrund or formless prime
matter co-eternal with God, as much Greek philosophy taught,
but the entire non-existence of anything. We may admit that
creation cannot be imagined by us; if it could it would not be
creation, for our mental images are invariably made up out of
elements in our past experience, and to nothing in our experience
does creation, in the strict sense, apply. But what we cannot
imagine we may nevertheless conceive, as we saw in discussing
the Anselmian definition of God. Creation is the calling into
existence of the being itself, not_merély-the~imposition
“of new
qualities
upon it; it istherefore the act of one who is Pure Being
and of no one else, and this is why there is nothing that we can
do which is precisely like it.
Some light is, however, thrown upon the meaning of creation
by the secondary and improper use of the word in ordinary
speech. When a great painter or musician is described as a
“creative artist,’ or when the noun “‘creation”’ is applied to some
triumph in the world of feminine fashion, in either case what is
implied is the note of novelty. Something has been brought into
the world which is different from everything that existed before;
the opposite of creative art is art that is purely imitative, and the
word ‘creation’? would, one imagines, hardly be applied to
utility clothes made by mass-production methods. In all such
cases the element of novelty is, of course, strictly relative; neither
Michael Angelo nor Mozart nor a Parisian dress-designer can
work in complete independence of all that has gone before, and
each of them requires some material upon which to exercise his
craft. But the fact remains that in the common use of the word
“creation” the idea of novelty is uppermost; the word is there
used in a sense which bears a distant and very imperfect analogy
to its primary and proper sense, in which it signifies the absolute
novelty that is involved in the production of being from complete
non-existence.
Since, then, ‘‘creation” is not the name of a change produced
in pre-existent matter but_of the act by which finite being is
\__ posited in existence, it has nothing to do with the question whether
1 Creatio est productio totius substantiae rei ex nihilo sui et subjecti. Cf. S. Theol., 1, Ixv, 3c:
Creatio autem est productio alicujus ret secundum suam totam substantiam, nullo presupposito quod
sit vel increatum vel ab aliquo creatum.
f|
GOD AND THE WORLD: ANALOGIA ENTIS 99
the world had a beginning in time
or not.1 We have been led to
affirm the existence of God as the ground of the world’s being;
as has been pointed out more than once, the postulation of a first
cause which is first in a merely temporal sense gets us nowhere.
The existence of a world that is changing and contingent necessi-
tates the existence of a God who is by his very essence changeless
and necessary, upon whose creative fiat not merely the world’s
beginning but its continued existence depends. Therefore, while
time is the inevitable mode under which finite beings exist,? God’s
existence is of necessity supra-temporal or, to use thétechnical
term, eternal. Etcrmity does not, it must be “insisted, merely mean
“going on for ever,” nor does immutability mean remaining
the same for an infinitely long time; eternity means existence
outside time, and immutability means independence of that subjection
1 Cf. §. Theol., I, xlvi, 2. S.c.G., I, xxxi-xxxviii. De Pot., iii, 14.“Tt has already
been remarked that St. Thomas caused something of a scandal in the thirteenth
century by maintaining that it is impossible to“prove by reason alone that the- world
had a beginning in time. There is, However, a modern argument, based upon con-
temporary physical” theory, which has been claimed as demonstrating that the world
cannot have been in existence for an infinite time, or at least that at some past epoch
not infinitely distant some extra-mundane intelligence must have intervened to impose
,-Some-ordered arrangement upon it. Since it involves theepusical, quantity known as
«eee ‘this has been dignified with the name of the-“entropological argument.” It
upon the highly general statement called the genet ees co)f Thermodynamics.
Briefly the argument is that the interchange of ¢energy betwee
6énn material'ssystems results
in a progressive eyening-out of the’ heterogeneity “of ‘theuniverse, analogous to the
process of shuffling a pack of cards: "Now
OWvif the universe had existed for an infinite
time , theshuffling would be complete and the universe would be entirely homogeneous.
But it is not entirely homogeneous, therefore it cannot have existed for an infinite
time. Taken in conjunction with the theory of theexpanding universe, this suggests
a date about _10,000,000,000_ years ago for the “beginning of the world.” (See
Eddington, New Pathways im1 Science, ch. iii; cf. The Nature of the Physical World, ch. iv.)
Professor Eddington’s attitude to the argument is very reserved; he writes as follows:
“Philosophically the notion of an abrupt beginning of the present order of Nature is
repugnant to me, as I think it must be to most; and even those who would welcome a
proof of the intervention of a Creator will probably consider that a single winding-up
at some remote epoch is not really the kind of relation between God and his world
that brings satisfaction to the mind. But I see no escape from our dilemma.” He then
discusses possible solutions, and leaves the matter undetermined. (New Pathways,
p. 59f.) (Cf. the discussion of Eddington’s attitude on this matter given by Professor
L. S. Stebbing in her refreshingly realistic work, Philosophy and the Physicists, ch. xi.)
It may be added that, for similar reasons to the above, arguments for or against the
existence of God from considerations of the evolution of the universe can hold only a
secondary place in Christian theism. For evolution is concerned with the becoming,
the feri, of the world, whereas we are concerned with.a.God who is not just a causa in
Jiexi, but a causa in esse. Cf. the passage from Garrigou-Lagrange quoted on p. 38 supra.
2Tt is unnecessary in this connection to discuss the difference made by the
scholastics between time (tempus) in the strict sense, which is the mode of existence
of material beings, and aeviternity (aevum), which is the mode of existence of pure
spiritual creatures such as the angels. Note also: “‘The being of things corruptible,
because it is changeable, is not measured by eternity, but by time; for time measures
not only things actually changed, but also things changeable; hence it not only
measures movement, but it also measures repose, which belongs to whatever is natur-
ally movable, but is not actually in motion.” (S. Theol., I, x, 4 ad 3.)
H
100 HE WHO IS
to.duration which is inherent iin temporal existence.? It is even
nips neitan- shecinmene eniemenooaee Sahat
more difficult to imagine eternity than to imagine creation, for in
the relation of creation between the Creator and the creature one
etpr OCESS, _ whereas _eternity 1
is
s som
of course, true that, as we view him, God ;appearsto bereheat to”
_change; because. the universe. thathe edag?
i ‘achanging
\universe,
ne
God’s action upon it when viewed fro wit
\to bea.chang
hanging y.action. .Nevertheless, the change that isobserved
is not a change in God but.
1t in his creation
; as St.Thomas’says, in
words that have been already quoted, ‘““Newness_of movement
is consequent upon the ‘ordinance of the eternal will to the effect
that movement be not. always.”
We are thus forced to Teject. the “‘receptacle theory” of time as
taught by. Sir Isaac Newton in-his famous scholium in. the Principia.
** Absolute, true,.and mathematical time,”’ he wrote, “of itself, and
- from its own nature, flows equably without regard to anything
external.... All things are placed in time as to order of situa-
\ tion.”’8 We agree, in contrast, with St. Augustine, _ St. Thomas
i and, it may be added, the, _Relativity : that time is
~+ impossible except in connection with the things ‘that €xist in it.4
( The world, we Shall say, 1is created notin. tempore but cum tempore,
\ while God’ s existence is independent of time altogether. It is
impossible iin speaking of God to avoid the use of terms that sug-
gest time, if only for the reason that all our verbs, as a matter of
mere grammar, are either past, present or future in tense; but,
even if we cannot imagine supra-temporal or non-temporal
existence, we can conceive it by divesting the words that we use of
their suggestions of temporality. With these precautions we can
apply to God’s mode of existence such terms as nune stans and can
1 Cf. St. Augustine: “Nor dost thou by time precede time; else shouldest thou not
precede all times. But thou precedest all things past by the sublimity of an ever present
eternity.... Thy years are one day; and thy day is not daily, but To-day ... Thy
To-day is eternity.” (Conf., XI, 16, tr. Pusey.)
2 S.c.G., II, xxxvi. See p. 62 supra.
3 Quoted by Whitehead, Process and Reality, p. 97.
# Cf. Eddington, Nature ofthe Physical World, ch. iii. It may be remarked that there
is nothing in the scholastic view ‘(thotgh there-is it!the Newtonian view) inconsistent
with the modern theory that, in a carefully defined sense, each being has its own time.
But it is necessary to make.a. careful distinction between physics‘and metaphysics. See
Maritain, on “Our Knowledge of Sensible Nature” in Degrees of Knowledge, ch. iii.
It-may also be noticed that St, Augustine tends to look upon time as_an impression
~ made by a changing thing upon the mind of an observer, and™ St. Thomas as a
Se Sg inherent in the changing thing itself; *cf.5€-g.; Conf., XI, 36, with S. Theol.,
oa:
5 The phrase is St. Augusin® s: De Civitate Dei, XI, vi.
_GOD AND THE WORLD: ANALOGIA ENTIS IOI
that to him all things are present simul et semel, once for all and
ultaneously.t
a further consequence of the doctrine of creation is that there
is no real distinction between God’s creation of the world and his
preservation of it. Both are aspects of one extra-temporal act by
which the world in the whole of its temporal history receives its
existence. If we consider the fact that, but for this act, the world
would not enter into existence, we give it the name creation; if, on
the other hand, we consider the fact that, but for the same act, the
world which exists would collapse into non-being, the word that
we use is preservation. “As the production of a thing into existence
depends onthe will of God,” writes the Angelic Doctor, ‘“‘so
likewise it depends on his will that things should be preserved; for
he does not preserve them otherwise than by ever giving them
existence; hence if he took away his action from them, all things
would be ‘reduced to nothing, as appears from “AyeuSHne “There-
fore a8 it was ini the Creator's power to produce them before they
existed in themselves; so likewise it is in the Creator’s power when
they exist in themselves to bring them to nothing.”? In other
words, God’s concern with the world is not to be thought of as
relating merely to the provision of the world with its initial impulse
into being, but as an incessant and intimate care for the beings to
which he has given all that they have and all that they are. This
is a truth which the Bible is never tired of asserting. “I have
graven thee upon the palms of my hands.” “‘He hangeth the
earth upon nothing.” ‘‘Are not two sparrows sold for a farthing?
and not one of them shall fall on the ground without your
Father.’ We may compare the words in which the thirteenth-
century English mystic, Julian of Norwich, describes one of her
revelations:
*‘He showed me a little thing, the quantity of an hazel-nut,
in the palm of my hand; and it was as round as a ball. I looked
thereupon with eye of my understanding, and thought; What
may this be? And it was answered generally thus: It is all that is
made. 1 marvelled how it might last, for methought it might
suddenly have fallen to naught for little[ness]. And I was
1 Cf. the discussion of the nature of time and eternity and their mutual relation in
VET: eager Time and Eternity in Christian Thought, lect. v, vi.
2 §. Theol., I, ix, 2c. The reference is to Aug. De Gen. ad lit., 1V, xii: Creatoris
namque potentia, et omnipotentis atque omnitenentis virtus, causa subsistendi est omni creaturae
quae virtus ab eis quae creata sunt regendis, si aliquando cessaret, simul et illorum cessaret species
omnisque natura concideret.” Cf. De Pot., v, 1.
8 Isa. xlix, 16; Job xxvi, 7; Matt. x, 29.
102 HE WHO IS
answered in my understanding: It lasteth, and ever shall [last],
for that God loveth it And so All-thing hath the being by the love of
God.
‘In this Little ‘Thine I saw three properties. The first is that
God made it, the second is that God loveth it, the third, that
God keepeth it.’”
And here we have all the answer that should be needed to the .
familiar accusation that the God of classical Christian philosophy
is a distant celestial autocrat who has none of the attributes
demanded by religion.
Two great questions remain to be discussed concerning the
creation of the universe. The first is: If God is in himself the full-
ness of being, how can there be anything else that is possible of
existence? The second is: Granted that God can create a world,
why should he decide to.do.so?..
(ry) It was asserted at the ieee ni of this chapter that the real
difficulty for philosophy is not to explain the existence of God but
to explain the existence of anything else. Whatever may be their
respective natures, it might be urged, surely God plus the world is
something more than God. Creation has increased the sum-total
of existence. Hence God _without-the world must be something
less than the fullness of being.
It must be replied to this objection-that the statement, ‘‘God
plus the world is more than God,” is not true of a God whois the
unique self-existent and infinite Being, in the sense in which we
have attributed those terms to him. Of any merely.finite God,
however.great he might be, the objection would hold. The sum
of any two finite beings is always greater than either of them taken
separately. But if God is literally infinite this simply does not hold.
God and the world being of radically different orders of reality
cannot be added together.?
To add beings together we must add them in respect of their
common qualities. Two cows plus five cows make seven cows, for
1 Revelations of Divine Love, ed. Warrack, ch. v.
2 J am tempted to bring out this point by a simple illustration from elementary
mathematics and to remark that, whatever finite quantity a may represent, infinity
plus a merely equals infinity. In other words, while adding infinity to a makes all the
difference to a, adding a to infinity makes no difference whatever to infinity. And I do
not think that the non-mathematical reader will be misled by this. But, in view of the
extremely elaborate nature of the mathematical theory of the infinite and the many
points in connection with it which are still a matter of violent controversy, it would be
almost certain to mislead any reader who is a trained mathematician, and to cause
him to assume that I am merely transferring to God the notion of the mathematical
infinite in its crudest form.
GOD AND THE WORLD: ANALOGIA ENTIS 103
they are all cows. Even two cows and five horses can be added
together in a certain way, for, although they are neither all cows
nor all horses they are all graminivorous quadrupeds; and the
answer is seven graminivorous quadrupeds. We might even,
passing to an extreme case add together the four lions at the base
of Nelson’s Column, the Thirty-nine Articles and the four cardinal
virtues, and say that as a result we had forty-seven beings con-
ceivable by the human mind, though the statement would be of
no practical importance owing to the excessively small common
content. But when it is a question of adding finite beings to God,
the sum simply cannot be made. In Thomist terminology God is
not in any genus and any argument that treatshimasifhe were |
is illicit-at the start.”
+ (2)Butifwe admit that God could create a world if he wanted
~ to why should he want to? And _ if he did decide to do so, why
should it be this particular world rather than any other; why this
particular one out of all the possible worlds? This is, we may
freely assent, the question which above all others is shrouded in
mystery; it must, however, be replied that, if it is true that we
cannot give a direct answer to it, the impossibility of giving a direct
answer is itself a consequence of the position that is being main-
tained and is not an extraneous objection coming from some
independent source. We must again urge that, if the doctrine of
God which is here maintained is correct, the question does not so
much demand an answer as refuse to admit one. The_very_basis
of our argument for the existence of God is that no reason can be _
assigned why the world should exist; hence we have been led to
postulate a God who. by.a.sheer.and-unconditioned.act.of.will has
given it existence. In so far as we are able to say why a will acts
as it does, we are limiting its nature as a will. It is possible to
assign motives for the acts which our human wills perform, pre-
cisely because their freedom is limited; but even in this case no
complete reason can be assigned, for, if it could, the act concerned
would be not free but necessary. In the case of God, whose will is
supremely perfect..andwhose.freedom-isabsolute, there. is no
1 See p. 9 supra.
2 One way of stating this is to say that, when God has created the world, there are
plura entia but not plus entis, more beings but not more being. The problem raises the
= very difficult question of the doctrine of analogy ;perhaps the best exposition of this is
to be found in Penido’s Le Réle de ’Analogie dans la Théologie Dogmatique. For a vigorous
attack on the doctrine of analogy, and indeed upon rational metaphysics as such, the
reader may be referred to N. Berdyaev’s Freedom and the Spirit, E.V., ch. i. See ch, x
~——¢nfra.
104 HE WHO IS
reason whatever that we can assign. We shall indeed maintain,
against the late medieval voluntarists, that God’s will must act in
accordance with his own moral nature, but that is all that we can
say. The only being that God wills necessarily is himself; and if
we are forced to say that, in some sense, the act by which God wills
the world must be included in the act by which he wills himself,
this does not make the willing of the world necessary, for the
reason on which such stress has been already laid, that, God and
the world being of radically different orders of reality, we cannot
class them together.1 Whether the act by which God wills himself
includes thewilling of this world or of some other world or of no
(world at all, in every case no difference is made to that supreme
~act For God and the world simply do not add up.?
For this same reason
the Leibnizian-argument_that, if God is
supremely
_good, the world which hehas made must be-the best
of all possible worlds is equally irrelevant.’ For, although
different possible worlds may differ widely among themselves, in
relation to the infinite Being which is God they all have the same
character of complete un-self-sufficiency.t In other words,
1 This point does not seem to have been given sufficient attention by Professor A. E.
Taylor in his remarks in The Faith of a Moralist, I, p. 244 f.
2 Cf. the following passage from Mr. D. M. MacKinnon’s very challenging (and by
no means slavishly Thomist) paper in Malvern, 1941:
“In metaphysics . . . the question of the why of contingent being remains always
unanswered. The goal of the metaphysician was not the achievement of a theoretically
satisfying system, such as certain forms of monism have claimed to provide. . It was
the derivation of the contingent from the necessary, but the question of why the
necessary should thus have generated the contingent remained always unanswered.
And there is a sense in which always it must. For there is no necessity why God
should have brought us into being. We are utterly unnecessary to him, we cannot
make of our existence (unless we deny its character) a matter of any necessity whatever,
yet we cannot deny that through revelation the character of our relation to God is
profoundly illuminated by the disclosure in an act, that is necessary to its achievement,
ofhis relation to us.” (op. cit., p. 87.)
3 “Now as there is an infinite number of possible universes in the ideas of God, and
as only one can exist, there must be a sufficient reason for God’s choice, determining
him to one rather than to another.
‘And this reason can only be found in the fitness, or in the degrees of perfection,
which these worlds contain, each possible world having the right to claim existence in
proportion to the perfection which it involves. .
“And it is this which causes the existence of the best, which God knows through his
wisdom, chooses through his goodness, and produces through his power. . . .
** |, . if we could sufficiently understand the order of the universe, we should find
that it surpasses the desires of the most wise, and that it is impossible to make it better
than it oh (Monadology, 53-5, 90. Philosophical Works of Leibniz, Everyman ed.
PP. 12, 20.
4 With the reservations made in the note to p. 102 I am tempted to refer to the fact
that in mathematics the various finite integers bear very varied relations to one
another but that in relation to infinity they each bear the ratio zero. Thus:
fe) I 2
== 90), == 105 4s Vos rete:
°° oo ee)
GOD AND THE WORLD: ANALOGIA ENTIS 105,
whether God creates the world, or no world, or a different world,
the result in relation to his own infinite Being is the same. No
world, however perfect, can add anything to him; none of them
can increase his beatitude, but any of them can nsanifet his glory.
“All things are well done for it is he who hath done them, says
the Christian reason. It is he who has made them because it is well
done and I know why, also it is difficult to think how to do them
better, says Leibnitzian optimism”’: this is Maritain’s summary of
the matter.1. And again: “Leibniz pretended to justify God by
showing that the work which proceeded from the hands of that
Yperfect Workman was itself perfect, whereas in reality it is the
(_ radical imperfection of every creature which best attests the glory
of the Uncreated.’”?
The truth is, as a matter of fact, as Gilson has pointed out in
expounding St. Bonaventure, that, while any world made by God
is bound to be good, the very fact of its finitude involves that God
could have made a better one. ‘‘There is no conceivable world,
however perfect, about which the same question could not be
raised as has been raised about our own. If God had made a
better world, we could always ask why he has not made one still
better, and the question would never be meaningless, for no term
of the series of possible worlds contains in itself the necessary and
sufficient reason for its realization. The only solution possible to
such a question does not reside in creatures but in God, and there-
fore it escapes us... . What he has given, he has given by pure
grace, in an act of posdads which allows of no dissatisfaction; the
/ rest. ishis secret.’’? As Fr. Sertillanges puts it, “The best possible =
_world is not possible, for God. could. always. improve-on
i it.’’4 i
To many present-day Christians the view that the creation of
the world is entirely unnecessary to God will, no doubt, seem in-
consistent with the Christian revelation of God as a God of love,
who has a most intimate concern with all his creatures; it may
suggest that God made the world, as it were, in a fit of absent-
mindedness and has little or no interest in it. Thus Dr. W. R.
Matthews asserts that the ‘‘ Deus philosophorum is not the God and
1 Degrees of Knowledge, E..T., p. 278.
2 Religion and Culture, E.T., p. 40. Cf. Ossuna: Quo majus est creatura, eo amplius eget
Deo, quoted by Bremond, Lit. Hi seof Rel. Thought in France, E.T., I, p. 11.
3 Phil. of St. Bonaventure, E.T., p. 175.
4 Foundations of Thomist Philosophy, E.T., p. 127. He points out, with St. Thomas
(S. Theol., I, xxv, 5, et 6 ad 1), that to say that God could have made a better world
is not to say that God could have made this world in a better way. Cf, Garrigou-
Lagrange, Dieu, E.T., II, p. 345 f.
§
106 HE WHO Is
_ Father of our Lord Jesus Christ. Those who thought out the
system,” he says, “‘in spite of their profoundly Christian experi-
ence, did not succeed in fusing the Aristotelian metaphysic with
the Christian Gospel. We cannot believe in the Deity who
emerges from their logic, not because he is too high but because
he does not really sustain the Christian values.”! And again,
while admitting that ‘‘to maintain that God in and for himself is
not self-sufficient is, from the standpoint of traditional Christian
theology, a gross heresy,” he asserts that, in his opinion, “‘the
doctrine of the self-sufficiency of God should be rejected”” and
that “the conception of the self-sufficiency of God in and for
himself is an abstract idea which cannot be allowed to dominate
our theology without disastrous results.’’?
Dr. Matthews is, it must be admitted, careful not to put God
and the world on the same level. ‘‘When we assert,” he writes,
*‘that it is of the nature of God to be creative and infer that every
moment of time must be filled with the exercise of his creative
power, we do not equate the product of creative activity with the
Creator.... It certainly_is.impliedin—our.argument that the
being of God as personal is. dependent..upon..the-existence-of-a
created order, and that we see no way of holding the personality
of “a Déity ‘prior to creation.’ But we must,” continues Dr.
Matthews, ‘‘make two remarks upon this which will remove the
real weight of the objection. Our argument most emphatically
does not imply the eternity of this physical universe in which we
are, nor of any universe; it will be satisfied by the admission that,
in any possible time, there must be created being of some kind.
And further, we do not suggest, nor can it be inferred from our
position, that God depends on creation in the same manner as
the creation depends upon him. -Created being depends upon
God in an absolute sense. It derives its existence wholly from him.
God depends on creation only in the sense that, being what he is,
it isa necessity of his nature to create.’”?
_.To deal with Matthews’s positive assertions first, it is surprising
/to\find a Christian theologian asserting that “‘the being of God as
/ personal is dependent upon the existence of a created order.”
1 God in Christian Thought and Experience, p. 104.
2 The Purpose of God, p. 173.. He adds: “I observe that Dr. Temple, in the valuable
Gifford Lectures [sc. Nature, Man and God] which he delivered in this place, adheres to
this venerable theological tradition, with the consequence that his views on purpose
and freedom seem to me to be obscure.”
3 God in Christian Thought and Experience, p. 206.
GOD AND THE WORLD: ANALOGIA ENTIS 107
Christian theology has always held that, in so far as the being of
God as personal can be said to depend upon anything, it is
dependent upon those eternal and internal processions by which
the unity of the Godhead is differentiated into the Trinity of
Father, Son and Holy Ghost.! In any case, the real objection to
Matthews’s view is not that it equates the product of creative
activity with the Creator, but that it deprives God of that status
of self-sufficiency which alone provides an explanation for the
existence of the world and so makes creation possible. And while
we may admit the truth of the two remarks which Matthews goes
on to make, they do nothing to remove this objection. The
assertion that, in any possible time, there must be created being of
some kind is no less repugnant than would be the assertion that, in
any possible time, there must exist this particular physical universe
in which we are. Nor does the denial that God depends upon his
creation in the same manner as it depends upon him help the
matter; such a statement is equally true of the relation between
any human artist and his work. There is, however, no doubt that
Matthews is voicing a widespread opinion when he says that the
Deus philosophorum is not the God and Father of our Lord Jesus
Christ.
In a certain sense we may readily allow this contention.\,
Traditional theology has not only admitted but.has.vehemently \
}
\
asserted that the God that philosophy displays to us is Jess than the
{
I
Christian God; thisiswhy it has maintained the.necessity for our ,
f
natural knowledgeofGod to be supplemented by revelation. But it ,
has not admitted that the God of philosophy is inconsistent with the
_ God of Christianity, as Matthews implies ; and this is the real point. /
1 Needless to say, Matthews does believe in the Trinity, but he does not consider
that the sphere of life contained therein is sufficient for the exercise of personality. He
argues that personality involves the pursuit of ideals, and hence an imperfect sphere
in which to pursue them. But, since God is himself perfect, that.sphere. cannot be
within himself. ‘If thereforé-we~hold~that God is personal, we.are forced to the
conclusion that he finds in the created world, or in the creatures, the sphere, distinct
from himself, in which his ideal ends are to be attained.”’ (God etc., p. 178.) But this
view of personality as fundamentally constituted by striving, while it would be
congenial to voluntaristic philosophies such as those of Fichte, Schopenhauer and
Hegel, has little to justify it; traditional Christianity asserts that the infinite and perfect
life of mutual self-giving of the three divine Persons is far more fully personal than
any striving to achieve ideals. And, incidentally, Matthews’s view would involve that
either God will ultimately achieve his ideals, in which case he will thenceforth cease
to be personal, or else he will remain for ever personal at the cost of never achieving
his ideals. (The second of these alternatives seems to represent Matthews’s own posi-
tion, op. cit., p. 202.) Later on, it is urged that there are strong reasons for believing
that “the thought of God as personal involves us in the conclusion that there are
distinctions within the Godhead” (op. cit., p. 193). But, if Matthews’s earlier argu-
ment is sound, it is difficult to see why.
108 HE WHO Is
Is the idea of God’s self-sufficiency
inconsistent with the idea of
his Love? At first sight we might think« so, but a
a little reflection
may lead us to a very different view. In the
ment_that_the. world. is.necessary..to.. God can have two very
different. meanings. It might mean that by some inherent meta-
physical urge God has to create a world whether he wants to or
not, or, on the other hand, it might mean that God’s sphere of
action would be limited and circumscribed wit world, and
so he creates it because he wants to. The first-méariing is clearly
inadmissible ;if God had as little voluntary decision in the creation
of the world as an automatic machine has in the delivery of a bar
of chocolate, we certainly could not describe creation as an act of
love. We must_assume, therefore, that the statement that the
creation_of the world is necessary tto “God. means that without it
something would be missing from the realm of God’s experience ;;
» that is to say, that he creates it because without it his activity
“would be incomplete.
But, it must now be said, if, in creating the world, God was
simply fulfilling his own being, then creation, while it might be
an act of love, would certainly not be an act of purely unselfish
love. Even taking into account the consequences that creation -
may be foreseen by God to involve—for example, admitting the
most crudely patripassian doctrine of the Atonement—the fact
would remain that, by creating a world, God was not only doing
something for the creatures composing it, but was also doing
something for himself.1_ We may take an illustration from human
parenthood. In consenting to procreate a child, the parents are
certainly not being entirely selfish, nor are they necessarily
oblivious of the sufferings which the child may cause them; their
motives are, however, inevitably (and rightly) mixed, and among
them is the conviction that they will be, on the whole, happier
with a child than without one. As the famous controversies
between St. Bernard and Abelard and between Bossuet and
Fénelon made plain, human love can never be totally disinterested,
not even our love for God ;? but to assume that God’s love cannot
be entirely disinterested is sheer anthropomorphismi,_ “So far from
"diminishing the love shown by God in creation, the doctrine that
creation is unnecessary to God enhances it. It is precisely because
1 Cf. p. 107, n. I supra.
2 See, e.g., the discussion in Gilson, The Mystical Theology of St. Bernard, App. II,
“Abelard,” and Baron von Higel’s extremely illuminating exposition and assessment
of Fénelon in The Mystical Element of Religion, I, p. 160 f.
GOD AND THE WORLD: ANALOGIA ENTIS 109
,
creation can give nothing whatever to God which in any way
~ enhances his beatitude, that creation is an act of entire giving on
the part of God. God would not be lonely or bored or idle if we
did not exist; his life as Trinity is a life of infinite activity, of
Ane:exhaustible fallness: In creatingthe world he gains nothing for nh
‘himself; that is why creation is an act of supreme love. Yo
The classical doctrine of creation is thus both mysterious and
paradoxical, but itisfar more profound than the alternative view.
Since creation adds nothing to God, there is a very true sense in
which God has made us for our own sake and not for his; thus
St. Thomas says that “‘God seeks glory, not for his own sake, but
for ours.”! Nevertheless, just because our whole being derives
from him we can attain our beatitude only by seeking him as our
last end, and the duty of doing so is the highest law of our being.
In other words, although created things can be of no utility to
God, it is of their very essence to be for his glory; and God’s
demand that we shall glorify him is not the selfish claim of a
tyrant but an act of the most astounding condescension, ‘‘It.is
infinitely more glorious for us to have been created for the glory
of Aim_who is,” writes Garrigou-Lagrange; “than to have been
\_ created for ourselves.”? That the self-sufficient God deigns to be
glorified by his entirely dependent creatures, whose service can
add nothing to him, is the supreme privilege and honour that God
has conferred upon them. When it is asserted that either God
demands our service, in which case he must be insufficient without
us, or else he is sufficient without us, in which case he can have no
need of our service, the answer simply is that the very formulation of
this dilemma assumes that God is finite. Ofany finite being, how-
ever great, the dilemma would hold; but it does not hold of God.
“The glory of God,” it has been said, ‘“‘does not consist in
receiving something from us which will make him richer! It con-
sists rather in giving us the means of being no longer nothing.”
/< Thus did the Word himself bid the people offer oblations,”’
wrote St. Irenaeus, “‘not that he had need of them but that they
\ might learn to serve God.’ And what is true of the glory which
“man, as a rational being, can give to God by his rational service,
is in its measure no less true of the glory which lifeless or sub-
rational living creatures give to God by the mere fact of their
existence.
1 §. Theol., 11 Il, cxxxii, 1 ad 1. 2 Dieu, E.T., p. 106.
ap: Charles, S.J.> Prayer for All. Times, E.T., II, p. 60.. * Adv. Haer., iv, 18.
110 | HE WHO IS ios
; If it is true, then, that in one sense God has made us not for
/ his sake but for ours, there is also a sense in which it is equally~
( true that we exist not for our own sakes but for his, for the value
‘of our service to God is set up precisely by this fact—and by this
fact alone—that God is ready to accept our freely willed offering
of it. We can find an illustration drawn from the realm of human
life which, however inadequate it may be, will help to make the
point clear. Of all the attributes of God the one which the
modern world is most ready to admit is that of fatherhood. Now
most fathers receive presents from their small children on their
birthdays, and receive them gladly, in spite of the fact that the
presents are usually quite useless and in any case have to be paid
for by the parent in the last resort. They are none the less readily
accepted because of that, and the normal human parent has a joy
in receiving such a gift which far exceeds the satisfaction obtained
froma much more expensive and useful present given by a business
client or even by a grown-up friend.
We must maintain, then, that God created the world by an act
of love for which no reason can be assigned except the free opera-
tion of his creative will. ‘‘God made the cosmos,’ wrote Abbot
Chapman in one of his letters,
““Why? It was’a very odd thing
to do'”’! And we shall never attain to that humility and wonder
which are essential to the true practice of Christianity unless we
realize that creation, so far from being what might reasonably be
expected of God is ae: most incalculable and, we might even say,
superfluous expression of the complete freedom and limitless
fecundity of self-existent Being. That creation is eminently
congruous with God’s nature we may readily admit—bonum est *
1 Spiritual Letters of Dom Fohn Chapman, p. 207.
2 It will be seen from what has been written above that there is a very close connec-
tion between the metaphysical problem of the existence of the universe and the moral
and ascetic problem of the duty of man to love and serve God. Mr. John Burnaby
has discussed it most interestingly in the Introduction, on “‘The Embarrassment of
the Aniti=mystic,2?.to his study of St. Augustine’s religious teachings: Amor Dei» He first
of all shows how Professor Macmurray emphasizes man’s duty of serving God to the
extent of making God for all practical purposes completely immanent in the world-
process, while, on the other hand, Professor-Nygren lays such an unbalanced stress
upon the transcendence of God as to deny that it is possible for man to love God-at all.
The Catholic. doctrine,.as expounded by Augustine, lies between these~extremes. It
insists as strongly as Nygren on the transcendence of God and as strongly as ; Macmurray
on the duty of man. But it insists that this transcendent God does condescend to be
served by his creatures, and that the duty that man has to fulfil finds its end not just
within the world-process but above it, in God. The heart of the answer is that man’s
ability to love God, like his very existence, is a free gift from God himself. (See
Burnaby, passim; J. Macmurray, Creative Society and The Structure of Religious Experience;
A. Nygren, Agape and Eros, E.T., especially Part II.)
BGOD AND THE WORLD: ANALOGIA ENTIS II!
diffusivum sui; but that it is in any way necessary to him we must
emphatically deny.
The doctrine that the world is unnecessary to God, and the
various corollaries that this involves, do not in the least imply that
God is uninterested in his creation, though it has often been
asserted that this consequence must follow from it. Thus Dr.
Matthews writes: “The main object of the Scholastic thinkers was
to disengage the conception of love in God from the taint of
*passio’, from the suggestion, that is, that God needs anything or
can be affected by anything outside himself’’; and he alleges that
“this is surely very near to a rejection of the belief that God loves
the world or human persons at all.”! This objection, like the
previous ones, rests upon an implicit denial of the absolute
infinity of God. There is no need whatever for us to overlook or
minimize the truth which the Bible throughout so plainly teaches :
that God enters into the most intimate details of the life of all his
creatures, that he rejoices in our happiness and sympathizes with
our sorrows, that he is glorified by our good acts and grieved by
our sins. We will go further and say that the intensity with which
our actions as personal beings affect him is infinitely greater than
that with which they affect our fellow human beings, for God, as
our Creator and Preserver, is present to us more closely than we
are present to ourselves.2, But when that has been said, it must
be added that even this is infinitely surpassed by the beatitude
which God enjoys in the interior fullness of his own divine life;
which it therefore can neither augment nor diminish. Therefore
“there is no incompatibility between the compassion and the_/
\impassibility of God.
Closely connected with this is the much misunderstood doctrine
that_God loves. things in proportion
to” théei® g@oodness.? Dr.
Matthews objects that ‘‘a love nicely proportioned to the merit of
the object seems too coldly reasonable to engage our admiration,”
it “falls short of the best human devotion.” We must take into
account, however, the fact that God confers upon every being that
exists all that it has and is, and that he preserves it in being,
however much, in its own freedom, it may violate his will. Behind
the act of love by which God loves it in accordance with its good-
1 God etc., p. 227. ; ; ‘
2 Solovyev is right when he says that “‘to assert that God does not ‘interest himself? }
in our material wants is to justify atheism by putting limits to the Godhead.” (God, /
Man and the Church, E.T., p. 44.)
3 Cf. S$. Theol., I, xx, 3. 4 God etc., p. 228.
112 HE WHO IS
ness, there lies the far greater act of love by which he has called it
sean soc’
into being from non-existence, an act which is in the fullest sense
of the word gratuitous, since before it existed it could have no
claim upon him at all, and while it exists can have no claim upon
him for its preservation. The fact is that not only does God love
whatever goodness we possess but he has himself conferred it upon
us. ‘‘Since our will,’ writes St. Thomas, ‘‘is not the cause of the
' goodness of things, but is moved by it as by its object, our love,
whereby we will good to anything, is not the cause of its goodness ;
but conversely its goodness, whether real or imaginary, calls forth
our love . . .: whereas the love of God infuses and creates good-
/ness.”! “There is only one reason,” writes Mr. Charles Williams,
( “why anything should be loved on this earth—because God loves
“6 '2 ae
: 0 sum up the discussion of this chapter, it has been maintained
that nothing less than a strictly infinite God can provide the
explanationof the world’s existence, and that, in consequence, the
world must be in the fullest sense, contingent and altogether
unnecessary to God. Various objections have been considered,
which all, in one way or another, complain that, on such a view,
God could not have the intimate interest and concern with his
creatures that is manifested in the Christian Religion. To this we
have replied that, while this would certainly be true if God was a
finite being, it is not true if God is infinite. In.other words, the
doctrine of an infinite God not only raises the difficulties, but also
solves them.3» The doctrine of a finité God, on thé other hand,
neither raises nor solves them ;but nor does it answer the question
which clamours for a reply: why anything exists at all.
And in the course of this discussion we have touched upon
various aspects of the relation of God to the universe.*
1 §. Theol., I, xx, 2c. 2 He came down from Heaven, p. 141.
3 It may be suggested that the failure to realize this truth is the main defect in
Professor John Laird’s very stimulating Gifford Lectures on Theism and Cosmology.
Cf. especially lectures iii and iv of that work.
4 The question as to whether creation adds anything 'to God is obviously similar to
the question as to whether, in the Incarnation, the humanity of Jesus adds anything
to the Divine Word, and, mutatis mutandis, the same answer is to be given. The
Incarnation involves, on the part of the Divine Word a human life that is every whit
as real and concrete as that of any man or woman who has ever lived; nevertheless,
since gua human it belongs to the created order, it cannot “add up with” the fullness
of being which the Word enjoys as the eternally begotten Son of the Father. In saying
this, we are not, of course, forgetting that the relation of hypostatic union set up
between the Word and the world by the Incarnation is essentially different from the
relation of “presence of immensity” which God has to the world through the act of
creation,
THE THEISM OF JAMES WARD 113
AppITIoNAL Note B To CHAPTER VII
GOD AND THE WORLD IN THE THEISM OF JAMES WARD
This is a convenient point at which to insert some remarks upon
the view of the relation between God and the world expounded in
the teaching of the late Professor James Ward, since Ward’s
reputation stands sufficiently high, through the magnificent refuta-
tion of nineteenth-century anti-theistic doctrine which he gave
in his first set of Gifford Lectures on Naturalism and Agnosticism, for
any divergence from his teaching to need some justification. In
almost the opening paragraph of the discussion of theism which
forms the second part of his later Gifford lectures, The Realm of
Ends, or Pluralism and Theism, he wrote asfollows:
“There are objections to all attempts to proceed altogether
a priori. It seems obviously puerile
to ask, for example, for a
sufficient reason why there is something rather than nothing.
_ This notion of being absolutely thoroughgoing, of building up a
metaphysic without presuppositions, one that shall start from
nothing and explain all, is, I repeat, futile. Such a metaphysic
has its own assumption, and that an absurd one, viz., that nothing
is the logical prius of something. Well at any rate, it may be said,
if we must start from something, let us at least start from what is
absolutely necessary, or rather let-us not stop till we reach it: let
us not rest in what is merely actual, for that can only be con-
tingent. But, paradoxical though it may sound, necessary being
is but another aspect of contingent being; for within the limits of
our experience only that is called really necessary which is inevit-
ably conditioned by its cause, and is thus contingent on this, that
is to say, follows from it.””4
Now, if by “attempts to proceed altogether a priori’? Ward
meant such argumentation in complete independence of all basis
in existent being as we find in St. Anselm, or in later days in
Descartes and Hegel, we might well agree. It is, no doubt, meta-
physically absurd to assume that nothing is the logical prius of
something. And it would be obviously puerile to ask why there
was something rather than nothing if the being about which the
question was asked provided the reason for its own existence.
1 Op. cit., p. 225.
114 HE WHO IS
Such a being all men, as St. Thomas would say, would unite in
calling God. But Ward condemns equally the argument which
refuses to rest in what is ‘‘merely actual’ (by which presumably
he means actual and finite) and ‘so only contingent; and he con-
demns it on the ground that, in our experience, necessary being
is necessary only in the sense that it is necessarily contingent on
some anterior cause. But, we must reply, classical theism has
never asserted that God is necessary in this sense; on the contrary, ~
iat is necessary precisely because he depends on nothing but him-
elf. The truth is, not that necessary being is but another aspect
/ of contingent being, but that contingent being, when it exists,
[oe necessarily the result—not, be it observed, the necessary result—of a oe
' creative act on the part of necessary Being._That is to say, given
\ the world, it necessarily follows that a God exists as its Creator,.but, ,
\¥ given God, it does. not necessarily. follow.that.a-world.exists~as- his.
creature. Itis this distinction that Ward overlooked.
After expréssing his agreement with the Kantian objections to
the traditional arguments for the existence of God, Ward rightly
remarks that there is nothing in our experience comparable with -
creation in the strict sense of the word. “The idea,” he says, “‘is
in fact, like the idea of God, altogether transcendent.” And he
immediately adds: “It is impossible therefore that experience
should directly give rise to it at all.”’ “But,” he continues in the
following paragraph, “it has been urged, the universe cannot
have existed for ever, since in that case, at any assigned moment,
an infinite time would be completed, and that is impossible. The
universe must then have had a beginning and so must have had a
First Cause.” And to this he replies that, “‘if this argument were
valid, it would apply equally to the existence of God. . . . Keeping
within experience we can only endlessly regress with no prospect
of ever reaching the beginning or of forming any concept of what
it was like.’ ae Ward failed altogether to understand the
issue. For, in the ace, so far from asserting that the world
cannot have existed es ever, the traditional position, at any rate
as it is stated by St. Thomas, refuses to admit that there is any
reason, apart from the Christian revelation, for supposing that the
universe had a temporal..beginning at all. And, in these second
place, the very essence-of that position lies
in its denial that-God
is merely the first: termin-acausal-series.. With every bit as much
emphasis as Ward it affirms that such a God would provide no
1 Realm of Ends, p. 232.
THE THEISM OF JAMES WARD 115
explanation of the existence of either the world or himself. For
this very reason it insists that he is outside and above the order of
the universe and is radically different in status from all the beings
that compose it.
The result of this approach is that Ward, while he is willing to
admit, at least as extremely probable, the existence of a God who
makes a cosmological unity out of a universe that is ontologically
a plurality, and indeed goes on to say that “the idea of God would
. . . be meaningless, unless God were regarded as transcending the
Many, so that there can be no talk of God as merely primus inter
pares,” is led to assert that “‘a God that was not a Creator, a God
whose creatures had no independence, would not himself be really
a God.” He develops this thought as follows:
“No theist can pretend that the world is coordinate with God:
the divine transcendence is essential to the whole theistic position.
No theist again assumes that creation involves external limitation.
But the point is that if creation is to have any meaning it implies
internal limitation. It is from the reality of the world that we
start : if this is denied, the divine transcendence becomes meaning-
less, nay, God, as the ideal of the pure reason, sinks to a mere
illusion within an illusion. On the other hand, if the reality of
Pa oS
the world be admitted, then this reality stands over against the ,
reality of God. God indeed has not been limited from without
but he has limited himself.”’? ~
The reply to this will now be obvious: it has already been given
on pages 102-5. But it may be remarked, in concluding this note,
that Ward never seems to have realized that there was any other
—alternative to his own pluralistic theism than some kind of
monistic absolutism. For him Reality was idential with Experi-
oo
ence, and the essence of theism was the doctrine that ‘“‘ beyond the
universe of the Many there is a single transcendent experient,
who comprehends the whole.” The doctrine of analogia entis* was
quite absent from his thought. Either, he implies, being is
univocal, and then God must be limited by the world which he
has made, or it is equivocal, and so in comparison with God the
world has no real existence. In following the former course, he
was, we may admit, choosing the lesser of two evils, but in it there
lies the one really unsatisfactory point in the whole of his system.
1 Realm of Ends, p. 241. Ibid., p. 243. 3 Ibid., p. 228. See p. 126 infra.
CHAPTER IX
THE DIVINE ATTRIBUTES
T is not the purpose of this book to give a complete and
systematic exposition of natural theology, and only those points
have been dealt with in detail which are directly relevant to
its main theme. It seems worth while, however, to devote a
chapter to some brief remarks on some points to which so far little
or no reference has been made.
It has already been said that, in spite of his assertion that
p
/
natural theology is ‘competent to emonstiate not what God is, but ~
that he is, St. Thomas Aquinas deduces from the fact of God’s
~ existence a remarkably large numberof facts about him.’ He
assumes that, although we cannot obtain by the exercise of reason
an essential or quidditative knowledge of God, we can know
certain of his attributes, by establishing first of all the fact that in
God essence and existence are identical and then deriving the
consequences of this identity. This has already been discussed;
here we shall merely offer some observations upon the problem ot
the divine attributes. :
The divine attributes have been defined as ‘“‘those absolutely
simple perfections, unmixed with imperfection, which exist neces-
sarily and formally, though in a higher mode, in God.’”? For the
sake of clearness it will be well to comment on some of the terms
contained in this form of words.
| (i)“ Absolutely¢ simple perfections; unmixed with imperfec-
“tion,” Perfections are indeed found in created beings, but’ they
are always limited. Creatures are good, but their goodness is
finite; they may know, but they are not omniscient. God, on the
other hand, is not merelygood, he is Goodness itself; he is not
merely. true, but isthe Truth, |
(il). “Which. exist “necessarily. The attributes of creatures are,
at least for the most part, contingent; they can be increased and
diminished, they can even be acquired and lost, without the being
which is their subject ceasing to be the same being. I can have
1 p. 69 f. supra.
2 Phillips, Modern Thomistic Philosophy, I1, p. 307. We shall not discuss the distinction
an the entitative and operative (metaphysical and moral) attributes of God (op. cit.,
ad loc
116
THE DIVINE ATTRIBUTES 117
my hair¢cut, I can become wiser or more foolish, without ceasing
to be myself. But-in-God_nothing isaccidental ;he is all that he —jx
oe a seewen oe Yous
Be bow”
Gu. 2k and. formally.” “Formally, that is as opposed to
rtually. Skegness is healthy, not in the sense that it enjoys
health but that it causes it; it is healthy virtually, Its residents,
presumably, are healthy formally, for they enjoy the health that
Skegness causes. Now--God-is.not.only.the cause of goodness,
beauty.andsso-on;he-is-the-Supreme Good and Beauty itself.
vefiv)“..,.. though in a higher modex”_Perfections éxist’in God,
ile
hnot-merely formaliter,but formaliter. Sa Conn amen
must, in some way, mean the same as goodness in his creatures;
otherwise, since our only immediate idea of goodness is derived
from the world of which we are part, the statement that God is
good would be simply meaningless ;we might as well say that God
was bad or that he was pink. But, as the Fourth Way of St.
Thomas insisted, goodness in God must be realized in an essen-
tially different and superior mode to that in which it is realized in
us. The_word eminentius expresses this fact, but it does not tell us
precisely what
wl that mode is, for it must clearly exceed the capacities
of our understanding. All wecan really say isthat God’s goodness isi
related to his infinite Being in a similar way to the way in which
our goodness is related to our finite being. This last statement is
ality ; this is not the place for a full discussion of the extremely sf
‘difficult problem of analogy, but, for the sake of accuracy it is
well to register the fact.1
The main problem in connection with the divine attributes is
how to reconcile their multiplicity with the simplicity of God.
Mr. G. K. Chesterton has a story about a lady who, after spending
some time on the Thomist exposition of the simplicity of God, was
heard to remark in despair that, if that was his simplicity, she
wondered what his complexity was like! She might have been
comforted to discover that there is no complexity in God, but it
must be admitted that, in view of the multiplicity of the divine
attributes—unity, goodness, truth, eternity, omnipotence, omni-
science, immutability and so forth—it certainly looks as if there
1 The doctrine ofanalogyisdiscussed at length in Penido’s book, Le Réle de I’Analogie
dans la Théologie Dogmatique. See also Garrigou-Lagrange, Dieu, E. Te II, p. 203f. The
Rev. W. W. S. March has pointed out how Bishop Berkeley was forced to adopt the
doctrine of analogy in order to provide a rational reply to the deists of his day. (See
article, ‘Analogy, Aquinas and Bishop Berkeley,” in Theology, June 1942.)
118 HE WHO IS
were. The difficulty is to steer a middle course between the
extremes of either, on the one hand, stressing the simplicity of
God in such a way as to deny that the various attributes mean
anything inherently different from one another, thus reducing
them to the level of purely subjective impressions made upon our
minds by the concept of God and having no objective counter-
parts in God himself, or, on the other hand, insisting on the
distinction between the attributes to the extent of depriving them
of any real common unity and of making them merely accidental
to the divine Being. How, it will be asked, can we assert that the
attributes are distinct from one another and yet all identical with
the divine essence? A simple illustration may take us some way
towards the answer, though, like all such parallels, it will land us
in the gravest error if it is pressed beyond the strict limits within
which it is applicable. A building may present quite different
appearances when it is viewed from different aspects, and yet
these appearances all cohere in the unity of the building. Their
differences do not mean that they are nothing but subjective
hallucinations ; they, all of them, each in its particular and partial
way, give a real knowledge of the building itself. Nor are they
disconnected from one another, for they are elaborately and
intricately connected by the laws of solid geometry and perspec-
tive; this can be seen from the fact that it is possible for the views
that we should describe as the east and south elevations of Lincoln
Cathedral to cohere in one object, while it is quite impossible for
the east elevation of Lincoln Cathedral to cohere with the south
elevation of the Temperate House in Kew Gardens. In other
words, the attributes of God do represent God, and are not merely
misleading impressions made by God upon our minds; while at
the same time they cohere and coalesce in the complete and
absolute simplicity of the divine essence. A more picturesque
illustration is provided by the story of the blind men who set out
to investigate the nature of an elephant. One of them walked into
its side and concluded that an elephant was a kind of wall; one
ran up against its tusk and asserted that it was a kind of spear;
one caught hold of its trunk, and said that obviously it was a
species of snake; while the fourth, who grasped the tail, said that
‘the others were all wrong and that obviously an elephant was
nothing but a piece of rope. The various characteristics which
they attributed to the beast were not subjective hallucinations;
they were genuine properties of the elephant. But their mistakes
THE DIVINE ATTRIBUTES 119
arose through their not understanding that such widely different
characteristics were all united in the elephant’s essence. Where the
illustrations break down is, of course, in the fact that the co-
existence of the various attributes in God is not due to any kind of
structural relatedness similar to the laws of geometry in space, but
to the unimaginable fullness and interior fecundity which is of
the essence of Absolute Being.t
A word on certain of the attributes may not be out of place.
It has frequently been alleged that a radical omniscience on the
part of God is incompatible with human freedom. If God knows
to-day what I shall decide to do to-morrow, how can mychoice be ~
free?) And if mychoice isgenuinely |free; how can God know now
What Lam going to do? ‘The controversies around these questions
raised by the Molinist doctrine of scientia media must surely repre-
sent one of the most elaborate and unnecessary discussions in the
history of philosophy.?. And if the questions are posed in these
terms it seems impossible-to-give-a-satisfactory- answer, for the
simple reason that the very posing of the questions introduces an
error into the discussion. For, if God’s existence is outside time,
it is strictly meaningless to talk about what God knows to- day,
God’s “to-day”? is eternity. It istrue to-day that God TEnowe
what I shall do to-morrow, but it is not true that God knows it_
~—lo=day: And so the question falls to the ground. But if, as a con-
cession to the limitations of the human imagination, we feel
een
1 Some néminalists taught that the distinction between the divine attributes was C
purely. nominal (so that, e.g., to them. Justice and mercy in God were pure synonyms) ;
Duns Scotus; in reaction, taught thatit was formal and actual..Cajetan, summing up the }
Thomist tradition, teaches that the attributes are formally in God, but that their ™ :
distinction is not formal but virtual. (See Garrigou- ee a Le Sens du Mystére et le
Clair-obscur intellectuel, p. 211 £.; cf. Dieu, E.T., I, p. 190f.
Dr. Edwyn Bevan, we may note, in his Ssymbolism and Belief, having stated correctly
the Thomist doctrine that perfections are formallyin God (p. 314), later on assumes that
this involves that the distinction between the attributes is purely nominal (p. 320 f.).
It is hardly surprising that he feels impelled to reject the doctrine of analogy altogether.
But the form of symbolism which he constructs after his resurrection of Dean Mansel
does not appear to be a satisfactory substitute (lect. xiv). His final argument for theism
clearly leads to nothing more than the idea of God as a regulative principle for human
life (lect. xvi), until in the last sentence of his work he suddenly asserts that “what
actually causes anyone to believe in God is direct perception of the Divine”’ (p. 386),
a surprising assertion in view of his earlier statement that ““even among religious
people mystics are only a small minority” (p. 346). Cf. the review-article by Bernard
Kelly, at S.D., in Blackfriars, December 1938.
2 taught-that God_has a special kind.of knowledge, scientia media, of the way
in whic “tech particular free agent will make his free choice in every possible situation,
and that this knowledge, whose object is the conditional future or futuribile, is speci-
fically different from his scientia simplicis intelligentiae, whose object is the pure possible,
and from his scientia visionis, whose object is the really €xistent,” See Garrigou-Lagrange,
Dieu, E.T., I, p. 59 f. A quite admirable discussion of divine foreknowledge is given
by Thomas Jackson, Works, V, p. 83 f.
120 HE WHO IS
obliged to speak as if God’s knowledge was in time, we shall be
compelled to say, however reluctantly, that God does to-day know
my future free choices, for the alternative is to deny God’s
omniscience tout court. But then we shall have to add quite simply
that there is a mystery beyond our power to unravel.
As regards the attribute of omnipotence, it has frequently been
remarked in recent years that, in the original Greek and Latin
Creeds, the words Tavtoxpértoop and omnipotens do not possess
the same nuance as the. English “word -almighty ;“that they assert
Godtobe theRuler ofallthings or powerfulin all matters, rather
than to be able to do anything. Such a contention may be
“admitted asa matter of etymology, for the simple reason that the
terms of theology are normally drawn from the vocabulary of
human speech and that, of no human being, however exalted,
would it be asserted, except perhaps in addresses which were
exaggeratedly flattering and insincere, that he was literally able
to do anything whatever. It must not, however, be concluded
from this that the early Church would have been ready to admit
the possibility that God was limited in his power. Nor is such a
limitation consistent with the argument that has been here
developed. If God is in any absolute sense the Ruler of all things
and powerful in all matters, if creatures depend upon God’s
incessant creative action for all that they have and all that they
are, if he is the very source of their existence and nature, then it is
absurd to suppose that his power over them islimited byanything
outside himself.” “Butthis must not be taken :as' denying that eee
1 For a discussion of the reconciliation of divine omnipotence and omniscience with
human freedom, the reader may be referred to Garrigou-Lagrange, Dieu, E.T., II,
p. 268 f. For some reflections on the historical and sociological bearings of the problem
see Maritain, True Humanism, ch. 1, and Freedom in the Modern World. I£.we maintain
/ that the Thomists are.right against-the. Molinists,.it.is.not because.they. provide a more
f complete answer (which is doubtful), but because they refrain from distorting the
question in order to make it more easily tractable.
Thomas Jackson, writing against the Calvinists, states the issue.admuirably :
“There is a fallacy, though the simplest one that ever was set to catch any wise man,
wherein many excellent wits of these latter ages, with some of the former, have been
pitifully entangled. The snare, wherein it were not possible for any besides themselves
to catch them, they thus frame or set: “Whatsoever God hath decreed must of necessity
come to pass: but God hath decreed every thing that is: therefore every thing that is
comes to pass of necessity. All things-are.necessary, at least in respect of God’s decree.’
The extract or corollary whereof, in brief, is this: ‘It is impossible for aught;that is not,
to be; for aught that hath been, not to have been; for aught that is, not to be; impos-
sible for aught to be hereafter, that shall not be.’ But if it be (as here I suppose) very
consonant to infinite wisdom, altogether necessary to infinite goodness, and no way
impossible for infinite power, to decree contingency as well as necessity; or that some effects
should be as truly contingent as others are necessary ;a conclusion quite contradictory
to that late inferred will be the only lawful issue of the former maxim, or major pro-
___THE DIVINE ATTRIBUTES 121
are_cérté ain limitations imposed upon God’s action by his own
nature; so that, for instance, he Cannotdecide-that-two
plus-two—
should be equal to five, or that cruelty should be a virtue instead
of a vice. For he is, by his very nature, the supreme Truth and
the Sovereign Good, and any violation of these-attributes-would
be not a perfection but a deficiency. St. Thomas neatly expresses
this in the dictum: “It is better to say that such things cannot be
done, than that God cannot do them.’ It is, indeed, one of the
marks—and, we might suggest, one of the strong points—of a
philosophy whose ultimate category is that of being, that it refuses
position matched with a minor of our choosing. Let the major proposition stand as
it did before, ‘Whatsoever God hath decreed must of necessity come to pass,’ with this
additional, ‘Nothing can come to pass otherwise than God hath decreed it shall or
may come to pass’: the minor proposition, which (if our choice may stand) shall be
consort to the major, is this, ‘But God hath decreed contingency as well as necessity’ ; or, ‘that
some effects should be as truly contingent as others are necessary : therefore of necessity
there must be contingency, or effects contingent.’ The immediate consequence
whereof is this: “There is an absolute necessity that some things which have not been,
might have been; that some things which have been, might not have been; that some
things which are not, might be; that some things which are, might not be; that some»
things which shall not be hereafter, might be; that some things which shall be here-
after, might not be.” (Works, V, p. 88.) And again: ‘‘Free it was for me to have
thought or done somewhat in every minute of the last year, whereby the whole frame
of my cogitations or actions for this year following might have been altered; and yet
should God have been as true and principal a cause of this alteration, and of every
thought and deed thus altered, as he is of those that de facto are past, or of that which
I now think or do.” (Ibid., p. 93; italics in original.)
This might almost be an expansion of the following passage in St. Thomas:
“Since then the divine will is perfectly efficacious, it follows not only that things
are done, which God wills to be done, but also that they are done in the way that he
wills. Now God wills some things to be done necessarily, some contingently, to the
right ordering of things, for the building up of the universe. Therefore to some effects
he has attached necessary causes, that cannot fail; but to others defectible and con-
tingent causes, from which arise contingent effects.”’ (S. Theol., I, xix, 8c.) 2
Two vital points have often been overlooked.in discussing the question: (1) that we
are concerned: with the relation of two voluntary beings—God and man—and not
with two lifeless mechanical forces which-could be compounded by a psychological
counterpart_of the parallelogram-law; (2) that because God’s creative act is at the
ontological root of the being of all his creatures it is possible for him both to conserve
their freedom and to maintain his own sovereignity, in the most complete and
harmonious balance. But that there isa mystery in all problems in which the will is
concerned we need not deny; indeed it is quite essential to assert it. (Cf. p. 103 supra.)
1 Whitehead has asserted that “‘it is perfectly possible to imagine a universe in which
any act of counting by a being in it annihilates some members of the class counted
during the time and only during the time of its continuance. . . . [There follows the
well-kndwnstory of the counting of the members of the Nicene Council.) Such a
story cannot be disproved by deductive reasoning from the premises of abstract logic.
We can only assert that a universe in which such things are liable to happen on a
large scale is unfitted for practical application of the theory of cardinal numbers.”
(Encyc. Brit., 14th ed., XV, p. 88.) But the mere fact that Whitehead describes such
an occurrence as “annihilation” involves an appeal to the accepted laws of arithmetic
as true. (Cf. Ayer, Language, Truth and Logic, p. 96, and Maritain’s remarks on non-
Euclidean geometry, Degrees of Knowledge, E.T., p. 201 f., especially p. 205, n. 1.)
28. Theol., I, xxv, 36.
122 HE WHO IS
to exalt one attribute or operation of God above another. To the
See who 4 taught..that..the..formal.. mnstituent of God was
eoieacent
and that will was essentially superior to intellect, it'was
inieity[ 22
to saythat the moral lawrested simply onmythearbitrary
decree of God and that ac ions,are_good. because.God has. cor
manded them; to the The
that was Tinduncnal ‘with the>necessary corollary Tar thé tio
law isneither anantecedent prescription’ to which God ds
i “bound
gated by him in an entirely arbitrary and capricious manner, but.
something inherently. rooted.-in-the-nature-of-man=as“reflecting
in himself, in however limited and finite a mode, the character
of the sovereign Good from whom his being is derived. The.moral
~Taw_is thus in its essence neither antecedent nor consequent to
“God; it is simply the expression of his own self-consistency. To
ff” SAYynHERE? ‘that God is bound by itis “frerety*to a ‘from‘o‘one
particular angle,.that God is God. —
"We may remark, i sing, that the contrast.generally drawn
‘between Thomist {ntellectualism and Siti vpbtnisae oe not
- wholly exact. The Scotist tradition is no doubt voluntarist in
spirit, and inthis senséScotus-may-be-considéréed asthe ancestor
of Kant, Fichte and Schelling. But, with all the stress that it lays
upon the intellect, the fundamental concept of Thomism is not
intellect but being, not truth but unity. For this very reason it is
able to preserve a balance between intellect and will, between the
true and the good, which otherwise is all too easily lost.1 And it is
not altogether fantastic to surmise that the belittling of the
intellect in comparison with the will, which has been so marked
a feature of post-Kantian philosophy, may be closely connected
with the rise in our own day of an attitude to life which is in effect
an idolatry of arbitrary power.
When we have said, however, that, because he is God, God
cannot do anything that isi in itself ether arbitrary or*immoral,
we have not denied the possibility of miracle.” Heré;asiti_other
matters, we must adhere closely to the doctrine of a God who, in
relition-10 the world, is” both” independent~of-it-and_yet_is
its
Creator,....he. idea of a miracle is essentially the idea of an
1 The extraordinary intricacy of the Thomist presentation of the structure of the
human act, as it is, for example, expounded in Gardeil’s La Crédibilité et I’Apologétique,
ch. i, while it will no doubt appear to many as unnecessarily subtle, is the result of a
determination to give due weight to both the intellectual and the voluntary elements
and to exalt neither at the expense of the other.
THE DIVINE ATTRIBUTES 123
occurrence which is a break with the order of nature, and this
involves necessarily that there must be an order of nature to be
broken with. We cannot defend the possibility of miracles by
arguing that nothing at all exists except by the deliberate creative
act of God and that miracles are no more surprising than any-
thing else, though it is by no means rare to find the possibility
of miracles defended on this ground. Such a view attributes our
surprise at what we should call a miracle simply to the fact of its
unfamiliarity to us. It makes the essence of a miracle purely
psychological. It is sometimes expressed by saying that everything
is miraculous, which is another way of saying that nothing is
miraculous. It implies a view of the world as being merely the
successive concretion by God of momentary configurations of
created being, which instantaneously collapse into non-existence
and have no more organic or causal connection with one another
than have the pictures that appear in rapid succession on a
cinema screen.! And once we allow ourselves to become sceptical
that the external world consists of persisting beings, possessing
determinate natures and exerting action on one another, it will
not be long before we find ourselves denying that there is an
external world at all; as Gilson has shown, there is a straight line
of development from Descartes through Malebranche to Berkeley
and Hume.” The very basis of the position from which we have
argued to the existence of God is the fact, which we have alleged
is evident to us as soon as we grasp the nature of the world of our
experience as it really is, that, on the one hand, the world consists
of really existing beings and that, on the other hand, these beings
are not self-existent. Looking back at this fact in the light of the
‘doctrine of creation, we can now see that it implies that God
‘creates finite being not by giving momentary existence to a
‘succession of entirely disconnected particulars, but by maintaining
lin being a universe which persists in time and in which there is
\genuine continuity and causality. ‘Thus God is not the only cause,
‘though - he is the only ultimate one; and he maintains the order
which is discernible in the universe a by annihilating the causal
relationship of beings towards one another, but by preserving it.
1 It may he objected that the pictures on the screen produce an illusion of continuity
and that therefore, for all we know, the continuity of the world in which we live may
be just as illusory. We cannot argue the point in detail here, but the heart of the reply
would lie in the fact that the very possibility of the illusion in the cinema depends upon
the existence of beings whose existence is not discontinuous, such as the screen, the
celluloid film and, above all, the audience.
2 See The Unity of Philosophical Experience, ch. vii, viii.
124 HE WHO IS
Were this not so, then the whole idea of an order of nature and,
indeed, of the existence of beings with definite characteristics
would be a complete illusion; and, among other consequences,
science would be reduced to the level of pure description, giving
us no valid insight into reality.1 The fact that potatoes nourish
me to-day would give me no reason to suppose that they may not
poison me to-morrow; and I should have no more objective ground |
for surprise if the cat which I can see walking across the garden
were to assume successively the forms of a purple cow, a chest of
drawers and the President of the Royal Society than I shall have
if it continues to exhibit its familiar feline appearance.
There is, then, a course of nature, and God has created a world
in which finite beings, while subject to generation, development
and corruption, have, in a certain relative sphere, determinate
characteristics and behave according to definite laws. They mani-
fest a combination of permanence and change: of permanence
because they are, in the strict sense, beings; of change because
they are only finite and relative ones. Each of them has a deter-
minate ontological status; yet, because
ofits finitade, itsfuture
depends not only upon its own particular nature but upon the
influence_of other finite. beings and.in.the.last.resort-upon..God
eeeapel
Ne
himself.2/ Its determinacy, being strictly relative, includes a cer-
‘tain indeterminacy; for only in God are existence and essence
estrone,
identified.
The question of the possibility of miracles, then, is simply the
question whether, in addition to the act by which God preserves
in existence the world of finite beings and the operation of their
1 It is a little startling to see how many scientists are prepared to save the autonomy
of science at the expense of denying that it gives any genuine knowledge about an
external world. It may be left to the despised Thomist philosophers to save from the
scientists the claim of science to be what its name implies. Cf., e.g., Eddington’s
Philosophy of Physical Science with Maritain’s Degrees of Knowledge, ch. iii. The scientists
of the last century tended to claim for science far too high a status of insight into
objective reality; their present-day successors often go to the other extreme.
The importance for theology of recognizing the reality of secondary causes is well
expounded by Thomas Jackson: “‘By this concession of some true power and property
of working unto natural agents, more is ascribed to the Creator of all things than can
be ascribed by the contrary opinion, which utterly denies all power or property of
working to the creatures; for he that denies any effects to be truly wrought by them
cannot ascribe their abilities or operative force (which in his opinion is none) unto
their Creator.” (Works, V, p. 279.) In other words, since we attribute causality, in
the supreme and primary sense, to God by a way of analogy based on the (secondary
and limited) causality which we see in the finite realm, any denial of secondary
causality will deprive us of our ground for attributing causality to God.
£
~ ® This fact is presumably what lies at the root of Whitehead’s philosophy of organism
—
with its “‘ principle of relativity”’ (Process and Reality, p. 30), though as will be seen later
on we do not give it the same interpretation as he does.
THE DIVINE ATTRIBUTES 125
secondary but real causality, it is also possible and proper for him
to intervene in thespheré of their operation by a direct initiative
of his power. That this is possible will be sufficiently evident;
that it is fitting may be less immediately so. But, if we remind
ourselves that operations are always directed to ends, it is difficult
to deny that, at any rate on certain occasions, God’s purposes
may be more fully achieved by such an intervention than without
it. To be able to predict precisely when a miracle would be fitting
is more than a finite mind has a right to claim; that would need a
knowledge of the relation of particular events to the ultimate ful-
filment of the divine purpose for the world which no one but God
—and, perhaps, human minds under the direct illumination of
God—possesses. We shall not expect miracles to happen with
very great frequency, since we can reasonably assume that the
order of nature which God has established is sufficient for the
normal operations of the world. But when a miracle is alleged to
have occurred, we need not deny its possibility; and, so far as
philosophy is concerned, the matter is then one for investigation
by ordinary rational and scientific methods.*
1 A very careful analysis and discussion, from the scholastic standpoint, of the
relation between the natural and the supernatural, the various senses in which the
word “supernatural” is used, and the relation between the supernatural and the
miraculous will be found in Verriéle’s book, Le Surnaturel en nous et le Péché originel, ch. i.
In view of the extreme looseness with which such terms as “supernatural” and
“miracle” are often used, the discussion there given is most valuable.
Dr. Tennant, in his Miracle and its Philosophical Presuppositions, gives a very convincing
vindication of the existence of an order of nature (though he overstresses its indepen-
dence through his doctrine of God as finite) and of creatures as dependent upon God’s
sustaining activity for their perduring existence (lect. ii; cf. Phil. Theol., I1, pp. 212,
215f.). But he looks upon miracles as due to an action by God similar to that by which
man makes use of natural forces rather than as a direct intervention by the Creator as
sovereign over the natural order.
CHAPTER X
TRANSCENDENCE AND IMMANENCE
/YHE God of traditional Christian theism is both transcendent
and immanent. He istranscendent because, as we have main-
tained in a previous chapter,‘‘a first cause wwho was himself in
even the very least degree involved inthe mutability, contingency
or insufficiency of the universe would provide no more in the way
of an explanation of the existence of the universe than it could
provide itself; such a God would provide a foundation neither for
himself nor for anything else.” He is immanent because unless
every finite being was sustained at.its ontological root by his
incessant «creative action—unless, to use the scholastic terms, he
was in it by “essence, presence and power”’—it would collapse
into non-existence dhioush sheer insufficiency ;it would, in Julian
of Norwich’s phrase, “‘fall to naught for littleness.”’ And both the
terms “‘ transcendent” and “‘immanent”’ are relative to the created
_>-world ;God is transcendent ¢o it and immanent in 2#.. Furthermore,
they are intimately related to each other, for they both arise out
of the Bee that the world is God’s creation. As Fr. Przywara puts
it, God, “‘as the pure ‘Is,’ is on the one side so inward tothe
. creation that the transient ‘is’of the creation is only from him and
in him—and yet on the other side, differentiated from the creation,
above it as the pure ‘Is,’ for whom no relationship to anything
which is ‘becoming’ is in any way possible.”’* The precise relation
of these two elements of transcendence and immanence, and their
consequences for religion, have been workd out with great pro-
fundity of thought and profusion of detail in the work from which
these words are quoted, and the doctrine ofanalogia_entis (that is,
the doctrine that creation is a similitude of God’s being, deriving
both essence and existence from his creative act, while being in no
way necessary to him) is there made the foundation of a general
theory of religion which is of quite exceptional significance. In
the present chapter we shall attempt a more modest task, and
1 p. 96 supra. 2 Cf. S. Theol., I, viii, 3.
3 Polarity, E.T., p. 33. An exposition of this difficult but most illuminating book,
by its translator, Dr. A. GC. Bouquet, appeared in Theology, December 1934. A short
discussion of Przywara’s teaching will be found in W. M, Horton’s Contemporary
Continental Theology, p. 65 f.
126
TRANSCENDENCE AND IMMANENCE bs 127
simply “give some discussion of the distortions that result when
either of the two characteristics is stressed to the partial or com-
plete exclusion of the other, concluding with some remarks on the
relevance of the balanced Catholic doctrine to two important
branches of theology.
The doctrine of transcendence in its extreme form results in the
view that God has no > realconcern \with the universe. He must
presumably have created it, or it would not be here, but it would
be-ridiculous
to suppose. that.he_can_have any present, dealings
with it. Historically this view was embodied in the Deism of the
eighteenth century, to which it was particularly congenial. For,
since it believed nature to be God’s handiwork and therefore to
manifest his character in the same way in which a sculptor or an
author is made known to us in his works, it provided the basis for
a sincere, if restrained and chilly, religious reverence, while at the
same time involving no untidy interference with the orderly
course of natural law as expounded by Newton and his successors.
aS Whitehead. says, “God made hisappearance inreligion 1under
shippedin whitewashed Churches’, ‘Whatever God might have
done in the past when the world first came into being, his present
relation to it is simply that of an artist to his work; in Addison’s
famous hymn it is as “their grand Original”’ that “the spacious
firmament on high and all the blue ethereal sky”’ manifest the
Deity, as rejoicing in “reasons’s ear,”’ they roll “‘in solemn silence
. round the dark terrestrial ball;’? and the attitude to God
which the contemplation of nature is expected to produce in us is
presumably not unlike the attitude to Sir Christopher Wren that
is induced by walking round St. Paul’s Cathedral.
The great instance of a religion of divine transcendence is of
course to be found in Mohammedanism. “The fundamental
conception of Allah among orthodox Muslim theologians is
negative. He is unique, as well as a unit, and he has no relation
with any creature that partakes of resemblance.’’? No practical
religion can, of course, deny all connection between the world
and God, and in Mohammedanism this doctrine of transcendence
is combined with an extreme voluntarism which makes God not
only transcendent but arbitrary and ruthless. The very name by
which the religion of the Prophet of Medina is known bespeaks
man’s complete helplessness before this implacable deity ;Islam is
1 Adventures of Ideas, p.157. * L. Bevan Jones, The People of the Mosque, p- 100.
128 HE WHO IS
submission, submission to the will of God. And by that enantio-
dromia, Which issomarked afeature of human activity, Moham-
médanism becomes the most militant religion in history, for once
the believer has made his submission he becomes himself devoted
to the divine will and so sees himself as an instrument of the divine
ruthlessness.
A religion of unbalanced transcendence allows no place for
intimate communion between man and God. In eighteenth-
“century Deism, worship becomes the orderly admiration by man
of the divine perfection; there is more than a suggestion of the
notion that worship is not so much something that man does for
the glory of God as something that he does for his own good; it
becomes what has been aptly called “the art of spiritual cosmetics.”
It therefore easily passes into a Voltairean cynicism. And, para-
doxically, just because it locates God at such a distance from the
world in which we live, it can easily deny that he has any effective
control over it, and then we are not far from the concept of a
God who is helpless before his own creatures. The practical out-
come of eighteenth-century thought has been described by
Professor Tawney in these words: ‘God had been thrust into the
frigid altitudes of infinite space. There was a limited monarchy in
heaven, as well as on earth. Providence was the spectator of the
curious machine which it had constructed and set in motion, but
the operation of which it was neither able nor willing to control.”
All that could be conceded—and, in a society where belief in the
Incarnation had not become entirely extinct, there was some
demand for this concession—was that, on very rare occasions,
the Creator might perhaps make some transitory intervention
in the affairs of his world, in the way in which an absétitee
landlord will sometimes visit his property to see whether it is
in order or to cope with some crisis. But, to quote Professor
Tawney again, concerning the divine Providence, “like the
occasional intervention of the Crown in the proceedings of
Parliament, its wisdom was revealed in the infrequency of its
interference.’”?
Side by side with this transcendence we can discern an apparent
immanentist tendency, but it is an immanentism of God not as
efficient but as formal cause. It is strongly manifested ii Pope’s=~
Essay on Man, which must surely be one of the most complacent
theological discussions ever written.
1 Tawney, The Acquisitive Society, p. 13. 2 Thid.
TRANSCENDENCE AND IMMANENCE 129
“All are but parts of one stupendous whole,
Whose body Nature is, and God the soul;
That, changed through all, and yet in all the same;
Great in the earth, as in th’ ethereal frame;
Warms in the sun, refreshes in the breeze,
Glows in the stars, and blossoms in the trees,
__ Lives through all life, extends through all extent,
Spreads undivided, operates unspent ;
____Breathes in our soul, informs our mortal part,
As full, as perfect, in a hair as heart:
As full, as perfect, in vile Man than mourns,
As the rapt seraph that adores and burns:
To him no high, no low, no great, no small;
He fills, he bounds, connects, and equals all 34
Have we, we might wonder, passed over into a religion of
complete immanence? Not really, for there is no suggestion that
man should enter into communion with this God who is the soul
of the world. The proper study of mankind is man, and virtue
alone is happiness below. Man’s ultimate duty, it appears, is
neither to a transcendent nor to an immanent God, though there
is enough talk about God and about virtue; it is rather to man
himself. For this is the conclusion of the discussion:
“That REASON, PASSION, answer one great aim;
That true sELF-LOVE and socIAL are the same;
That virTUE only makes our Bliss below,
‘ 292
And all our Knowledge is, OURSELVES TO KNOW.
We might add that, to an eighteenth-century philosopher, no
very intimate union was implied by describing God as the soul of
nature; Descartes had seen to that. For the classical Christian
doctrine_of man_as a psycho-somatic unity had given way to the
Cartesian. view.that the relation between body and soul was
purely external and accidental; man was, in Maritain’s phrase,
“an angel driving a machine,’’* and mind and matter were so
gssentially disparate that even the word “‘driving”’ is almost an .
/exaggeration. The God of deism, in short, oscillates between a
(genuine transcendence and a spurious immanence; but he is far ,
Pires being the God of Christian theism.
The.transcendentalism of Islam is of a very different order. In
the first place, Mohammedanism is a historical religion. It is
1 Essay on Man, Epistle I, ix. 2 Ibid., Epistle IV, vii,
3 Religion and Culture, p. 24.
130 HE WHO IS
based upon a revelation alleged to have been made through an
Arab born in Mecca about the year 570 and to be contained in ©
the book which he left behind him. For it there is one God, and
Mohammed is his prophet. The extreme voluntarism of the
transcendental revelationism of orthodox Islam reduced religion
almost entirely to the unreasoning performance of certain specified”
duties; God has demanded their fulfilment and that is all that
y
of Islam
there is to say about it. But very early in the histor
another trend becomes evident in reaction from this, and finally
establishes itself as the tradition of Sufism. It has both a philo-
sophical and a mystical side, and on the latter it represents the
attempt to satisfy a desire for union with God for which the pure
Koranic doctrine makes little or no provision. Perhaps its most
interesting manifestation is in the Persian mystic, al-Hallaj, who
has been made the subject of a remarkable and sympathetic study
by Fr. Maréchal.! Hallaj was executed in a.D. 922 at Baghdad as
a heretic, on the charge of having denied the Koranic doctrine of
the divine transcendence and of claiming to have achieved an
actual identification of himself with God. The conclusion of
Fr. Maréchal’s investigation is that Hallaj had, in all probability,
attained to a genuine mystical union with God of a very high
order. In support of this he points out that Hallaj had a profound
veneration for Jesus of Nazareth and that much of the doctrinal
content of the Koran itself derives from the Judeo-Christian
tradition. He sees, therefore, no insurmountable objection to the
possibility of the achievement of the height of mystical experience
by such a thoroughly sincere seeker after God as was Hallaj, and
he believes that the evidence shows its actual realization. But the
task of reconciling Koranic Doctrine with Sufism has been a
perpetual embarrassment to Muslim theologians, and it seems
to be generally agreed that even the great attempt at a synthesis
on the part of al-Ghazali in the eleventh century was not success-
ful?
1 Studies in the Psychology of the Mystics, E.T., p. 241 f., “The Problem of Mystical
Grace in Islam.” Cf. also the essay on “Islamic Mysticism” in Mr. Christopher
Dawson’s Enguiries into Religion and Culture. For a short summary of the history and
characteristics of Sufism, see L. Bevan Jones, The People of the Mosque, III, v, vi, or
R. A. Nicholson in E.R.E., XII, p. 10 f., s.v. “Sifis.” Maréchal’s study is based upon
the researches of M. Louis Massignon.
2 See Maréchal, op. cit., p. 273; Bevan Jones, op. cit., p. 146 f. For a study of the
early development and interrelations of early Christian and Muslim mysticism, see
Margaret Smith’s Studies in Early Mysticism in the Near and Middle East, This writer has
also published a detailed study of Hallaj’s predecessor, Mohdsibi, under the title An
Early Mystic of Baghdad.
TRANSCENDENCE AND IMMANENCE 131
ive very different type of revelationist transcendentalism is
represented by such typical figdtesofthe modern coiitinental
Protestant dogmatic revival as Nygren, Barth and Brunner. In
spite of their not inconsiderable differences—Nygren is a Lutheran
and the other two Calvinists, and Barth and Brunner have them-
selves parted company—they agree in denying that man can enter
mysticism
into any_real union with God and hence they.assert.that
isimpossible.1_ Man can have faith in God—justification
by faith
is for these writers the very-essence.of religion—man can even obey
God, though his obedience is bound to be vitiated by sin, but to
love God or to be united with him is altogether beyond man’s
power. It is interesting to see how this thesis is worked out by
Nygren in his great study, Agape andEros. God loves man in spite
of man’s worthlessness, indeed the divine agape is precisely love for
that which is worthless ;but for man to love God is hardly proper,
in spite of the scriptural texts. Faith, not love, is man’s proper
attitude to God, though apparently nobody between St. Paul and
Luther knew this.? In the course of his historical study Nygren
most scrupulously and charitably defends the great Catholic saints
from practically all the charges that Protestantism has ever brought
against them; there is only one fault of which he cannot acquit
them, that of trying to love God. For to him the assertion that
man can love God and achieve real union with him is clearly the
supreme heresy; it is a contradiction of man’s creaturely character
and asin against humility. In Barth this same trait takes the form
of denying that man can have any knowledge of God that counts
for anything except by divine revelation; no effort from man’s
side can do anything to pierce the barrier that creaturehood and
sin together have set up between man and God. And the whole
of Barth’s Gifford Lectures consist of a vehement and sustained
denial that the science about which he is lecturing—namely,
natural theology—has a right to exist.2. Brunner, it must be
1 I use “mysticism” here in the strict sense of a quasi-experimental knowledge of
God. Cf. Maritain’s definition of mystical experience as “a fruitive experience of the
absolute.” (Quatre Essais sur Esprit, p. 132.) A discussion of the various meanings,
proper and improper, which the word has borne and of how it may best be defined will
be found in Dom C. Butler’s Western Mysticism, p. 1 f. Cf. Maréchal, op. cit., pp. 286-7.
2 Thus Nygren refers to ‘‘the tendency . . . towards a weakening of the idea of
Agape in the Johannine conception.” (Op. cit., I, p. 117; cf. II, p. 24 f.)
3 The Knowledge of God and the Service of God, especially ch. i. It is interesting to note
as many of the continental Protestant theologians are forced on theological grounds
condemn their own subject-matter. Nygren writes about mysticism, Barth about
ae theology, Troeltsch about the social teaching of the Christian Churches, but
ch of them has to maintain that what he is writing about is an aberration,
K
132 HE WHO Is
admitted, is less violent than Barth about hess corruption,
but he is none the less opposed to mysticism. For, according
to him, mysticism means either that man can jump over the ©
gulfseparating him from God or that he can find God as im-
manent in his soul. And, in either case, the transcendence of God
is denied.1
/ What can be said in reply to this we shall see later on. It will
cebe well now to see what happens when an exaggerated stress is
\ laid upon immanence.
* The logical outcome of a purely immanentist theology is
/ pantheism, thedoctrine that God and’ the world-are” “simply
identical. But in practice such a view is very difficult to maintain,
if only for the reason that the multiplicity of the world and the
evil with which it is infected seem clearly to contradict it. Pan-
theism, therefore, tends to take certain modified forms ; 4t-may;
with some.doctrine.of maya, teach that the world of appearance
is merely illusion, it may hold on to belief in the reality and deity
of the world but take refuge in the notion that deity is imper-
sonal—both tendencies seem to be present in philosophic Hindu-
ism ;? it may, as in Taoism, reduce God to the position of a mere
principle of order,to“Which human life ought to conform ;3
or it may combine, in an unnatural union, all three tendencies.
But there is one caution which it is most important to bear in
mind.
Europeans have become so accustomed to interpret the notion
of God in the context of their own tradition that it is almost
impossible for them to divest it of the fully developed characteristic
of personality which it has acquired in the theology of Christen-
dom. They tend, therefore, in — such a matter as the
God. “Ifthe God in whom we believe were merely immanent,”
f
H
™.they ask themselves, ‘“‘what would follow?” To which the obvious
reply is that the God in whom we believe could not be merely
immanent, for he is transcendent as well. Before we can even begin
to discuss the great Asiatic religions we must divest our minds,
1 The Mediator, p. 292 and p. 293, n. I.
2 Cf. the discussion of Buddhism and Brahmanism in Coomaraswamy’s Buddha and
the Gospel of Buddhism, III, iv.
3 Cf., e.g., the Tao-te-King of Lao-Tse, printed in full in The Bible and the World,
Ppay f. Any suggestion of a personal deity is still more remote in Confucius; cf. the
nalects.
TRANSCENDENCE AND IMMANENCE 133
so far as is possible, of concepts which are incompatible with
them.
We can illustrate this point by reference to the much discussed
question as to whether Brahmanism and Buddhism _are_ really
inconsistent. To the European mind it seems at first sight per-
fectly obvious that they are, and that, from the standpoint of
Brahmanism, Buddhism ought to appear as a pernicious heresy.
For Brahmanism places before 1man as his ultimate beatitude the
identification of himself with the ground of the world—the soul
becomes the One, the Atman is the Brahma—while Buddhism sets
before him the ideal of complete extinctionofpersonality. But we
must be careful. For what is this ground of the world, and what is
the world of which it is the ground? In Christian thought it would
be a personal God, and the world would be his creation ;more-
over, men would be individual, substantial, personal beings. For
Brahmanism, on the other hand, the ground of the world is cer-
tainly not personal, and the world itself is of doubtful status; it is
we are told, _majya;—appearance, but what does that mean?
Appearance of something that really exists, or mere illusion?
And what men?”of They, too, are impermanent, and their per-
sonality is only superficial and. accidental. Is there really any
ontological difference between absorption into the impersonal
ground of an illusoryworld and complete extinction? Admittedly
there is a_difference between the technique of Brahmanism—at
least in some of its forms—and that of Buddhism, for the former is
an attempt to identify oneself with the ground orthe world, while
the latter is an attempt to escape from the world altogether. Yet
even here the difference is perhaps not so marked as might appear.
For the Brahmanist,)identification involves an entire break with the
world as we experience it in order to attain union with its ground.
The things that surround us are mere appearance or illusion, and so}
must be altogether rejected, whether this is done by the way of in-|
tellectual gnosis (jnana), disinterested activity (karma), or devotion |
(bhakti). A Christian mystic may reject the things of this world in
order to attain union with a transcendent God. But if God is purely
immanent, what sort of union can be attained by such renuncia-
tion? Is not the world renounced in fact identical with the deity
sought after? And ifso, how can we find him by this renunciation ?
It might well appear that the negativism of Buddhism is the logical
outcome of Brahmanism rather than the contradiction of it.
For if all existence is illusion, is it not better to try to escape
134 HE WHO IS
from it altogether rather than to seek to be identified with its
ground?!
_-For_a Christian there are few phenomena in the history of
“religion more puzzling than the contrast between Gautama
_ Buddha and his teaching. It is quite impossible to read the
‘description of his life and doctrine in the Buddhist scriptures
without being overawed by the sheer moral beauty of his character
and the utter genuineness of his desire to lead his fellow-meninto —
the way of salvation. But what is this salvation that he offers
them? Extinction. And what is his ultimate philosophy? Sheer
atheism. And the.final issue of it all is a kind of despairing
scepticism. For, Bi addhism 1teaches, man will not achieve beatitude,
merely by desiring it, even if beatitude is extinction. He will only
achieve it by being indifferent as to whether he achieves it or not.?
And in that case, what can it matter whether he does achieve it?_
Blessed_is he that expecteth nothing, for he shall not be dis-
appointed; this is the end of the matter. Parinirvana may or may
not be extinction ; nirvana is mere indifference.®
In practice, Taitheisn is¢ bound to be mitigated in i some way,
for the distinction and conflict between the various beings in the
universe makes it impossible to identify either each of them
separately, or the totality of them all, with God without making
some qualification. This point is brought out well in a parable by
the nineteenth-century Hindu teacher, Sri Ramakrishna:
1 In his essay on “ L’ Expérience mystique naturelle et le vide,” in Quatre Essais sur Esprit,
Maritain suggests that the Buddhist nirvana is fundamentally the same as the Brahman
mukti, though inadequately interpreted owing to the phenomenalistic philosophy of
Buddhism (p. 166). He works out in detail the view that this “void” is a metaphysical
experience of the substantial esse of the soul by a negative (or, rather, annihilating)
connaturality. Its.object.is.theAbsolute,.as.Atman.or self. It thus bears upon the soul
itself and, through the soul, upon God as the source of the soul’s being, but, on account
of its entire negativity, cannot distinguish between them. Hence the inclusion within
Atman of both the human and the Supreme Self. The Indian mystic is thus seeking
the Absolute within his soul by a stripping-off of the soul’s operations which leaves him
equally unable either to distinguish God and the soul from each other or to confuse
them with each other, since the experience is purely negative.
2 Cf. Buddha and the Gospel of Buddhism, II, iv. “Nibbana.” ;
3 The distinction is sometimes drawn between Parinirvana, the “‘“Complete or Final
Dying-out”’ at death, and Nirvana, the “‘Dying-out” which can be achieved in this
life: op. cit., p. 122. Itiis difficult for a Westerner to discover whether Buddhism can
validly be considered a gospel” of ontological suicide; in’a letter'to the writer, Dr. A.
Coomaraswamy asserts that no such view can be substantiated from the Buddhist
texts. In any case, general statements about anything as many-sided as Buddhism are
bound to be over-simplified. For a full discussion of the various views as to the exact
nature of irvana see Mrs. Rhys Davids, Buddhism, ch. vii.
Whitehead remarks that “Buddhism is the most colossal example in history of applied
metaphysics. ... Christianity. . has always been a religion seeking a metaphysic,
in contrast to Buddhism which is a metaphysic generating a religion.” (Religion in the
~-.MMaking, p. 39-)
; TRANSCENDENCE AND IMMANENCE 135
“The master said, ‘Everything that exists is God.’ The pupil
understood it literally, ‘but not inthe right spirit.While he was
passing through the street he met an elephant. The driver
shouted aloud from his high place, ‘Move away! Move away!’
The pupil argued in his mind, ‘Why should I move away? I am
God, so is the elephant God; what fear has God of himself?’
Thinking thus, he did not move. At last the elephant took him
up in his trunk and dashed him aside. He was hurt severely,
and going back to his master, he related the whole adventure.
The master said: ‘All right. You are God, the elephant is God
also, but God in the shape of the elephant-driver was warning 4
you from above. Why did you not pay heed to his warnings?’ ”’? /
As examples of modern attempts to construct a Christian
theology on an immanentist basis we may mention the works of
Professor Nicholas Berdyaev and Fr. Alexis van der Mensbrugghe,
both members of the Eastern Orthodox Church and both strongly
influenced by the ‘‘sophiology,” or teaching concerning the
Divine Wisdom, which looks back to the fourteenth-century mystic
of Mount Athos, St. Gregory Palamas,? and which became
prominent in Russian theology in the last century through the
thought and writings of Vladimir Solovyev.
Berdyaev launches a violent attack upon what he describes as
*‘naturalist metaphysics and theology,” in which term he includes
practically the whole of the historic theology of Christendom. For
him the truth about religion is to be found by exploring the con-
tent of the religious consciousness: “‘an abstract metaphysic can-
not exist, but a philosophy or a phenomenology of the spiritual life
is possible.”? He thus condemns rational theology no less
vigorously than does Karl Barth, though from a diametrically
opposite position; and it is well known that Berdyaev and Barth
consider each other to represent the extreme of theological error.
To Barth rational theology is anathema because it is an attempt by
sinfulman to comprehend a God who is incomprehensible; grace
and nature have nothing in common. To Berdyaev, on the other
hand, rational theology is anathema because it turns into an
object of discussion a God who is already comprehended in the
depths of the human spirit; grace and nature are really identical.
Berdyaev accuses the Thomist tradition of a vicious dualism and
1 “The Sayings of Sri Ramakrishna” in The Bible of the World, p. 163.
2 See the article on “‘The Ascetic and Theological Teaching of Gregory Palamas,”’
by Fr. Basil Krivoshein, in Eastern Churches Quarterly, January—October 1938. See
P. 140, n. 1, infra. 3 Freedom and the Spirit, E.T., p. 6.
136 HE WHO Is
also of a vicious monism, “Religious metaphysics and theology,”
he says, maintain “‘an opposition between the Creator and
creation and between grace and nature. But in making this
opposition... creation is naturalized and objectified, with the
result that the Creator himself is subjected to the same process.’’!
We must not, he urges, rationalize religion. ‘Spirit is not a
substance, an objective reality, | in the same sense as other sub-
stances. Spirit.is_life, experience, destiny. | A purely rational
metaphysic _¢
of spirit is impossible. Life isonly disclosed in
experience. ’”? ‘Hence to prove the existence of God is impossible,
\for “the reality y of the spiritual
Pp world and of the divine do not
correspond in any way to the reality of our sense-perceptions and
{
i
our thoughts. ... It is impossible to ask the question whether
\\ there is a reality which corresponds to the experience of the great
saints, to that of the mystics, to that of men who live on a higher.
spiritual plane, for that is a question arising only within the sphere
of psychology, naturalism, and a naive, and non-spiritual,
realism.,’”? :
The result of this denial of the validity of rational theology is
that discussion becomes impossible, for there is no common ground
from which Berdyaev and his critics can even begin to argue.
There are pages in which Berdyaev appears to be stating what is
practically a Thomist thesis, but they are followed by denuncia-
tions of Thomism; there are pages that read like pantheism, but
pantheism is violently repudiated. This is hardly to be wondered
at, since, if Berdyaev’s fundamental contention is correct, his own
formulations fall under the same condemnation as those of his
opponents; for he, no less than they, is attempting to express in
the categories of human speech what, on his own theory, is
ineffable. We come to the heart of the matter when he states his
ultimate postulate that freedom is-prior to being,* for this means
that it is prior to God. Adopting the distinction made by Eckhart
between God (Goit) and Divinity (Gotthert) in a way from which,
one imagines, Eckhart would have fled in horror, and identifying
Divinity with Boehme’s Ungrund, he asserts that “both philosophy
and theology should start neither with God nor with man (for
there is no bridge between these two principles), but rather with
the God-Man. The basic and original phenomenon of religious
1 Freedom and the Spirit, p. 6. Cf. the vigorous attack on “objectification”? in Solitude
and Society, II, ii.
2 Freedom and the Spirit, p. 9. 3 Freedom and the Spirit, pp. 10, 11.
4 Freedom and the Spirit, p. 119. Of. Solitude and Society, p. 24.
TRANSCENDENCE AND IMMANENCE 137
life is the meeting and mutual interaction between God and man,
the movement of God towards man and of man towards God.”
But we must beware of giving such statements as this the meaning
which a rational theology would attribute to them, since, as we
have seen, the very idea of a rational theology is denied. The real
problem that Berdyaev raises, though he fails to face it, what
is the
relation of theology to reality can possibly be once this denial has
been made. Berdyaev would not admit that he is merely describ-
ing the psychological phenomena of the religious consciousness,
yet it is impossible on his own principles to see what else he can be
doing. Denying the doctrine of analogy, he insists that his mode
of_thought is symbolic.2. Yet, since he maintains that his is “‘a
bolic mode of thought which brings us face to face with oi
mate realities,”’* it is not easy to see how this denial of analogy can
be justified.
~~ The conclusion would seem to be that Berdyaevis attempting the
impossible. There are few modern writers whose work is more
provocative and stimulating, and there is probably none who is
able so profoundly to read the signs of the times. With his desire
to find a place in his discussion of religion for the whole content of
the religious consciousness, and to avoid that type of abstraction
which substitutes bloodless categories for the full experience of
life, we cannot but sympathize. But we cannot admit that he has
succeeded. Indeed he himself seems to be prepared to confess as
much. “I recognize,”’ he writes, “that there is something essential
which I cannot put into words.... I put my problems in the
form of affirmations. But my thought as it moves within my own
being is that of a man who, without being a sceptic, is putting
problems to himself.” The only question is whether the very way
in which he is putting them may not deprive him of the possibility
of getting an answer.
Berdyaev’s-immanentism seems to derive at least as much from
German idealism as from Eastern Orthodoxy; this could hardly
be said of Fr. Alexis van der Mensbrugghe, who claims that his
views are firmly based upon Platonism, patristic theology, Byzan-
tine iconography and the thought of Russian Orthodoxy.’ He
1 Freedom and the Spirit, p. 189. Cf. p. 194. Cf. also the remarks on the “‘theandric
idea” in Solitude and Society, p. 40of.
2 Freedom and the Spirit, ch. ii, “Symbol, Myth and Dogma.” 3 Op. cit., p. 83.
4 Freedom and the Spirit, Introduction, p. xviii.
5 From Dyad to Triad,p. 2. See also a review by the present writer and a reply by
the author in Sodornost’, fe aeaer 1935 and March 1936.
138 HE WHO Is
maintains what he describes as a doctrine of analogy, though it is
very different from the analogia entis of the scholastics, and this
doctrine becomes a universal principle, which is applicable to
morphology, mythology, logic, ontology, axiology, and theology
alike. There are two modes, he asserts, in which the One differen-
—*
tiates itself: a “horizontal”, threefold differentiation of Modality,
“and a “vertical”. Baolelddifferentiation. ofee In its theo-
meee
God’s Existence from ‘his ESSE or Wisdom (Sophia: In this
Sophia there is included the created universe,
both in its ideal
essence and in its concrete reality. For the ideal essence and the
concrete reality of the world are not admitted to be distinct from
each other. “There is no first ‘conceptual realization’ in God,
and after that, apart from that, later on, another ‘realization in
reality.’ God does not think first and create afterwards. His Act,
as that of apure. Spirit, is absolutely simple ; if he thinks a Torn
that form is created ipso facto;his
his thinking is
is creative. We, on the
contrary think first and make afterwards, because we are only
secondary agents (causae secundae). We can only work on a given
material already existing.... But that is exactly where we are
only transformers, not creators in the adequate sense of the word.”*!
Several comments seem to be called for on this passage. The
temporal presentation—“‘later on’’ and so on—is, we may sup-
pose, merely for verbal convenience. But the denial of any dis-
tinction whatever between God’s knowledge of things as possible
and. his creation of them as actual presumably involves the conse-
quence that everything that is possible is actual as well, or, to put
this in another way, that this world is the only possible one. And
the comparison with human work, while the assertion that is made
must be admitted, is quite irrelevant, for no one (except possibly
David of Dinant, whose “ravings” were so sternly rebuked by the
Angelic Doctor?) has claimed to find within the God of Christian
theism the matter out of which the world is constructed.
But the crux of the matter is reached in the question of the
nature of evil. “‘Dualism (or Pluralism), writes Fr. van der
Mensbrugghe, ‘‘is wrong in seeing Creation as a movement ad
extra, the raising of a second subject of Being. . . . Sophian or Dual
Orthodoxy [his own theory] is right in seeing Creation at once as
a movement ad intra sed ad infra. Its definition of Creation is
1 Sobornost’, March 1936, p. 20. 2.$.¢.G., 1, xvii.
TRANSCENDENCE AND IMMANENCE 139
Internat Inferiorization of Being.”! The obvious objection to-this
doctrine is that, if it is true, evil, as indeed everything in the
es =
‘created world, becomes something happening in the interior of”
To this criticism Mensbrugghe replies as follows :
“Thee theory. does indeed deny absolute evil; but the relative
evil consisting 1 in splitting the lines of action a ‘Wisdom along
the double axis of the Better and the More is amply safeguarded,
and is, I think, sufficient to explain the evil in this world..
Peccability could only be predicated of Sophia as far as it has
nothing to do with rebellion or even with sluggishness in answer-
ing God’s Will, so far as no meaning of moral guilt be attached
to it.... There is thus as much difference between Sophia’s
Sin and Moral Sin as between Original Sin in the babe and
Actual Sin in the grown-up. ... Sophia is sinless in her service
of God (except for that ®80e4 which makes her oscillate between
Better Service or More Service), although physical and moral
evil arise in and through the relation of the created forms
between themselves.’’?
It is difficult to be satisfied with this reply, for the mutual relations
of the created forms are, on this theory, themselves included in
Sophia. If creation is, as is alleged, internal inferiorization of
Being, so that “the Container needs the Contained, the Absolute
needs the Relative in order to become God,’’? then moral evil
must itself be within God, for there is nowhere else for it to be.
The idea that there is anything which is in any sense outside God
is rejected quite definitely: ‘‘once and for as long as an actual
‘outside’ of God is granted, any further dependence or unity of
Origin cannot prevent Dualism 7n actu.’
It is easy to see what is the danger that Mensbrugghe fears from
the admission that anything is exterior to God; it is the setting up
of a dualism in which God and his creatures appear as ranged on
the same level in some sort of medium anterior to both; indeed, he
explicitly says so, when he quotes a phrase from Pringle-Pattison
about the “impropriety in placing God and men in the same
numerical series, and in speaking as if we and God together, in a
species of joint ownership, constituted the sum-total of existence.”
(Berdyaev, it will be remembered, made just this objection to all
1 From Dyad to Triad, p. 61. 2 Sobornost’?, March 1936, pp. 20, 21.
3 From Dyad to Triad, p. 62. 4 Sobornost’, March 1936, p. 21.
5 Ibid. The quotation is from The Idea of God, p. 389.
140 HE WHO IS
systems of rational theology.) But this is precisely what the doc-
trine of creation as we have interpreted it in line with the Thomist
tradition refuses to do. A God who is merely a primus inter pares
provides, as has been repeatedly urged, no ground for either his
own existence or that of anything else; but it is not true that the
only alternative view is that creation is internal to God. According
to the doctrine of analogia entis, God as self-existent Being is alto-
gether distinct from the world, while the world is entirely insuffi-
cient and dependent, although, at the same time and indeed for
this very reason, it is most intimately interpenetrated by the
_creative act through which he is present to it at the heart of
' its being. And we shall argue, in concluding this chapter, that
only a doctrine in which transcendence and immanence are com-
bined in this way is capable of providing a satisfactory basis for
religion.?
pooner
1 It should be pointed out that Mensbrugghe is by no means the only modern
Russian theologian who has discussed the problem of creation in the light of the
conception of the Divine Sophia, There is a line of development going back through
Fr. Paul Florensky to Vladimir Solovyev. And a sophiologyin many respects different
from that of Mensbrugghe has been worked out by the distinguished Dean of the
Russian Theological Academy in Paris, the Archpriest Sergius Bulgakov; his only
work on the subject in English, apart from a few articles in Theology (July and August
1931, January 1934), is The Wisdom of God. Eor him, the divine ousia and sophia are
identical ;sophia is the self-revelation of the Godhead and belongs to all three Persons
of the Trinity. Sophianity (which is identical with theandrism, the ultimate unity of
Godhead and manhood which is manifested in the fact that man is made in the
image of God) is a general metaphysical and theological principle, which provides a
particular understanding not only of the doctrine of God, but also of cosmology,
anthropology, Christology, pneumatology, Mariology and ecclesiology. Bulgakov
insists very strongly upon the distinction between God and the world, and upon the
fact that the world is not necessary to God ; “‘‘The world as such maintains its existence
and its identity, distinct from that of God. . . There is no such ontological necessity
for the world as could constrain God himself to create it for the sake of his own
development or fulfilment; such an idea would indeed be pure pantheism”’ (op. cit.,
p. 110). Distinction is made between the uncreated sophia, which is the locus of the
divine prototypes of all possible creatures, and the creaturely sophia found in the actual
world. “Here we have at once Sophia in both its aspects, divine and creaturely.
Sophia unites God with the world as the one common principle, the divine ground of
creaturely existence. Remaining one, it exists in two modes, eternal and temporal,
divine and creaturely” (op. cit., p. 112).
Bulgakov’s supporters claim that his treatment solves the problem of the relation
between God and the world, whereas the treatments of both the Barthians and the
Thomists are unsatisfactory. Berdyaev, as we have seen, claims also to steer a course
between these two positions ;he dissents from sophiology, though from a quite opposite
reason from that of most of Bulgakov’s critics, namely that he considers Bulgakov to be
too conservative! It must be added that Bulgakov’s thought is extremely difficult for
a Western mind to follow; also, it has given rise to a painful domestic controversy in
the Russian Church, which led to a condemnation by the Moscow Patriarchate in
1935 (see Orient und Occident, March 1936). The opinion may be ventured that, what-
ever its inadequacies in the realm of trinitarian theology and Christology (and on this
see the review by J. P. Arendzen in Eastern Churches Quarterly, January 1938), in the
cosmological sphere Bulgakov’s sophiology would appear to approximate to the
scholastic doctrine of analogy, though the obscurity of the language makes a precise
TRANSCENDENCE AND IMMANENCE I41I
For a-religion of pure transcendence. places-God.too-far-away
from
us for him to be relevant to our life; if there is no line of
communication between him and us, he ceases to be our concern.
And if the attempt is made to correct this deficiency by super-
imposing an extreme revelationism, as in Barthianism, so that,
although there isno way up from us to God, God has blasted a
way down to us, God’s accessibility has been saved at the cost of
making him unintelligible. If the way between earth and heaven
is an entirely one-way street, God may indeed utter his word, but
we can never understand it, for it comes to us simply as a bolt from
the blue and there is nothing in our experience with which we can
relate it. On the other hand, a religion of pure immanence places
God too close to us to be of any real use. It immerses him in the
very predicament in which we ourselves are entangled. ;.iJt_makes
him neither our judge nor our saviour—and it is noteworthy that
neither Brahmanism nor Buddhism offers us a saviour ;at best they
tell us how we may hope to save ourselves. Dr. Demant has shown
with devastating clearness what this involves in the sphere of
human history, but it is equally true in the sphere of personal
piety.
““God conceived as the immanent spirit in the march of his-
tory,’ writes Dr. Demant, “makes religion an appendage of
events and secular movements. Such a religion can neither
interpret human history nor help man to make it. It registers
religiously what transpires under a natural and secular egis.
But, the abstract transcendentalism of theological Protestantism
leads to the same attitude. For the actual work of living in the
world the dogma that all is equally under the divine ‘No’ is as
useless as the dogma that all is equally under the divine ‘Yea.’
In either case man has to find his criteria of judgment and action
assessment practically impossible. It may be noted that he rejects Solovyev’s doctrine
of sophia as “undoubtedly syncretistic” (p. 23).
Owing to the inaccessibility of the material it is very difficult for Western theologians
to investigate sophiology in detail, though a thorough examination in the light of
traditional Western theology would be of great interest and value. A long discussion
of the cosmology of Solovyev, whom he convicts of “univocity,” was given by Dom
Theodore Wesseling in Eastern Churches Quarterly, January—October 1937, and this was
followed by a series of articles by Fr. Basil Krivoshein on St. Gregory Palamas,
January—October 1938. The words of so rigid a “‘westerner” as Maritain may be
,worth quoting here: “I would add that Greek and Russian piety, which differs
apparently from Catholic piety not so much in divergences of dogma as in certain
_ characteristics of spirituality, is much less hostile, in my opinion, to the philosophy of
' St. Thomas than might at first be supposed. It approaches the problems from another
angle and the scholastic presentation as a rule irritates and offends it. These are merely
. questions of modality.” (St. Thomas Aquinas, E.T., p. 70.)
142 HE WHO IS
from some source extraneous to his central religious conviction.
pectey‘had to posit that, outside the realm of grace and faith
|whiels
whieh-affected only man’s inner orientation, worldly affairs were
iby God entrusted to the princes ;Galvinism, with its valiant effort
Ito make the world Christian, succeeded only in making Christian-
|ity worldly. Much modern Christianity is avowedly confined to
| encouraging a ‘change of heart’ and a change of nothing else.
| The historic scene is left to ‘the experts’ who have inherited the
| function of Luther’s princes.’’}
~ Again, Dr. Demant remarks that the Christian mind has
“split, in its innermost outlook, into cosmic interpretations which
bring God within the world process and purely redemptive theolo-
gies which take man as religious out of it, leaving his actual
existence at the mercy of its floods.’
In contrast with these one-sided distortions,
“‘the dogma which sees existence in terms of Being and Becoming,
Eternity and Time, Creator and Creation, gives us the right to
make choices in the actual world without identifying our choice
with the absolute good.’’®
We can follow out this same point in the realm of Christian ,
piety. Here, as before, the doctrine ofa God who is both
immanent and transcéndent ‘produces a tension which has seemed
to many to be paradoxical and even inconsistent. God is, on the
one hand, to be approached by me only if I leave all created things
beneath me in a cloud of forgetting, while I strive to pierce the
cloud of unknowing which separates me from the ineffable super-
essential Deity;+ on the other hand, the one and only place in
which I can find him is in the “‘apex”’ or “depth”? or “‘fine point”’
of my soul.> Furthermore, these two apparently entirely opposite
movements do not seem to those who describe them to involve any
contradiction ; certainly St. Francis of Sales was not conscious of
any antithesis between his own doctrine and that of the great
Spanish Carmelites.
1 The Religious Prospect, pp. 180-1. 2 \Ibid., p. 214. 3 Tbid., p. 154.
4 Cf. The Cloud of Unknowing, passim.
5 Cf. Blosius: ‘‘‘Therefore should he ‘introvert’ himself—that is, should turn himself
into his own soul, and dwell there in his own heart—for there will he be able to find
God.” (Book of Spiritual Instruction, ch. iii, 3.) And St. Francis of Sales: “Le second
moyen de se mettre en cette sacrée présence, c’est de penser que non seulement Dieu
est au lieu ot vous étes, mais qu’il est trés particuliérement en votre cceur et au fond de
votre esprit, lequel il vivifie et anime de sa divine présence, étant 14 comme le coeur
de votre coeur et l’esprit de votre esprit.” (Intro. a la Vie dévote, II, ii.)
TRANSCENDENCE AND IMMANENCE 143
The fact is that the two movements, although their descriptions
are so different, are, as regards their object, the same, for that
object is the one transcendent and immanent God. For this
reason, even from the psychological point of view, their difference
is much less than might be expected. For the transcendent God
who is sought in the cloud of unknowing is present already in the
y tepth of the soul; while the immanent God who dwells in the
_ centre of our being is infinite, perfect and self-existent.
~ There is, however, a sense inwhich the transcendence of God is
primary, for Gods‘isthegiver of existenceto the beings in which he
is immanent. The primary act is not from man to God, but from
God to man; the Barthians are right about this. As Przywara says,
A
“for the Catholic foundation of religion, which is the analogia \
}
ents, all other bases for religion count as immanental. The God of j
Catholicism alone-ts the truly transcendent deity.... Even for the highest |
mystical union, God remains the One who dwells in Light unap-
proachable; even in the most intense form of mystical illumina-
tion, as it occurs in St. John of the Cross, the experience is of this
inaccessibility, of the ‘Night’ as well as of the unio mystica.” Asa
/result, “* “Catholic immanentism’ isthe immediate consequence of
/ the rigid super-creatureliness of the Universal Creator, while
/ ordinary immanentism simply confines him to being within his
j creation.”’+
“It is this, and not Mr. Aldous Huxley’s theory that God is really
impersonal, that explains the phenomenon of the Dark Night in
Christian mysticism. The closer the mystic approaches to a vision
of the pure Deity, the more the object of his contemplation exceeds
the capacity of his vision. Hence we find in the writings of St. John
of the Cross the most terrible descriptions of spiritual desolation
side by side with the insistence that it is precisely in this desolation
that the soul is both being assimilated to God and being given a
clearer knowledge of him.
1 Polarity, E.T., pp. 38, 41, 42. It is important to remark that, when the Catholic
doctrine_of the relation between God and the world is neglected, the distinction
between immanéncé and transcendence very largely vanishes too. Mr. E. I. Watkin
points out that ““ultra-transcendental pantheism, when it regards the finite as but a
manifestation of an underlying Godhead, ipso facto regards the finite as a mode of that
Godhead, and becomes thus identical with ultra-immanentist pantheism. Catholic
mysticism,” he continues, “bars both passages to this common error. It bars the
immanentist approach by insisting on the absolute distinction of finite beings from
God, in virtue of their essential finitude. It bars the transcendentalist approach, and
therefore the mystical modernism that is taking that way, by its doctrine of special
relationships, including, as it does, its doctrine of personal identity between a created
being and God, in the Incarnate Word.” (Philosophy of Mysticism, p. 72.)
144 HE WHO IS
- “O guiding night,
O night more lovely than the dawn;
O night that hast united
The lover with his beloved
And changed her into her love.’’?
Thus, writes Przywara, Catholic piety is
‘“‘a never-ending (relatively infinite) unveiling of the absolute
infinite God in and above the never-ending (relatively infinite)
self-evolution of life in him. It is bounded by two extremes, the
oneimmanentism, which is an actual self-evolution of God in man,
for which God is only another name for creaturely evolution ; the
other, transcendentism, which is the lifeless static adoration before
a wholly remote eternal Deity, for whom human life with its self-
mutation has in general no inward religious meaning except at
the most the negative one of pure abandonment.... The
repose in movement of Catholic transcendentality is the decisive
formula. Two attitudes interpenetrate here to form a_ twofold
unity: (i) Experience of the infinity of God in the endless rhythm
of life, and (ii) Adoration of the same infinity above the endless
att
rhythm of life.’’?
This, then, provides the theological basis of Catholic mysticism.
The psychology of it has received perhaps its profoundest treat-
ment in Fr. Gardeil’s work, La Structure de Ame et [Expérience
Mystique. As we shall see, it involves just the combination which
we have expounded of divine transcendence and divine imman-
ence.
Gardeil begins by considering the essential structure of the
human soul. It is a mind, a mens, a special kind of created spirit,
and as spirit it bears in itself the image of the Supreme Spirit,
God its Creator. As spirit it has a natural desire for God, but as
finite spirit it can tend to him only according to the idea which it
can form of him, and it is totally incapable of raising itself to this
infinite object of its desire. Nevertheless, what it cannot do for
1 Amada en el Amado transformada. The poem is quoted in full in Spanish Mysticism, by
E. Allison Peers, p. 227.
2 Polarity, E.T., pp. 61-2. Przywara specifies three fundamental modes in which
the human ego orients itself in relation to the ultimate object of its experience:
immanence.(Zustdndlichkeit), which looks for God in the depths of the soul; transcen-
derice (Gegenstdndlichkeit), concerned with the “wholly Other”; and, cutting across
these, transcendentality (Tatstdndlichkeit), which postulates no existing object, but is a
pure striving. Transcendentality tries to hold immanence and transcendence in
tension, but such tension is unstable and explosive. Only in Catholicism is it firmly
achieved, for there the unifying principle is God himsélf, who is both transcendent
and immanent.
TRANSCENDENCE AND IMMANENCE 145
itself God can do for it, for, leaving on one side sin, which is a
distortion of the soul and not an element in its true nature, there
is nothing in the soul that is repugnant to the vision of God. The
soul is merely helpless. :
Now the soul is, as we havejustseen, a created analogue of God,
and, byhis presence of immensity, "God is more present. to. it than
it is to itself. Hence we may expect that the best analogue that we
can find to the soul’s knowledge of God is its knowledge of itself.
This self-knowledge has one very remarkable characteristic;
namely that, although the soul is, ontologically considered, more
immediately related to itself than to any of the beings which it
apprehends (since it zs itself, entitatively, while it merely knows
them), it knows other beings before it knows itself. Indeed, it comes
to know itself only as a result of those acts in which it turns
towards external beings.1 (It may be noted that, absurd as it
might seem, children learn to recognize external objects before
they learn their own self-identity.) Therefore, by analogy, and
a fortiori, we may expect that the God who is more interior to the
soul than it is to itself will also be known only as a result of the
soul’s acts.
But as a result of what t acts? The act of self-reflection indeed
eens is a created analogue of him. But “ee a knawledae will be
only a knowledge of him in his effects, not a knowledge of him in
his own proper being. It will be of the same kind as the soul’s
knowledge of God through his effects in the external world, though
of a vastly higher degree. Is there, we therefore ask, any kind of
human act through which we may come to know God himself in
the same sort of way as the way in which, in self-reflection, we
come to know our own selves? We know already that God is near
enough for this, for he is present by immensity at the very heart
of our being, in the most intimate way conceivable. But is there
any way by which we can be made to see him there, any way by
which our faculty of knowledge can be brought to exceed the
limitations of its finitude and be raised to the apprehension of this
infinite Being?
It is asserted that there is, and its name is grace. Just as,
through the operation of his natural acts of knowledge directed
towards the external world, man can come to know not only the
1 Professor John Laird expounds this point very usefully in his Mind and Deity,
p. 80 f.,though he also allows for a direct reflexive self-awareness.
146 HE WHO IS
external world but himself, so, by sanctifying grace, he can come
to know God. And this means that, in addition to his presence of
immensity, God must be made present in a new and higher mode
in the soul, for only God can know God adequately, and so man
can only know God in a way which is analogous to God’s know-
ledge of himself if he is elevated into God’s own act of self-know-
ledge. Just as by nature the soul knows itself as the principle of its
natural acts, so by grace it knows God as the > principle. ofits super-
natural-acts. Even so, this indwelling of God in the soul by grace,
which is offered to anyone who is prepared to live by the theological
virtues, is not enough to give us a knowledge of God that is in all
respects parallel to the knowledge of ourselves which we have
through our natural acts.
‘““When, by psychological knowledge, the soul reflects upon the
act which it has just performed, it seizes this act in its entirety—
in its term, namely the concept, in its movement or intention of
knowledge, in its efficient principle, namely the soul itself in so
far as it is activated by its act—for all this is in itself actually
intelligible. On the other hand, in the intentional life of living
faith, all that can be actually apprehended are the object of this
life, constituted by the truths of the Faith (which are themselves
representatives of the divine reality), and the activity of living
faith itself.’’+
For anything more than this the ad hoc initiative of God himself is
necessary, and in this thesoul, whatever it may have previously
done in the way of preparation,is
i | entirely passive. The soul can
neither demand nor expect this; it is a pure giftfrom God, and it
raises the soul to the highest pont of union attainable in this life,
that of the Spiritual Marriage or the Prayer of Union. The pre-
dominant part in this is played by the virtue of charity and the
gift of wisdom.
“The Holy Spirit... communicates to the soul that loves
God his knowledge of divine things, that is, in particular, what is
deepest, most formal, most God_(if we may venture to use the
phrase), for thé Holy Spirit comprehends God in his depths and
in an infinite mode. . The human spirit, having placed itself
by its use of the gift of wisdom under the rule of the Spirit who
uncovers before it the depths of God, allows itself to be joined to this
1 La Structure de ?Ame, U1,p. 185. In this discussion I have tried to reproduce the
substance of Gardeil’s thought, without either making unnecessary use of technical
terms or referring to the many important side-issues which he raises,
TRANSCENDENCE AND IMMANENCE 147
interidr God without any concept, tanquam ignoto et inaccessibili,
and in this attitude of absolute renunciation of seeing or of making —
for itself ideas of God, it lets itself be carried towards God by the7!
Holy Spirit.... No doubt the soul will then contemplate —
nothing, but if it truly shares in the divine wisdom, it will do
better than contemplate; in so far as it has become one spirit
with God, it will feel, touch, and experience immediately the
substantial presence of God within itself.’’! .
Such an experience can come rarely, and then only transiently,
fin this life; and, as has been said, it is a pure gift from God.
\Nothing remains beyond itbut the Beatific Vision, of which it is
‘itself a kind offoretaste. —
Tt will be readily seen how, in this description of the life of the
soul in grace, the divine immanence and the divine transcendence
are balanced, and how unjustified is the complaint of the Barthians
that Catholic mysticism, and indeed Catholic religion as a whole,
is just an attempt of man, in defiance of the divine transcendence
and his own creatureliness, to climb up to God. It does indeed
find place for both immanence and transcendence. God is
immanent in the depth of the soul and it is there that he is tobe
found. But he is immanent there, not as contained in it,but rather
“as containing it; not in the sense that he is limited and restricted
by man, but in the sense that man, at the very root of his being, is
altogether dependent upon God. God, then, is immanent, but
the unveiling of this immanent t God iis not. The work of man, All
that man can do without grace is to know God as the author of
certain effects, and he can only know this because God ts the author
“of them}; not because man can, as it were, put God under his
scrutiny. Anything beyond this is the work of sanctifying grace,
of the elevation of man by God to a participation in God’s own
life ;and the very essence of grace is that it is gratuitous, a free gift
of God which man has neither the right to demand nor the power
to appropriate by his own efforts. Finally, in mystical experience
—and it is, it will be remembered, against this that the Barthians
raise their strongest protests—the human soul is devoid of all
activity whatever, even of the least active co-operation; it loves
God, as St. John of the Cross says, ‘‘not through itself, but through
himself; which is a wondrous brightness, since it loves through
the Holy Spirit, even as the Father and the Son love one another.’”*
1 Tbid., II, pp. 258-60.
Living Flame of Love, 2nd redaction, III, 82. Works, III, p. 206,
L
148 HE WHO IS
‘The substance of this soul, although it is not the substance of
God, for into this it cannot be substantially changed, is neverthe-
less united in him and absorbed in him, and is thus God by
participation in God.”! And the unanimous teaching of all
Catholic mystical theologians is that, whatever man may do in the
wayof preparing himself for mystical union with God, and how-
‘ever large or small may be the proportion of souls that God calls
to it (questions which have received much discussion in recent
_ years”), the actual elevation to mystical prayer is a pure and
unconditioned gift of God for which man has nothing more than
an entirely passive or obediential capacity.
Non-Catholic mystical doctrine nearly always falls into one of
two extremes. It may, on the.one hand, envisage the mystical
union as the elevation of the soul“by its own powers to the level
of a deity who is conceived as purely transcendent, the “‘flight of
the alone to the alone”’ of the neo-Platonists, and in this case there
is a virtual denial of the creatureliness of man. It may, on the
contrary, set forward a technique for the identification of the
human personality with a God who is located purely in the depths
of the human soul, in which case there is a virtual denial of the
divine transcendence, as in the teaching of Brahmanism and of
such Western mystics as Jacob Boehme and, presumably, Angelus
Silesius. Against both these types of mystical theory the opposition
of Barth, Nygren;~Brunnéer and Niebuhr is justified. Their
criticismdos not, however, touch the general theory of Catholic
mystical theology, with its two fundamental doctrines that God is
both transcendent and immanent and that mystical union is
achieved not by any human effort, but by the pure and uncon-
ditioned act of God himself, who elevates the soul into a participa-
tion of his own divine life. It is most regrettable that so many,
both of the supporters and of the opponents of mysticism, in recent
years have taken as representatives of Christian mysticism such
very untypical figures as Eckhart or even Angelus Silesius, rather
than St. John of the Cross or St. Teresa of Avila. Thus, on the one
side, Dr. Inge describes Eckhart as “the greatest of all speculative
1 Living Flame of Love, 2nd redaction, II, 34. Works, III, p. 158.
2 Cf, e.g., the works of Poulain, Saudreau, Farges, etc., and the balanced dis-
cussion by Garrigou-Lagrange in Perfection Chrétienne et Contemplation.
3 It is important to notice that mystical experience does not arise, as Dr. Edwyn
Bevan among others supposes (see Symbolism and Belief, lect. xv), merely from a peculiar
type of temperament on the part of the mystic, though this may dispose towards its
reception, but from the initiative of God.
TRANSCENDENCE AND IMMANENCE 149
mystics,”! and Berdyaev singles out Eckhart and Angelus
Silesius for special approval.? On the other side, Brunner claims
to show that mysticism and Christianity are incompatible by
arguing for a substantial identity in the teaching of Eckhart and
of the tenth-century Indian Sankara,? while Niebuhr similarly
uses Eckhart in order to show that mysticism necessarily contra-
dicts human creatureliness.t None of these arguments, in fact,
penetrates to the heart of the question, for reasons which have been
made clear. Finally, it may be remarked that Catholic mystical
doctrine in no way neglects the centrality of Christ, in its emphasis
on union with God, though it has often been alleged to do so.
While, from the psychological aspect, there is, as we have seen, a
radical distinction between the obscure and mediate knowledge
of God which the Christian non-mystic possesses and the quasi-
experimental knowledge which is conferred upon the mystic by
God himself, the ontological foundation of the two is the same,
namely union with the divine humanity of Christ by sanctifying
grace.® By incorporation into the manhood of Christ, who is the
eternal Son of the Father, the soul is given by adoption a real
participation in the Sonship of Christ and so enters into the life of
the Trinity. This is simply the other side of the indwelling of God
in the soul to which the Johannine Gospel testifies—“If any man
love me, my Father will love him and we will come unto him and
make our abode with him.” It is what St. Paul describes in the
text, “Your life is hid with Christ in God.” There is thus a
fundamental continuity between the state of the ordinary Christian
and that of the mystic, or even of the saint in heaven who rejoices
in the Beatific Vision, for, as St. Thomas says, ‘‘Grace is nothing
else than a beginning of gloryin us.” Because of our finitude,
however, this-capacity fora» quasi-experimental knowledge of
God cannot be realized by our own efforts, but only by the
deliberate act of God himself. When, where and upon whom he
will think fit to bestow this gift is known to him alone.
1 Christian Mysticism, p. 148. 2 Freedom and the Spirit, p. 194.
3 The Mediator, E.T., p. 110, n. 2.
* The Nature and Destiny ofMan, I, p. 61 f. (cf. pp. 134, 145).
5 It must be noted that, while there is a real continuity (and indeed, according to
many mystical theologians, an identity of essence) between mystical union and the
Beatific Vision, mystical union, like all earthly knowledge of God, is knowledge by
faith, not by sight. Cf. the teaching of St. John of the Cross and Mother Cecilia as
discussed by E. I. Watkin in The Philosophy of Mysticism, p. 214 f. Only in heaven is the
| veil entirely taken away; identity of essence is not incompatible with diversity of
» mode.
6 §. Theol., II II, xxiv, 3 ad 2.
CHAPTER XI
THE COSMOLOGY OF WHITEHEAD
HE outlook which has been adopted in this book has been,
in the broad sense of the word, cosmological ; that is to say,
in considering the problem of the existence of God, we have.taken._
as our starting-point the cosmos,.the world or universe of which we
ourselves are part. Recent years have seen two quite outstandingly
portentous cosmological investigations, in the work of Dr. A. N.
‘Whitehead and in that of Dr. F. R. Tennant. It will therefore be
well to add to our discussion a comparison of the position at which
we have arrived with those of these two distinguished thinkers, so
that points of agreement may be registered and an attempt made
to account for any differences that may appear.
The definitive statement of Dr. Whitehead’s philosophy is to be
found in his Gifford Lectures delivered at Edinburgh in 1927-8
and published in 1929 under the title Process and Reality: An Essay
in Cosmology. Its groundwork was laid as far back as 1919, in two
books entitled An Enquiry concerning the Principles of Natural Know-
ledge and The Concept of Nature, while it was immediately preceded
by three other works which appeared between 1926 and 1928,
namely, Sczence and the Modern World, Religion in the Making and
Symbolism, its Meaning and Effect. No substantial development is
apparent in the later works, Adventures of Ideas, which was pub-
lished in 1933, and Modes of Thought, published in 1938.1
~ It is interesting to reflect that the cosmology of Whitehead and
‘ the work of the logical positivists both derive from the monu-
mental work, Principia Mathematica,which appeared in three large
volumes from 1910 to 1913 as the joint product of Dr. Whitehead
-and Mr. Bertrand Russell, although the final conclusions at which
Whitehead has arrived are of the kind which every good logical
positivist is in honour bound to reject as ‘‘metaphysical’’ and
hence meaningless.
Like all other cosmologists, Whitehead is confronted with the
problem of a universe in which multiplicity is interlocked with
1 In the present chapter these books will be referred to in the footnotes by their
initials. Attention may also be called to Miss D. M. Emmet’s valuable book, White-
head’s Philosophy of Organism and to the two articles by Mr. S. E. Hooper in Philosophy,
July 1941 and January 1942.
150
THE COSMOLOGY OF WHITEHEAD I5I
unite ond persistence with fluidity, but in contrast to many he
makes the most determined efforts to see that none of these data
are neglected for the sake of the others. It is thus very difficult to
know whether to describe him as a pluralist or as a monist, as a
Parmenidean or as an Heraclitean. There is hardly a philosopher
of note from whom he does not borrow, or whom he does not
correct, but his main affinities are with the group of seventeenth-
and eighteenth-century philosophers who stretch from Descartes
to Kant. He designates his system by the name “The Philosophy
of Organism.”
ne Supesenis notion of Whitehead’s philosophy is that of
of,sorepresumably for its dynamic implications (though it may
_be observed, in passing, that, for the philosophia perennis, being is
anything but lifeless, since the supreme Being is God, who is Life,
Power and Love). Creativity is the “principle of novelty..? It is
“that ultimate principle by which thé many, which are the
universe disjunctively, become the one actual occasion, which is
the universe conjunctively.”’ For ‘‘it lies in the nature of things
that the many enter into complex unity.”! We are reminded here
(though not by Whitehead) of the scholastic doctrine of poten-
tiality and act.
The units out of which the universe is made up are cé
“actual
entities’’or ‘‘actual occasions’’; these are, as it were, atoms of
creativity. But, though atoms, they are not isolated; in each of
them all the others are mirrored, either in positive prehensions
as being relevant to the prehending entity or in negative pre-
hensions as being discarded.?, However, the actual concrescence
of an entity is not a purely passive reception by it of influences
from other entities; it is itself self-creative, and what it ultimately
becomes on emerging into concrete existence depends upon its
own subjective aim. And out of this perpetual self-creation and
interdependence of actual entities the history of the physical world |
is
s built up. 3
1 P.R., p. 28.
2 Elsewhere Whitehead writes: “The organic starting-point is from the analysis of
process as the realization of events disposed in an interlocked community. ‘The event
is the unit of things real.” (S.M.W., p. 212 (178). Cf. M.T., p. 205 f.) Again, “it
belongs to the nature of a ‘being’ that it is a potential for every ‘becoming’ ”’ (the
“principle of relativity” (P.R., p. 30). It must be noticed that for Whitehead
“relativity” does not mean “‘non-absoluteness” so much as “uniyersal relatedness.’’)
His meaning of the word has a different nuance to that which it bears in’thé famous
“Theory of Relativity.”
152 HE WHO IS
It must be observed that, according to Whitehead, actual
entities are atomic in their temporality as well as in their spatiality.
Their creationis a perpetual perishing ;; they do not persist in time,
for-witht
time
ime they die. The. notion ofa® substance (ong
physical objects are simply ropical constructs from sequences of
particulars, isolated sense-data strung together somehow or other.
Whitehead’s view is somewhat similar, but, through his doctrine
of interrelation ofactual entities, he is able to maintain that when
a nexus of actual entities has what he calls‘social: order” each
ig
entity absorbs into itself, by a ‘kind of“inherit nce,”the. defining
characteristics of its predecessors, so that, while the individual
emia perishes, its'character persists.? (This is, of coursé, as true, in
Whitehead’s view, of lifeless as of living objects, but it is not
difficult to see here his predilection for the Buddhist rather than
the Christian doctrine of Imire asserting itself.) Such a
nexus he calls an “enduring object,”28.but. he.does not seem to
to get “persons” out ‘of | “occasions” by the notion of “ aaa
ference of energy.”
_We cannot, however, he asserts, account for an ordered universe
simply by means of actual entities. An aggregate of actual entities,
each of which was pure creativity, would be formless chaos. ‘The
understanding of actuality requires a reference to ideality.”
There must therefore be a realm of ‘“‘forms”’ or ‘‘ eternal objects,”’
which actual entities prehend as well as prehending one another.
So out of actual entities and eternal cbjects Whitehead proceeds
to build up the world, and four-fifths of Process and Reality is
devoted to the description of the way in which, from these two
fundamental types of entity, in which creativity manifests itself,
the universe of our experience emerges. It is not necessary for our
present purposes to go into this in detail.
Once the category of eternal objects has been introduced, the
question of their nature inevitably arises. They cannot have any
concrete existence of themselves, or they would be actual entities:
“the metaphysical status of an eternal object is that of a possibility
for an actuality.”’® And Whitehead is far too much of an empiricist
p. 83 supra.
2 Fe this reason Whitehead is able to make a more satisfactory attempt than
Russell at constructing a theory of induction. (P. Re p. 288 f.; cf. M.T., p. 133 f.)
BPR pr aee 4 AT, ch. xi. 5 S.M.W., p. Q21 (185).
6 S.M. W., p. 222 (186).
THE COSMOLOGY OF WHITEHEAD 153
to attribute reality to pure abstractions. His eighteenth Category
of Explanation, the “Ontological Principle,”’ insists that “actual
entities are the onlyreasons’ ;1 “all real togetherness istogether-
ness in the formal constitution of an actuality,””72~Where, then,
arethe eternal objects? In the actual entities, we ‘might Teply;
and so in a sense they are, for the actual ‘entities prehend them
and they are ingredient into the actual entities. But if that
is all that is to be said, we are back where we started, with
the actual entities, and we agreed that actual entities by them-
selves would produce mere chaos, not a cosmos. What is the way
out?
“Ttis at this point that Whitehead introduces a unique actual _
entity whose function it is to be the non-temporal locus.of all.tne F°
eternal objects; and to it he gives the name God. Actual entiti
~~.
~-prehend the eternal objects through the valuation of those objects
by God. “The differentiated relevance of eternal objects to each
“instance of the creative process requires their conceptual realiza-
tion in the primordial nature of God. He does not create eternal
objects ;for his nature requires them in the same degree that they
require him.’* ‘‘Apart from God, there could be no relevant
novelty. Whatever arises in actual entities from God’s decision,
arises first conceptually, and is transmuted into the physical
world.”’4
Thus the cosmological scheme, in its fundamental elements, is
complete. In Religion in the Making, Whitehead sums it up as
follows :
“The temporal world and its formative elements constitute for
us the all-inclusive universe.
“These formative elements are :
“1. The creativity [embodied in actual entities] whereby the
actual world has its character of temporal passage to novelty.
@“2.)The realm of_ideal entities, or forms, which are in them-
selves not actual, but are such that they are exemplified in every-
thing that is actual, according to some proportion of relevance.
‘3, The actual but non-temporal entity whereby the indeter-
mination of mere creativity is transmuted into a determinate
freedom. This non-temporal actual entity is what men call God.
—the supreme God of rationalized religion.’
PRAT: 33- 2 P.R., p. 44. 3 P.R., p. 363; cf. p. 353. # P.R., p. 229.
154 ) =HE, WHO.IS
Again he writes:
“God_is that non-temporal actuality which has to be taken
account of in every creative phase..
“Godi3 the one_sys
systematic, complete fact, which is the ante-
cedent ground conditioning every creative act.
“He transcends the temporal world, because heiis an actual
fact in the nature of things. He is not there as derivative from
the world; he is the actual fact from which the other formative
elements cannot be torn apart.’’}
Are we then to hail Whitehead as a Christian theist because he
thus introduces a being (or, as he would say, an actual entity) to
which he gives the name God? We may well hesitate to do so
when we see what he has to say about God. Let us turn again to
Process and Reality, ‘There we read:
“It is as true to say that God is permanent and the World
fluent, as that the World is permanent and God is fluent.
“It is as true to say that God is one and the World many, as
that the World.is.one and God many, —
“Tt is as true to say that, in comparison with the World, God
is actual eminently, as that, in comparison-with God, the World
is actual eminently. :
“Tt is as true to say that the World is immanent in God, as that
God is immanent in the World.
“Tt is as true to say that God transcends the World, as that
the World transcends God.
“It is as true to say that God creates the World, as that the
World creates.God.
“God and the World are the contrasted opposites in terms of
which Creativity achieves its supreme task of transforming dis-
joined multiplicity, with its diversities in opposition, into_con-_>
crescent unity, with its diversities in contrast.... For God
the conceptual iisprior to the physical, for the World. the physical
poles aré prior to the conceptual poles:.
“God and the World stand over souiast each other, expressing
the final metaphysical truth that appetitive vision and physical
enjoyment have equal claim to priority in creation.’’?
Again we are told that “God isthe primordial creattre,’? and
that “‘God’s existence is not generically different from that of other
actual entities, except that he is primordial in a sense to be
* R.M., pp. 94, 154, 156. 2 P.R., pp. 492, 493. _ A PR p. 42.
THE COSMOLOGY OF WHITEHEAD 5 155.
gradually explained.”’! We have certainly travelled far from the ,
Thomist doctrine of God as self-existent Being.
For Whitehead’s God is: essentially an evolutionary deity.
Although-he-is in.a.certain sense superior to the world, he is very
much dependent. upon..it,.. and the historic process is a kind of
mutual adventure in which God and the world are taking part.
“The World’s nature is a primordial datum for God; and God’s
nature is a primordial datum for the World. Creation achieves
the reconciliation of permanence and flux when it
t has reached its
final term which is everlastingness—the Apotheosis of the World.
. Neither God, nor the World, reaches static completion. Both
are in the grip of the ultimate metaphysical ground, the creative
advance into novelty. Either of them, God and the World, is the
instrument of novelty for the other.’
Theresseems to be an inconsistency here. God was described
as a non-temporal actual entity, presumably because temporal
actual entities are inevitably perishing; yet here we are told that
he himself develops as time goes on. Nor can the reply be made
that, while God is himself unchanging, from the standpoint of a
.. being within the changing world he appears to undergo change.
_ This might be asserted of a transcendent.God, but the Philosophy
of Organism knows nothing)of a transcendent God. “Any proof,”
writes Whitehead, “which commences with the consideration of
the character of the actual world cannot rise above the actuality of
this world. It can only discover all the factors disclosed in the world as
experienced. In other words, it may discover an immanent God,
but not a God wholly transcendent.”? ~The Thomist notion of a.
pears quid é:est, as contrasted with a demonstration quid est, »
is thus rejected, In Ths r primordial nature’’ which me possesses
“apart from the world, God is not even conscious. ‘“‘Viewed as
primordial, he is the unlimited conceptual realization of the
absolute wealth of potentiality. ... As primordial, so far is he
from ‘eminent reality’ that in this abstraction he is ‘deficiently
actual.’... Thus, when we make a distinction of reason, and
consider God in the abstraction of a primordial actuality, we must
ascribe to him neither fullness of feeling, nor consciousness.’’4
“The consequent nature of God is conscious ; and it is the realiza-
tion of the actual world in the unity of his nature, and through
the transformation of his wisdom.’® ‘‘The completion of God’s
1 P.R., p. 103. 2 PR.
wis p..
D:19% 3 R.M., p. 71 (italics not in original),
4 P.R., p. 486. 5 P.R., p. 4
156 HE WHO IS
nature into a fulness of physical feeling is derived from the
objectification of the world in God.”! (We are reminded of
Dr. Matthews’s assertion that without a world God could not be
personal. )
It is important not to be filed by Whitehead’s description of
God as “the principle of concretion.”* This does not mean that
he*is thé Creator; for everyactual entity, as a manifestation of
Creativity, creates ‘fisell As Miss D. M. Emmet says, ‘Whitehead
would not agree with St. Thomas that creation proceeds by
emanation from the primordial cause, God, but would only claim
that God provides the final causation for hie self-creation of actual
entities, and also the initial limitation upon mere creativity in
virtue of which there can be any order, or process of creative
advance whatsoever.”? In one place after another, Whitehead
>
E|
makes it plain that, in his view, God ‘merely provides the world
with the vision which it is tofollowin its creation of itself. “The
power
by which God sustains
the world is the power of himself as
the ideal. He adds himself to the actual ground from which every
creative act takes its rise. The world lives by its incarnation of,
~ God in itself. . . He is not the world, but the yaluatio
“world. In qostrachon from the course of events, this valuation is
“a necessary metaphysical function. Apart from it, there could be
no definite determination of limitation required for attainment.
But in the actual world, he confronts what is actual in it with what
is possible for it. “Ths he solves all indeterminations.’’* “The
metaphysical «doctrine, “here expounded, finds the foundations of
the world in the esthetic experience, rather than—as with Kant—
in the cognitive ‘and ¢conceptive experience. All order is therefore
esthetic order, and the moral order.is merely certain aspects of
esthetic order."The actual world is the outcome of the esthetic
y order, and the esthetic order is derived from the immanence of
\Gods" “God is the measure of the esthetic consistency of the
‘world. There is some consistency in ¢réative action, because it is
conditioned by his immanence.’’ It is true that in the earlier
book Science and the Modern World there are passages that suggest
leanings to a more orthodox view,’ but these are not returned to
1 P.R., p. 488. 2 $.M.W., p. 243 (203); Pa SNP: 345.
8 Whitehoad’s Philosophy of Organism, p. 121. ER.M., pp. 156, 159.
Ook. M.epa tod. SOR: » P. 99:
7 SM. W., PP. 249-51 (207- 8). But even here the “habit of paying to [God]
metaphysical compliments” is condemned.
°
THE COSMOLOGY OF WHITEHEAD 157
in Process and Reality. There we are told that “each temporal
entity,
in one sense, originates from its mental pole, analogously
to God himself. It derives from God its basic conceptual aim,
relevant to its actual world, yet with indeterminations awaiting
its own decisions.”? ‘‘He is the unconditioned actuality of con-
ceptual feeling at the base of things; so that, by reason of this
primordial actuality, there is an order in the relevance of eternal
objects to the process of creation.” In his pumordal nature © \
/“he As_the ture forfeeling, | the.eternal urge. of desire,”® while
Vea the “great companion—the fellow-sufferer who ‘under. J
/
nds. 994 7 i
“Tt will be clear from this that Whitehead’s doctrine of God is a
doctrine ofimmaneénce; this isnot altered by such remarks as “It
is as trué to say that God transcends the World, as that the World
transcends God,” or “He transcends the temporal world, because
he is an actual faét in
i the nature of things.”’® God is not needed by
Whitehead as the self-existent infinite Being upon whose love and
power the world depends for all that it is and has; he is needed |
merely as the locus of the eternal objects without which actual =~
occasions would not be able to effect their own self-creation. God
provides for them a finalCause ‘of an esthetic kind; it is note-
worthy that in one place Whitéhead quotes with general approval
the famous passage from Aristotle about God moving the world
as the object of its desire.6 God also appears as some kind of.
formal cause, since the eternal objects of which he is the locus are
ingredient into the actual occasions which prehend them. But he is
certainly not, in Whitehead’s thought, what Christian theism
asserts that he is, namely the creative efficient cause of the total |
being of everything other than himself. It is therefore hardly sur- |
prising that Whitehead’s final attitude to.God.isnot.one.ofadora- °
tion so much as of sympathy... T. E. Hulme has remarked that it is
at the end of their discussions that philosophers usually give
themselves away. ‘‘If you ask,”he writes, ‘“‘what corresponds to the
pantry which betrayed the man in armour, I should answer that
it was the /ast chapters of the philosophers in which they express
their conception of the world as it really is, and so incidentally
expose the things with which they are satisfied. How magnificently
1 P.R., p. 317. 2 PLR Pp .4480; :P.R., p. 487.
4 P.R.».P2497- 5 R.M., p. 156. P.R., p. 487.
158 HE WHO IS
they may have been clad before, they come out naked here.”? It
is in his fina] description of God as “the great companion—the
fellow-sufferer who understands” that Whitehead reveals the
essential immanentism of his thought.”
But it is important not merely to point out the inadequacy of
Whitehead’s system from the point of view of Christian theology,
but also to show precisely why this inadequacy has arisen. Unless
we can do this we shall merely have done something towards
discrediting Christian theology. It appears, as a matter of fact,
that, for all its po Whitehead’s..cosmology--suffers-from
in the sense “of Self-creativeness, can be taken to be the funda-
mental characteristic of finite beings, and that, in consequence,
whatever may be needed in the way of a God to provide what is
lacking in them as an esthetic final cause or formal cause, they
have no need of any efficient cause outside themselves. “The
concrescence of each individual actual entity is internally deter-
mined and is externally free.”? But-this simply means that
Whitehead has concentrated_his thought. so thoroughly upon.the i
way in which things behave as never really to inquire why they are.
He has never properl y understoodwhat finitee being ts and so has ~
never-apprehended-its radical contingency. 4. This may be in part
due to the highly individualistic and questionable nature of his
metaphysical doctrine, but it can hardly be entirely the result of it.
For even if the account of finite beings which is given in Process
1 Speculations, p. 20. On the previous page Hulme writes: “A man might be clothed
in armour so complicated and elaborate, that to an inhabitant of another planet who
had never seen armour before, he might seem like some entirely impersonal and
omnipotent mechanical force. But if he saw the armour running after a lady or eating
tarts in the pantry, he would realize at once, that it was not a god-like or mechanical
force, but an ordinary human being extraordinarily armed. In the pantry, the
essence of the phenomena is not arms, but the man.”
2 We may add that, while, for the reasons suggested below, Whitehead’s God falls
short of the demands of a thoroughgoing theism, he nevertheless yields more to those
demands than Whitehead’s philosophical doctrine would seem to allow. Professor
L. S. Stebbing, in a very critical review of Process and Reality, asserts that ‘‘ Professor
Whitehead’s indefensible use of language becomes nothing short of scandalous when
he speaks of ‘God’.” (Mind, October 1930, p. 475.) And in view of Whitehead’s own
description of God’s existence as “the ultimate irrationality” (S.M.W.,p. 249 (208) ips
it is a little surprising to find him criticizing the positivists because they “work
[themselves] into a state of complete concen with an ultimate irrationality.”
(M.T., p. 202.) shalahseo
4 Itis noticeable, for example, that in The "Concept of Nature, ch. i and ii, in spite of
a most salutary insistence that natural philosophy is concerned with data that are
external to the mind, Whitehead goes on to discuss the structure of these data without
any investigation of their ontological status.
THE COSMOLOGY OF WHITEHEAD 159
and Reality be accepted, we still want to know why they are there
at all. To say that they are units of Creativity explains nothing,
unless we are also told why they are the particular units that they
are, manifesting the particular kinds of Creativity that they do.
The one thing that they are clearly not is self-explanatory, but
this is the one fact that Whitehead never allows himself to think
about. He postulates God as the ground of rationality,* but never
as the ground «of
being, and as a result his whole system ‘is left
bombinating in a_vacuum,
The second deféct is, it seems, not unconnected with the first,
and may indeed partly explain it. Whitehead apparently has no
conception of the doctrine of analogia entis. The result is that,
having found it necessary to introduce a being which
he calls
“God’” order
in to «compensate for a certain evident Iack of self-
explanation in the world, ke then assumes that God must be
dependent upon the Tarend correlatively to its dependence upon
a
im. His God is thus both evolutionary and finitey “It is as true
o say that God creates the World, as that the world creates
od”’ ; ‘a process must be inherent in God’s nature, whereby his
infinity is acquiring realization.”? As we have already seen,
without the doctrine of analogy it is impossible to provide an
explanation of the existence of finite being, for a God who is
limited by the world no more provides a reason for his existence
than the world does for its own. This has already been expounded
at some length in chapters vi and viii; there is no need to repeat
the contents of those chapters here. All that need be remarked
is that, since it is impossible to answer the question “Why does
the. world exist?’”? without the doctrine of analogy, the fact that
Whitehead Has fio stich doctrine may be one reason why he refrains
from discussing that question. Be that as it may, he certainly fails
to discuss it; and, as it is the ultimate problem of cosmology, this
must be taken as showing a radical deficiency in his system.
For it is a deficiency rather than a positive error. All that
Whitehead has to say about God’s relation to the world as its
formal and its esthetic final cause may be accepted in substance
as a restatement of the Christian doctrines that the ideas of all
created beings are to be found in a more eminent mode in God and
iceoe re toms 1toaes as its last end. And if Whitehead
15.M.W., p. 249 ei
2 A.I., p. 357. Cf. the discussion of “Deity” as “that factor in the universe whereby
there is importance, value, and ideal beyond the actual.” (M.7., p. 140.)
160 HE WHO IS
had taken the great tradition of Christian scholasticism as seriously
as he has taken the ancient Greeks and the post-Cartesians he
might have gone far towards providing the modern philosophical
world with the synthesis which it so urgently needs. Professor
A. E. Taylor has given it as his opinion that ‘‘Dr. Whitehead’s
work would be even better than it is if it were influenced a little
more by St. Thomas anda little less by Spinoza.’’! By taking as
his fundamental category a concept of ‘“‘Creativity’’ which is
essentially becoming, rather than a concept of being which, in its
analogical fecundity, bridges the gulf between the finite and the
infinite, he has condemned his cosmology to imprisonment within
the realm of finite existence and has thrown away the only key
which could release it. Professor Taylor accuses Whitehead of
“unconscious tampering with his own sound -priiciple that all
possibility is founded_on.actuality,”’ and remarks that “the attempt
to get back somehow behind the concreteness of God to an élan
vital of which the concreteness is to be a product really amounts to
a surrender of the principle itself.’’? But, in spite of its wrong-
headedness, the Philosophy of Organism of Dr. Whitehead remains
as one of the most impressive and massive cosmological construc-
tions that have been produced, and the Christian theologian may
learn much from it, not merely through the valuable positive
insights which it manifests but perhaps even more through its
ultimate inadequacy.
1 “Some sien gh on Process and Reality,” in Theology, August 1930, p. 79.
2 Art. cit., p. 78.
CHAPTER XII
THE COSMIC TELEOLOGY OF TENNANT
E now proceed to consider the approach which is made by
Dr. F. R. Tennant and which he elaborates with extreme
thoroughness in his great two-volume work, Philosophical Theology.
He is one of the most radically empirical and genetic of modern
philosophers in method, and he states his view of the whole duty
of the philosophical theologian in the most uncompromising
way. ‘The classical proofs of the being of God,” he writes,
“sought to demonstrate that there is a Real counterpart to a pre-
conceived idea of God.... The empirically-minded theologian
adopts a different procedure. He asks how the world, inclusive
of man, is to be explained. . . He will thus entertain, at the
outset; no~ such AEE positions as that the Supreme Being, to
which the world may point as its principle of explanation, is
infinite, perfect, immutable, supra-personal, unqualifiedly omni-
potent or omniscient. The attributes to be ascribed to God will
be such as empirical facts and their sufficient explanation indicate
or require.” ‘Tennant is the irreconcilable antagonist of all
attempts to deduce the existence of God from purely logical
premisses or from the bare concept of abstract being; St. Anselm
and Hegel are equally rejected. But it may be doubted whether
he has taken sufficiently seriously the classical tradition of
scholasticism with its doctrine of analogia entis ;certainly he neglects
to give any systematic discussion of the presentation of theism by
St. Thomas, and the seven references to the Angelic Doctor which
occur in his work are all related to merely incidental points. But
the relevance of these remarks will become clearer later on 1m this
chapter.
In his first volume, which occupies nearly two-thirds of the
whole work, Tennant makes an extremely thorough examination
of the basic philosophical principles upon which any consistently
empirical theology must rest, and much gratitude is due to him for
performing this task; his epistemological doctrine is also ex-
pounded, more shortly and with less elaboration, in his later book,
Philosophy of the Sciences. We must, he insists, start from our own
1 Phil. Theol., II, p. 78.
161
162 HE WHO IS
experience, for we have nothing else to start from; the ordo
cognoscendi must precede the ordo essendi. And he makes a most
valuable distinction between that which is ‘“‘psychically”’ immedi-
ate to us in our crude acts of perception and that which is recog-
nized as “‘psychologically”’ immediate after subsequent reflection
upon those acts.!_ Much error, he points out, has arisen from the
neglect of this distinction (“the psychologist’s fallacy’), and
throughout his work the distinction between the psychical and the
psychological, denoted by (y) and (fs) respectively, is never
allowed to be forgotten, He plainly affirms that our knowledge of
the world is derived from the impressions of particulars received by
our senses. ‘‘There is no psychological or scientific basis,” he
writes, “for the opinion that universals exist save in rebus, or for
identifying the valid with the Real. ... Universals are obtained
by subjects from the particulars in which they are implicit; we
know of none that, in the last resort, are not so obtained. In the
order of knowing, they presuppose thought; not thought, them.
Consequently, universals are not known by acquaintance, or with
(ps) immediacy; the (y) immediacy, with which they eventually
come to be explicitly apprehended, is an outcome of process and
practice.”»2 Proceeding from this starting-point, he builds up, by
a very penetrating psychological discussion, a most impressive
case for the existence of the human soul as a persistent knowing
and willing entity; both Russell’s doctrine of the soul as merely a
logical construct out of particulars and Whitehead’s theory of
nexus with “social order” receive implicit refutation here.* “The
soul,” he-writes, “cannot be™ phenomenal. It is that to which
phenomena appear, and is known otherwise than is the pheno-
menal. It is rather the one known being that must be called ontal
or noumenal, if we are to avoid indefinite regress ;or the one ontal
thing that is assuredly known.” He does not, it is true, assert, as
Gardeil does, that the soul, in reflection upon it acts, has an imme-
diate, though obscure, experimental perception of its own essence ;®
“the apprehension of it is mediated discursively,” he says, and
“by construction rather than by pure inference.’® But of the fact
1 Phil. Theol., I, p. 46.
2 Phil. Theol., 1, p. 64. It is important to note that, when words like “real” and
“object” are spelt with a small initial in Tennant’s discussion, they refer to the content
of immediate experience of percipient subjects; when spelt with a capital they refer
to the “common world” of many percipients, whether derived from that content by
oS, construction or assumed to exist in its own right behind the phenomena.
28°Phil Theol., 1, ch. v. See pp. 83, 152 supra. 4 Phil Theol., I, p. 97.
5 Gardeil, Structure de V’Ame, Il, p. 112 f. 6 Phil. Theol. i, Page:
THE COSMIC TELEOLOGY OF TENNANT 163
he has no doubt, and, while he asserts that “‘ Kant was right in
teaching that internal experience is only possible through external
experience, he agrees with Descartes and Locke that“we have
clearer and [frmorecertain ky
}10wledge. of our.own existence than we
have of the existence of things.’”4,
“A discussion of Tihical Walbe is
j succeeded by a chapter on
Thought and Reason, “in which the categories of substance and
cause are vindicated. ‘Substance, as unknowable substratum,
can well be spared; ... but the concept of a substance remains
indispensable. It assigns the ground of the conjunction of parti-
culars, which resort to logical concepts, such as class or series,
simply ignores. It is the determinedness, as to order, of our sensa,
not they themselves or their mere occurrence (which might con-
ceivably be fortuitous but is not), that suggests and calls for an
interpretative concept. Thus the concept is not fashioned by the
mind itself without objective call.”? Again, “‘the concept of
constancy of sequence may not logically imply that of necessary
connexion or determination ; but the fact of constancy of sequence,
and of the non-emergence of effect unless the cause be forth-
coming, bespeaks or suggests a sufficient ground. As reasonable
men, we cannot dispense with the causal category as thus expres-
sing determination of one event by another, whatever we may do as
rational logicians.’’* And the accusation of “‘anthropomorphism”’
is faced cheerfully. ‘““When we adopt the anthropic category of
cause, we may be contenting ourselves with analogy where we
cannot have logical cogency: with the ‘regulative’ where no
‘constitutive’ function is forthcoming. But if one then sneakaina. \
a man,’ one at least knows what one is doing and what one ise]
talking of.’’4
There follows a discussion of the theory
of knowledge, which is,
in many respects, of quite extraordinary value. All attempts to
make the act of perception terminate with the sensum, whether that
be conceived as a state of the perceiving soul, as a mere isolated
particular, or as a “permanent possibility of sensation,’ are
rejected, and the Kantian doctrine of an unknowable thing fer se
is found to be grossly inadequate. “While we... have acquain-
tance only with the sensible or phenomenal, we areabate rapport
with, and phenomenal knowledge of or about, the noumenal.
\1 Phil. Theol., 1, p. 78. This last statement is not, of course, altogether in agreement
ith the position maintained in the present work; see p. 145 supra. But what we are
concerned with here is simply Tennant’s defence of the existence of the soul.
2 Ibid., I, p. 179. 3 Ibid. I, p. 181. Ibid., I, p. 181.
M
164 * HE WHO Is
We do not, in this latter sense of ‘knowledge,’ know the pheno-
menal only; rather, we know the noumenal through the pheno-
menal.... The thing per se, we conclude, must be credited with
far more of responsibility and character than Kant,,influenced by
rationalistic and idealistic propensities,meted out to it.”2 It will
be seen how close Tennant is here to the scholastic doctrine of the
species as being the medium rather than the terminus, the quo
rather than the quod, of perception ;nor must we be misled by the
fact that he called his theory of knowledge by the name of
phenomenalism. ‘For by this he does not mean that phenomena
are all that we know, but that phenomena are phenomena, that is to
say, appearances of noumena which they reveal. “We have,”’ he
says, “‘already substituted for the dictum that we only know
phenomena, the more correct statement that we know the
noumenal through the phenomenal. The phenomenal is, so to
say, the utterance of the ontal to us; if the noumenal shines forth, ESNIPS OOD SE
or appears tous, as the phenomenal, it cannot be totally unknow-
able. And why should the appearing beassumed tobe a veiling,
rather than a revealing, a distortion or caricature, rather than
representation? Philosophers have been wont to assume that
phenomena only serve the purpose of deception. But their
pessimism is groundless.’’?
Tennant next passes on to consider induction and probability,
and it is here that we find the first serious divergence from the
position that has been taken up in the present work. Agreeing
substantially with Keynes’s treatment of induction and proba-
bility, he remarks that probability, in order to be amenable to
logical and mathematical handling, must be defined as a relation
between sets of propositions; no proposition is probable or
improbable in itself, but only in relation to certain other proposi-
tions. This raises various difficulties, in particular that of relating
this purely objective relation between propositions to the fact that
the probability with which we are concerned in actual life involves
our purely subjective knowledge or ignorance of them. And the
crux of the matter appears when we ask what is the basis of the
probability which we are forced to ascribe to the ultimate body
of knowledge relative to which the probability of all our other
beliefs is to be assessed, since, ex hypothesi, there can be no ‘‘more
ultimate’? body of knowledge to which it can be referred. ‘If
probability is rational, the rationality is in turn problematic.
1 Phil. Theol., 1, pp. 247, 248. 2 Thid., I, p. 252.
THE COSMIC TELEOLOGY OF TENNANT 165,
Probability becomes a logical relation to probabilities.” Tennant
then concludes that purely logical probability is not the probability
with which we are concerned in life, and that, if the process of
induction, which is essential to human action, is to be justified, it
must be on the ultimate basis of a probability which is, to use his
term, alogical.*
It will be remembered that in chapter vii above, we main-
tained, in substantial agreement with Fr. M. C. D’Arcy, that,
while induction cannot be justified as a purely logical derivation
of propositions from one another, it can be justified if we assume
that, by its very nature, the human mind has the capacity (which,
like every other human faculty, may of course sometimes function
erroneously), when confronted with an adequate body of empirical
fact, of apprehending, however partially and obscurely, the nature
of finite beings, and hence of understanding to a greater or less
degree the laws of their behaviour. Tennant does not, however,
take this course,? presumably for the reason that, while, as we
have seen, he thinks that Kant did not go far enough in attributing
to us knowledge of things fer se, he never fully accepts the doctrine
that, through the phenomenal, we perceive the noumenal; he
treats the ‘“‘so-called phenomenal World of common sense and gpape
nee,
1 Phil. Theol., I, p. 283. Cf. Tennant’s small book, Miracle, p. 19 f.
2 We may note that, while Professor C. D. Broad is on the whole ready to agree with
Tennant’s doctrine of alogical probability (see review of Phil. Theol., Vol. 1, in Mind,
January 1929, p. 99), Professor J. Laird considers it to be “‘really incoherent” and to
involve a vicious infinite regress. (See review of Phil. of the Sciences, in Mind, January
1933, P- 111.)
3 Thus he writes elsewhere: “I would submit that the proposition which we are
able to assert without possibility of refutation is not that there must be some certain
knowledge [sc. of actuality, and not merely of logical implications] which even the
sceptic presupposes, but that we possess presumptive knowledge, the ultimate pre-
suppositions of which may in turn be found to be presumptive” (Phil. of the Sciences,
p. 38), and he goes on to elaborate an epistemological doctrine on the basis of this
notion of “‘presumptive knowledge” (op: cit, lect: ii); But, we must reply, if human
knowledge is, of its very essence and not only accidentally, presumptive, there is no
justification for the investigations by which we attempt to discriminate between
genuine knowledge and error; indeed, the notion of human knowledge as being
essentially and universally presumptive would seem to be meaningless. The fact
surely is, not that we have presumptive knowledge, but that.we presume that_we have know-
ledge. If presumptive knowledge” be a convenient term to indicate this, we may
accept it, but if all our knowledge of actuality is presumptive, we have no possibility of
discriminating between presumptive knowledge which is true and presumptive know-
ledge which is false. And we may add that, while our () certitude of truths arrived
at by induction is different from (ps) certainty, our (ps) reflection upon it reveals, not
that it rests merely upon alogical probability (though this factor need not be altogether
excluded), but upon an intuition of the nature of finite essences. Tennant’s genetic
approach to knowledge, which he expounds so impressively, would seem to supply a
very useful tool for discovering when the intellect is in error, but does not, it would
seem, succeed in replacing ‘‘knowledge”’ by ‘‘ presumptive knowledge” as the faculty
of the intellect.
166 HE WHO IS
conceptual science” as something inevitably made by us “‘out of”
our various private perceptual worlds and the realm of things
per se, and not as, in the scholastic sense of the word, an abstraction
from the latter realm, by which (apart from the possibility of error:
- due to sheer human fallibility) it is genuinely manifested to us,
however partially, in its actual nature;1he speaks of “manipula-
tion of our knowledge-data’’? rather than of “abstraction,” and
when he does speak of ‘‘abstraction”’ it is rather in the sense of
“construction.”’? At this point Tennant imports into his system
» the idea of “faith”’ as essential to epistemology. He declares that
his faith is essentially of the same kind as the faith of religion, and
suggests that to the catalogue in the eleventh chapter of the Epistle
to the Hebrews we might add “By faith, or by hope, Newton
founded physics on his few and simple laws of motion; by faith
the atomists of ancient Greece conceived the reign of law through-
~out the material world; and so on indefinitely.’’4
Without inquiring how far this conception of faith is identical
with that of Christian theology, or indeed with that of the New
Testament, we may observe that at this poit there occurs the
one considerable break in the rationality of Tennant’s system.
Faith,” he tells us, ‘‘is venture dictated by human interest: it is
not mere prudence or probability, for these cannot be, till faith
has substantiated somewhat of the hoped for; it is not confined to
the realms of moral value and religious ideas, but infects all existen-
tial and theoretical knowledge. ... Science postulates what is
requisite to make the world amenable to the kind of thought that
conceives of the structure of the universe, and its orderedness
according to quantitative law; theology, and sciences of valuation,
postulate what is requisite to make the world amenable to the kind
of thought that conceives of the why and wherefore, the meaning
or purpose of the universe, and its orderedness according to
teleological principles. Both are necessarily interpretative, anthro-
,pic, interested, selective.”> And in an eloquent passage he argues
ithat, on the basis of this doctrine of faith, theology and science
‘have the same epistemological
Pp g status.®
We might doubt whether theology should be grateful for being
placed upon the same epistemological level as science at the cost of
accepting such a doctrine of “‘faith” as this. It must also be
noticed that, in spite of the vigour with which he expounds it, it is
1 Phil. Theol., I, p. 255. 2 Phil. of the Sciences, p. 83. 3 Ibid., p. 133 f.
* Phil. Theol., L p. 298. 5 Thid., I, p. 299. ; 6 Ibid., I, p. 303.
’
THE COSMIC TELEOLOGY OF TENNANT 167
by no-means easy to determine precisely what the nature of
“faith,” as Tennant understands it, is. His explicit repudiation
of “rational’’ theology, by which he means any theological system
resting on a purely logical and a priori basis,1 and, it must be
added, his criticisms of the ordinary arguments from religious
experience as involving a confusion between (w) and (fs)
immediacy,? seem to be well-founded; but whether an empirical
approach to theology necessarily requires the insertion of “‘faith”’
at the point where he makes it seems highly questionable, and some
of its results will be seen as we consider how he proceeds from it to
formulate an argument for the existence of God.
This task is undertaken in the second volume, which opens
with a discussion of the conformity of the world to law and of the
sense in which the world may be said to be “rational,” and then
discusses the empirical approach to theism. Argument of a
primarily ‘“cosmological”’ type, such as St. Thomas’s first three
“Ways, based upon the fact that the world exists although it
does not provide the explanation of its own existence, are rejected,
though by implication rather than by explicit repudiation. There
is, indeed, in an appendix to the first volume an incidental denial
of the argument e contingentia mundi,’ which seems to show a
misunderstanding of the sense which contingency bears in it, but
on the whole it appears that Tennant’s avoidance of arguments
based upon the mere existence of finite being is related to the fact,
mentioned above, of his timidity in accepting wholeheartedly
the real existence of extramental objects as substantial beings.
*“The soul or active subject,” he writes, “‘is the one kind of thing
per se that we have as yet [sc. at the beginning of Volume IT] been
enabled to assert to be knowable. And, previously to inquiry, the
ontal that lies behind or beyond physical bodies and phenomenal
events may be supposed to consist of soul-like monads, or to be of
essentially different nature.’ Even ‘‘other selves . . . are neither
directly apprehended nor provable otherwise than by cumulative
pragmatic verification.” The consequence of this is that
Tennant’s inquiry about the world, instead of being directed to
the question why it exists, is concerned with the way in which it
enstramonnniecoreont
A
behaves; and he is able to argue that the design which he claims |
to see in it implies at least the probable existence of a designer|1
' 1 See, e.g., Phil. Theol., II, p. 79. 2 See p. 19 supra.
3 Phil. Theol., 1, Appendix: Note L, p. 410, 4 Phil. Theol., II, p. 26.
5 Tbid., II, p. 78.
168 HE WHO Is
without having to decide whether its ontological nature iecaadt
the existence of a self-existent Being as its creative cause.
The chapter on ‘‘The Empirical Approach to Theism,” which
forms the heart of Tennant’s second volume, bears the sub- title
“Cosmic Teleology.” The keyword of the argument is “purpose,”
and five fields of Tactare examined iin which purpose has been
claimed aasevident¢_(1)There is first the epist ical_adaptive--
ness of things to th ht, in scholastic language the convenientia
entium ad intellectum. In view of the interpretative function which
he ascribes to the mind, Tennant feels justified in concluding only
that ‘“‘things,..or.their.ontal.counterparts, have so much affinity
/ with us as to be assimilable and to be understood, >but that this
(“does not of itself testify that the adaptedness is teleological. es
A2) Then there is the adaptiveness that is seen in living organisms,
‘concerning which Tennant remarks that the overthrow of the
Paleyan type of argument by the Theory of Evolution does not
deprive the adaptedness of teleological significance. ‘“‘The dis-
covery 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. It has also served
to suggest that the organic realm supplies no better basis for
teleological argument of the narrower type than does inorganic
Nature.”? (3) He next considers, therefore,
the adaptation of the...
inorganic, envifonment.to. organic life. .Against the contention that
' different inorganic conditions from those that actually hold might
have been equally compatible with life, albeit life of a different
kind, he urges that “the necessary environment, whatever its
nature, must be complex and dependent upon a multiplicity of
coincident conditions, such as are not reasonably attributable to
blind forces or to pure mechanism,’’? and that the fact that the
inorganic world is such that life with its purposiveness can emerge
from.it is an indication of purpose in the inorganic world itself.
(4y‘The treatment then proceeds to deal with the esthetic quality
of Nature, and particular stress is laid on the fact that it is just
where Nature is most “mechanical” and least touched by man
with his attendant abominations of factories, motor-cars and the
like that she is most beautiful. That is, Nature speaks to man of a
purpose beyond herself, and, while coercive force is not claimed
for it, this argument is pressed as providing sound confirmation of
the purposiveness that has been already discerned. “ Theistically
1 Phil. Theol., II, p. 83. 2 Ibid., II, p. 85. 8 Ibid., II, p. 87.
THE COSMIG TELEOLOGY OF TENNANT 169
regarded; Nature’s beauty is of a piece with the world’s intelligi-
bility and with its being a theatre for moral life; and thus far the
case for theism is strengthened by esthetic considerations.”?
(5) Finally, TTennant considers the moral status of man. He will
not admit that a sound argument for the existence of God can be
built up on moral considerations alone. For him, they “supply
the coping-stone of a cumulative teleological argument...
Isolated from other facts, and used only for the sake of their
alleged a priort preconditions, they can supply no rational proof
of the existence of God . . .: the valid and the existent have been
confounded.”’? Kant and Rashdall receive short shrift, the latter
being accused of ambiguity in his use of the word “existence” or,
at least, of having illicitly assumed that “‘to exist’? involves ‘to be
in some mind” and of having argued from this that, since “the
ideal is not fully apprehended as to its content by any individual
[or] realized in any human life . . . there must therefore be a
Divine Mind in which its ‘existence’ is to be located.’? For
Tennant the significant fact is that ‘‘ ‘man is the child of Nature.’ ’*4
“Her capacity to produce man must be reckoned among her
potencies, explain it how we may. And man is no monstrous birth
out of due time, no freak or sport [; he] is like, and has genetic
continuity with, Nature’s humbler and earlier-born children,’® ,
“Nature and snGral man are not at strife, but are organically|
one [, and] the whole process of Nature is capable of being
regarded as instrumental to the developement of intelligent and
moral creatures.’’ /
Even from so sketchy and imperféct an outline as the above it
should be clear how weighty are the arguments marshalled by
Tennant in support of his contention that “the only idea.of a
world-ground that yields an explanation of these facts in their
totality [s¢."*the multitude of interwoven adaptations by which
the world is constituted a theatre of life, intelligence and moral-
ity”] would seem to be that of an efficient, intelligent, ethical
Being.’ But, as he is careful to point out, “when we call this
Being ‘God’ we are borrowing a name, but not any idea, from
religion.’’® Not even any “premature assumption of monotheism
is intended. What is claimed to have as yet been reasonably
established is that cosmic purposing is embodied in the world.
1 Phil. Theol., I1, p. 93: 2 Ibid., II, p. 100. Ibid., II, p. 97.
4 Ibid., IT, p. 100. 5 Tbid., II, p. ror. Ibid., II, p. 103.
7 Thid., II, p. 121. 8 Ibid., II, p. 121.
170 HE WHO IS
But oneness of purpose does not imply numerical oneness of
purposer,”’! and the question of the unity of God is left over for
later consideration.
We must further notice that the argument is not alleged to be
coercive; all through it Tennant is dogged by his doctrine of
‘faith,’ which, it will be remembered, he asserts to be implicated
in all human theorizing. At the end of his investigation he tells
us that ‘‘not only theism, but any metaphysical theory or world-
view whatsoever, can at best claim to be a reasonable belief ulti-
mately grounded on the alogical probability which is the guide of
life and of science, and verifiable only in the sense that it renders
the known explicable. No a priori, rational, logically coercive, or
ese proof is possible.”’? His ultimate claim ismerely that
“if philosophy can show that theistic belief, such as some take to
be the essence of Christianity, is reasonable in the sense that it is
continuous with the faith of science and is on a par, in respect of
its intellectual status, with the probabilities which are involved in
all explanation and in all other knowledge concerning Actuality,
it will have provided an answer to a question which will never fail
to be of vital interest to human beings.”’$ He stsumsass - s judgment
on his argument in these words: i
*‘Our world is . . . a cosmos, at least in the humblest sense of
the word, and the original determinateness of its terms or posita
is such as to make it intelligible. This, of course, constitutes a
teleological proof of theism no more than does the existence of
the world afford a causal or cosmological proof. The mystery of
mysteries is that something exists; and if the one underived or
uncaused existent be God, the creator of all things else, God is
‘the last irrationality,’ and creation is the next to the last expt:
cability. To replace absolute pluralism by theism is to reduce an
indefinite number of separate inexplicabilities to these two alone;
and so far economy, and therefore explicability of a kind, is
secured. It is of no important kind, however: for there is no more
wonder about a self-subsistent plurality than about a self-
subsistent individual. But when the intelligibility of a cosmos,
rather than the mere existence of a world of any sort, is the fact
to be considered, teleological theism evinces more conspicuously
its advantage, in other respects than that of economy, over
absolute pluralism. For over and above the forthcomingness,
conceived as self-subsistence, of the many existents, is their
adaptiveness, inherent in their primary determinateness and their
1 Phil. Theol., 11, p. 121. 2 Tbid., II, p. 249. 3 Tbid., II, p. 250.
THE COSMIC TELEOLOGY OF TENNANT EL
relations, to the requirements of intelligibility. This further
particularizes their determinateness and so bespeaks more of
coincidence in the ‘fortuitous.’ ’’+
The last sentence but one in this passage might, if it stood in
isolation, be taken as a restatement of the Thomist Fifth Way, as
asserting that the very existence of finite being involves a final as
well as an efficient cause.? But the rest of the passage makes it
plain that it is the intelligibility of the cosmos as a whole, or at least
of large sections of it, which is being considered, as is, of course,
shown by the previous statement of the whole argument. The
really surprising feature is the way in which Tennant assumes that
to postulate God and creation, rather than a plurality of finite
beings, as the ultimate categories of explanation of the fact that
something exists, is to effect merely a numerical economy of
concepts ;° this is presumably possible only in view of his doctrine,
on which we shall comment shortly, that God is himself a finite ~
being. Tennant is here in line with his master, James Ward, |
as is seen in their agreement that creation is necessary to God.! \
To describe God as “‘the last irrationality” is inevitable in such a
case; it is parallel to Whitehead’s assertion that ‘‘God is the |
ultimate limitation, and his existence is the ultimate irrationality.”*” “
But the assertion of traditional theism, that God, in contrast to all
if other beings, is self-existent and infinite provides not merely a,
\ numerical economy but an explanation. Any other,God could |
~Only be a last irrationality, but the. Christian God is the oné énitire —
Rationality-through-which alone all else becomes.rational: This,
has already been expounded® and it need not be repeated here.
Tennant’s discussion of cosmic teleology is followed by two
chapters dealing with the Idea of God: Creation, Eternity, Infini-
tude and Perfection, the Absolute, Divine Personality, and Self-
1 Phil. Theol., 11, p. 104. 2 See p. 54 supra.
3 Cf. the following passage in Philosophy of the Sciences :
“We must recognize at the outset that an alogical factor enters into the foundations
of all our knowledge. The posita, of which I have already spoken, are prior to logic
and determinative of all possibility. No reason, let alone a logical or an a priori
reason, can be assigned for their being what they are, or indeed for their being at all.
Their particular determinateness is inexplicable”’ (p. 87).
It is, of course, precisely because no logical or a priori reason can be assigned for the
existence and nature of the posita that traditional theism affirms them to be the
creatures of a self-existent Being. It does not admit that because there is an element
in the world which is alogical in the sense of not being explicable by mere implication
between propositions, it is therefore alogical in the sense of being metaphysically
inexplicable. 4 See p. 113 supra.
5 Science and the Modern World, p. 249 (208). 6 See ch. vi supra.
172 HE WHO IS
limitation. His demolition of all such absolutisms as make the
world merely an appearance or an experience of God would
seem to be unanswerable, as is his recognition that the only
alternative is some doctrine of creation. And he is rightly anxious
to ensure that if terms are used of God they shall have a deter-
minate meaning and shall not be merely vague honorific epithets
savouring, in the phrase which he quotes from Hume, more of
panegyric than of philosophy. But his disregard of the doctrine
of analogy, with the consequence that all terms derived from our
experience of finite beings must be applied to God in precisely
their original sense if they are to have any significance in that
application at all, makes his discussion of the divine attributes,
however illuminating in itself, to some extent irrelevant to the
issue.t It will be sufficient here to see what he has to say about
the attributes of infinity and perfection.
Tennant distinguishes three 3 main uses ofthe word “infinite”’
Christian theology. It’ tan 1mean either “devoid” of all ‘defining
limitations,” “with the “implication of indefiniteness or indeter-
minacy ; it gan mean “the limitless in number, time, or space:
that which cannot be reached by successive acts of addition or
' division”’; or it can have a third meaning, that of “completeness
or perfectness, and Tennant accuses St. Thomas, in common
~ with other Christian theologians, of confusion in his use..of the
om When St: Thomas says that the being of God is called
“infinite”? because it is pure, self-subsistent or not received, he is
using the term simply to imply non-limitation by other self-
subsistents; and to pass on from this notion of infinitude as self-
subsistence to the connotation of ‘‘an omnitudo of all qualities
however mutually incompatible” would be, as Tennant clearly
sees, not only a change in meaning, but a change from sense to
nonsense. There is, however, a sense in which God is an omnitudo,
which follows immediately from his self-subsistence and which does
not involve any contradiction, and this sense is precisely what is
implied in the traditional theistic doctrine. It has been observed
earlier that ipsum esse subsistens and q
quo majus nihil cogitari potest are
mutually equivalent ;* if God is the one self-subsistent,.then-heis
1 It is interesting to note how deeply, through the neglect of the traditional doctrine
of analogy, the notion of creation as necessary to God, has implanted itself in modern
philosophical theology, even when this would claim to be “orthodox.” We have found
iin ae Tennant, and Matthews; it occurs in Pringle-Pattison too (The Idea of God,
ect. Xv1
2 Phil, Theol., I, p. 140. 3 See ch. ii supra.
THE COSMIC TELEOLOGY OF TENNANT 173
necessarily the omnitudo, not of all1 qualities 1 however incompatible,
but ofall perfections. ~St. Dhomealis quite explicit that the infini-
tude of God does not involve the possession of incompatible
qualities. “Whatever implies contradiction,” he says, “‘does not
come within the scope of divine omnipotence, because it cannot
have the nature of possibility. Hence it is better to say that such
things cannot be done, than that God cannot do them.”! Thus
God’s infinitude leads immediately to his perfection ; butit does not
seem fair to accuse St. Thomas, as Tennant does, of “merging”’
the two terms.? Their close connection is inevitable, but that is
simply because God’s infinitude and his perfection are in them-
selves both identical with his essence and differ only in that they
are different aspects of it. Nor is it the case that St. Thomas
confuses the use of “infinite” as meaning self-subsistent with its
use as meaning indefinite or indeterminate, because this latter
notion is quite foreign to his thought about God. To the Angelic
Doctor, Pure Being is not vague or indeterminate,-but.supremely
active and rich;? to him, as Tennant realizes, Pure Beingis the
most perféct of all things, because it actuates allthings, and when
we remember that; to the Catholic theologian, this Pure Being is
none other than the eternal and ever-living Trinity, we see at once
how false is the notion that in Catholic theology the infinitude of
God carries with it the least suggestion of indeterminacy or
indefiniteness. There is no need to deny that words like “‘infinity”’
and “‘perfection”’ have often been applied to God with vague and
overlapping meanings, but, however we precisely define them, dst
Paneer
there are two mutually implicatory facts for which they stand, |
namely, that God is unconditioned by anything outside him- |
self and that he is maxime ens, the plenitude of all that supremely | eee
1s.
In his treatment of creation;~[ennant finds himself in asimilar~~
position to that_ of \ Whitehead, hough he does not force it
to the same extreme-or- “express it in so sharply antithetical a —
form.
“The human imagination,”’ he writes, “‘most forcibly represents
to itself the dependence of the world on the will of God by sup-
posing God and time to precede creation. But, as early Christian
thinkers found, there is no reason to convert this representation
into a theistic tenet. One of them regarded creation as an
1 §, Theol., I, xxv, 3¢. 2 Phil, Theol., Il, p. 141. 3 $.¢.G., I, xliii.
174, HE WHO IS
endless regress, world preceding world; another taught that
creation was not in time, but the world and time were
created together.... A further refinement is made when God
is conceived as essentially the world-ground or creator ;not another
cause in the series, or a being who might or might not have
created. God gud God is creator, and the creator gud creator is
God: or ‘God without the world is not God.’ ’’?
And this last 1
is Tennant’s own view.
ae ee
‘Thus conceived,”’ he continues,
s, “the theistic
idea of creation
is free from the old puzzles concerning temporal relations. If God
be a-world-ground, “there never could have been no world. The
‘possibilities,’ prior to any Actuality, between which the Deity
has been thought to have chosen, are but hypostatized abstrac-
tions.... If we are to speak in terms of time, the theistic
doctrine may be summed up in the statement thatthe world is
coeval with God and is contingent. on his determinate _nature
sacaceai
inclusive ofwill.
SSO
w oo
WE
_ There are various suggestions scattered through his work which
make one hesitate to ascribe to Tennant the obvious force of his
statements in the pages from which the above extracts are taken.
Like Ward, he is clear that the world derives its existence from ,
God: “The world ++ vis God’s ‘utterance.
“Its nature depends on
his nature and* will, not his naturé”on that of a self-subsistent
universe.”’®. Such a statement is far more satisfactory than
Whitehead’s dictum that “‘it is as true to say that God creates the
World, as that the World creates God.’4 Yet Tennant tells us
that ‘‘God without a world, or a Real other, is not God but an
abstraction,’’> in much the same way as the primordial nature of
Whitehead’s God is not “eminently real’ but “deficiently
actual.’ And it seems clearly fallacious to argue that “if God
be a world-ground, there never could have been no world,”’ for
it is perfectly possible to suppose that God is, by his very nature,
a world-ground in the sense of being able to create a world, but is
only incidentally a world-ground in the sense of having actually
done so. From the contemplation of a picture we can say that
there must have been an artist who was able to paint it; but we
cannot say that he could not help painting it, or that “‘there
1 Phil. Theol., 11, p. 128. Tennant repeats this last assertion in three other places,
viz., pp. 156, 168 and 183.
2 Ibid., II, p. 129. 3 Thid., II, p. 209. 4 Process and Reality, p. 492.
5 Phil. Theol., Il, p. 168. 8 Process and Reality, p. 486.
THE COSMICG TELEOLOGY OF TENNANT 175
never could have been no picture.” Tennant is, in fact, making a
9
needless capitulation to Broad’s belief that the theistic position
implies that the existence of the world follows from the existence
of God, though, unlike Broad, he sees the necessity as meta-
physical rather than logical.1 The fact is that, having rejected,
as traditional theism also rejects, the idea that creation necessarily
implies a temporal beginning, Tennant concludes that the world
is thereby shown to be necessary to God. All that has been said
in chapter viii above is relevant here, and we may sum it up in
Gilson’s words that ‘‘itis quite ris an eminently
- Christian-God; but a God whose very essence isto be a creator is
net-a Christian God at all. The essence of the true Christian God
is not _to create but.to.be.”?-Like Sct etl ‘Tennant assumes
that, because-we-can-say that the world must have been Gicateds
by eg we Can infer that God ima have created the world,
4 ignoring the fact that the antecedent conditions presupposed a
\ these two statements are by no means the same. We agree that, Seer
oe
s the world exists, the world must have been created by God, but
\ we cannot conclude that, if God exists, God must create the world,’
\Tndend, the true conclusion is precisely the opposite. Because it is
the very insufficiency of the world that implies the existence of a
Creator, it follows that this Creator must be himself self-sufficient
and therefore under no obligation to create. The world demands
as its ground a God who need not have made it. We may sum
this up by saying that we have indeed to make our first approach
to God by examining the finite world, but that this very examina-
tion, while it shows us that God has necessarily the power to create
and has as a matter offact made use of this power, shows us also
that his use of it is not necessary but is an act of free and uncon-
strained will. This point has been sufficiently elaborated already
and needs no more mention here.
Because of his implicit denial of analogia entis and because of the
explicit denial of divine infinity which results from it, Tennant is
forced to envisage God as related to the world in essentially the
same way as finite beings are related to one another. He cannot
say, with the scholastic tradition, that, whether God creates this
world, a different world, or no world at all, his own infinite perfec-
tion is unaffected, or that God and the world do not “add up.”
In consequence he finds himself, in order to save the divine wisdom
and goodness, under the necessity of maintaining that, taking into
1 See p. 57 f. supra. 2 God and Philosophy, p. 88, quoted on p, 97 supra.
176 HE WHO IS
account the conditions under which God has to work, this isthe_
best of all possible worlds ;! this gives a rather unrealistic flavour
to his discussion of the problem of evil. Again, since terms have to
be applied univocally to God and to creatures, he-has to set limits
to God’s power and knowledge ;God has limited his own activity
in creating a world in which there are beings exerting causality,
and his conservation of the world is exercised rather in preventing
these spontaneously acting independent beings from disrupting
the cosmos than in preserving them in existence by a continuous
act of creative power. God’s relation to us thus becomes far less-
intimate than traditional theology has taught. ‘‘Theism,’’ writes
Tennant, “does attribute to God’s creatures delegated spontaneity
such as conceivably might develope into erratic tendencies. And
it is only with the qualification which this attribution involves that
the theist can use such expressions as that in God ‘we live and
move and have our being’: to take these words more literally
would be to identify the world-ground with the world instead of
to insist on the inseparableness and distinctness of God and his
world.”? There thus appears an interesting paradox. Because,
through his doctrine that creation is necessary, he has lessened the
ontological distinction between God and the World, Tennanthas
perforce to exaggerate their practical independence in order to
avoid confusing them altogether; while traditional theology,
secure in its fundamental postulate of God’s transcendence as self-
existent being, can without fear of any concession to pantheism or
determinism, teach that, without any violation of human freedom,
God is the primary causeof every human act.
It is unnecessary to follow up these points in further detail ; their
bearing is obvious. But one further point must be briefly examined.
This_is the question as to whether God_is-a_unity ora plurality.
Tennant is not very much impressed by the argument that “‘if
we are to regard goodness and love as essential attributes of
God ... this implies that within or besides himself there must
eternally have been an other to be known and loved with all the
fulness of his capacity to know and to love,’’? apparently because
of his view that God always had a world to occupy himself with.
“God without a world, or a Real other, is not God, but an
1 This is altogether different from the Thomist doctrine that, while God might have
made a better world, he could not have made the present world in a better way.
Cf. S. Theol., I, xxv, 5, et 6 ad 1; Garrigou-Lagrange, Dieu, E.T., II, p. 345.
2 Phil. Theol., II, p. 212. 3 Ibid., II, p. 169.
THE COSMIG TELEOLOGY OF TENNANT 177
abstraction.”! But, as we have seen, he does not claim that his
argument necessarily leads to monotheism. “Oneness of purpose
does not imply numerical oneness of purposer.”” And while he
rejects the doctrine of the Trinity in its orthodox form on the
ground that it postulates for the three Persons a status inter-
mediate between the substantival and the adjectival,? he sees no
objection to the idea of “God”? as a self-subsistent society of divine
beings with equality of nature, identity of essence and harmony of
will. And his subsequent inquiries do not, so far as one can see,
throw any further light on the question: “‘the issue,” we are told,
“‘does not lie within the sphere of fact-controlled thought.”’* The
plain fact is that he is not able, and apparently does not wish, to
rule out the possibility of polytheism, though it is a highly refined
polytheism which he prefers to describe as “‘the pluralistic con-
ception of deity.”> That he is prepared to face this is a great
tribute to his courage and honesty; there are few philosophers
who would be willing to risk the charge of polytheism, with the
uncomplimentary associations which it carries, even if they hoped
to be able to show that those associations were gciliysoms to
their particular case.
We may bring this discussion of Dr. Tennant’s Cosmic Teleo-
logy to an end by remarking that, if, as we have seen, it affirms the
existence of a God who is less than the God of traditional theism in
that, while the universe is dependent upon him for its existence,
nevertheless he is bound to create it and once he has created it is
limited by it, this is only what we might have expected from the
method of investigation adopted. For, if we restrict ourselves to
asking questions about the way in which the beings which compose
the world behave—and this is what Tennant’s investigation of
cosmic purposing amounts to—we are most unlikely to arrive at
the notion of a genuinely transcendent cause. As Whitehead,
who is far more consistent than Tennant in following this pro-
cedure, says, ““Any proof which commences with the character of
the actual world cannot rise above the actuality of this world... .
1 The argument is that, while “‘a self or subject does not depend, qué an existent or
as to its being, upon any object to which it stands in the relation of presentation or
attention,” yet “relation or rapport is essential for any existent’s assuming the réle of,
and being, a subject.” (Phil. Theol., II, p. 167.) This is, of course, precisely the argu-
ment which, in conjunction with the notion of znternal relations, a trinitarian will use
in support of the doctrine of the Trinity. See p. 186 infra.
2 Phil, Theol., 11, Appendix: Note D, p. 267 f. 3 Tbid., II, p. 170 f.
4 Tbid., JI, p. 172% 5. Thid. Tle psa7zo:
178 HE WHO Is
_””It may discover an immanent God, but not a God wholly trans-
(__cendent.’! And the transcendence which in places Tennant
verbally ascribes to God is far from being transcendence in the
traditional and thoroughgoing sense. It is only when, dis-
regarding the prohibitions of a philosophical method which
mistakenly bases itself upon the procedures proper to the natural
sciences, we ask the metaphysical question, “Why do finite beings
exist at all?’’ that we can arrive at the recognition of the true
transcendence of the Deity. And this is a valid question, which
needs asking. Time after time Tennant comes near to asking it,
but he always fails to do so. In addition, the unfortunate intro-
duction of “‘faith”’ at the end of his first volume renders the whole
of the subsequent argument professedly tentative and speculative.
As a result he arrives at the merely probable existence of a God
who is less than the God of religion: a God who, whatever else he
may be, is certainly not adorable. If Tennant’s contentions were
at all points irrefutable, we should have to accept them and make
the best of a bad job. But reasons have been advanced for sup-
posing that this is not so. The plain fact is that, in his magnificent
refutation of all attempts to build up theology on purely a prion
arguments, Tennant never seems to envisage any other alternative
than a God who is essentially implicated in the world-process, and
therefore never takes seriously the traditional doctrine of a God
who is discovered as the creator of the world and yet is at the same
time altogether self-sufficient.2, But when we have said this, we
are far from denying the importance that attaches to the systematic
working out in such careful detail of the implications of a radically
empirical approach to the problem of the design in the universe.
| 1 Religion in the Making, p. 71. (Italics not in original: I suspect that Whitehead
_mentally stressed ‘“‘actual”’ rather than “‘character” and so missed the real point of
his own remark!) This-inherent limitation of any approach which is primarily teleo-
logical rather than cosmological appears in Dr. W. R. Matthews’ development of the
‘teleological argument in his Purpose of God. His final conclusion is that ‘‘the doctrine
of the self-sufficiency of God should be rejected” (p. 173).
2 In an interesting article in the Harvard Theological Review, July 1942, Dr. Graham
Frisbee examines Tennant’s attempt “‘to show that the charge that the empiricist is
unable to ‘prove’ the existence of more than a finite God is damaging on neither its
philosophical nor its religious side,” and concludes that “‘ whatever else he has or has
not done, he has not shown that.”
THE MORAL ARGUMENT 179
ADDITIONAL Note C To CHAPTER xu
THE MORAL ARGUMENT AND THE PROBLEM OF EVIL
This book would hardly be complete without some reference to
an approach to theism which has become increasingly prominent
during the last century, namely that which adopts as its starting-
point the fact and the content of the moral consciousness of man.
Historically this development is, of course, to be traced to the
influence of Kant, who, having, as he thought, demolished all
forms of speculative argumentation for the existence of God as
involving, in one way or another, the fallacy of the ontological
argument, took refuge in the moral sense of man as affirming the
existence of a Supreme Moral Being as a necessity for the practical
reason. Anything like a full discussion of the moral approach
would fall outside our present scope, for, except in so far as it
enters as one element into the‘‘Argument ee Perfection”? which | <—
is the Thomist Fourth Way, it™is almost entirely absent from
‘traditional theology. Furthermore, an adequate discussion of it
‘would require a book of its own, and not merely a few pages in a
book whose primary concern is with something else. What is,
however, both relevant and necessary to our present task is to give
some consideration to the relation between the moral approach
and that of traditional theology; that, and that only, will be
attempted here.
It has been argued at some length in the preceding chapters
that the fundamental question for theology is, “‘ Why do any finite
beings exist at all?’’ and not merely, ‘‘ Why has the realm of finite
beings the particular characteristics which we observe in it?’’,
though, as we have recognized, it is in practice impossible to ask
the one question without at least thinking about the other. In |
addition we have tried to show that the common modern notions
of a finite God or of a God to whom the creation of the world is
necessary arise precisely from concentrating on the second question
rather than the first; that is one of the main points of our dis-
cussion of the works of Whitehead and Tennant, as well as of the
writings of several other philosophers who have been referred to
in less detail. It may, however, be added, in qualification of this
statement, that if there 7s any question about the nature, rather
N
180 HE WHO IS
than about the existence,of finite beings which might offer some
hope of arriving at the existence of a God who is, in the fullest
sense, independent of his creation, it will presumably be some
question about the nature of the human mind; for, among all the
beings of which we have experience in this world of ours, the
human mind is, so far as we know, the only one which has the
capacity of transcending the limits of the finite order and of
directing its operation to affairs which are not altogether bound
up with finite being.1 It was asserted above, in our discussion of
mysticism, that, while the natural object of the finite human
intellect is finite being, its only adequate and completely satisfying
object is the infinite Being which is God. In a similar, but not
quite identical, way we can assert that, while the immediate object
of the human will, in its fallen state, is finite good, its only
adequate and completely satisfying object is the supreme Good
which is God.? We cannot, however, argue with any safety from
this fact of unsatisfiedness alone to the existence of a perfect Good
in which man’s moral needs are to be satisfied. If we already have
reasons to believe in the existence of God, we can see why man is
unsatisfied with finite good and can also see that God is the one
Good in which man can find full satisfaction. But until we have
those reasons such an inference is at best extremely precarious.
_ This reservation is very similar to the objection which Tennant
_ brings against Kant when he asserts that “‘if the summum bonum has
its possibility of realization guaranteed by the concept itself, Kant
in principle employs the ontological argument in ethics after
demolishing it in theology,”’ since Kant “merely argues that God,
a regulative idea for the theoretical reason, is a postulate for the
practical reason: zf the moral order of the world (as Kant con-
ceived it) is to stand, and what ought to be is to be, God must
1 The possibility of the existence of pure spirits or of rational psychosomatic beings
other than man need not be taken into account here, as in any case the content of
their moral experience is not accessible to our examination.
2 “Similar, but not quite identical,” because, whereas the intellect bears imme-
diately upon finite being because of its own finitude, the will of fallen man bears
immediately upon finite goods not because of its finitude but because of its fallenness.
The essential difference is that intellect is a faculty of possession, while will is a faculty
of tendency; we can know only what is present to our mind, whereas we can desire
what is absent. Even the intellect of the saint cannot have God-in-his-essence as its
object until it has been altogether “deified” by God himself in the Beatific Vision (or
in some transient foretaste of it), while even imperfect Christians, if they are in a state
of grace, can have God-in-his-essence as the object of their will. (This does not, of
course, mean that the degree in which the will loves its object is necessarily perfect.)
With this explanation the statement in the text may stand.
THE MORAL ARGUMENT 181
exist.” And it is of importance to notice that most philo-
sophers who have elaborated the moral argument in recent years
have refused to take man’s moral sense in isolation as the basis of
their treatment. Thus Dr. W. R. Sorley wrote that “the argu-
ment .. . may be looked upon as a special and striking extension
f the cosmological argument,”? while Tennant affirms that
oral considerations ‘“‘supply the coping-stone of a cumulative .
teleological argument.”? Again, Dr. William Temple, in his
massive work Nature, Man and God, while he adopts the moral
alue or the Good as his fundamental category, builds up his case
r theism not upon the mere occurrence of man’s intuition of the
Good, but upon the fact that the world is of such a character as
to give rise to minds in which, among other properties, this
intuition is found. And, while Professor A. E. Taylor, in the first
volume of his Faith of a Moralist, seems to go further than any of
the writers just. mentioned towards finding a foundation for
theistic belief in the moral consciousness of man alone, it is signi-
ficant that when, in his essay on ‘“‘The Vindication of Religion,”
he sets out to give a complete presentation of the case for theism, ~
the moral argument forms only one strand in a threefold cord, of~
——
which the other two strands are formed by the cosmological and
teleological arguments and by the argument from religious
experience respectively. Nor must we forget the vigorous defence
of the cosmological approach in his long article on “Theism”’ in
the Encyclopedia of Religion and Ethics. It should also be noticed
that, in developing the moral argument, he is careful to insist that
what he is dealing with is not the bare concept of morality, but
the moral life as a concrete constituent of human experience.
It appears from such instances as the above that, while the moral
approach is to be accepted as the great constructive addition
which modern philosophy has made to the edifice of traditional
theology, it depends for its full force upon the prior, or at least the
simultaneous, acceptance of a more metaphysical investigation.
We must add that the consequences of a religious kind which the
moral argument will involve will depend very largely upon the
form that this metaphysical investigation has taken. If the moral
argument is taken in conjunction with a cosmological doctrine
1 Phil. Theol., 11, pp. 97, 96.
2 Moral Values and the Idea of God, p. 348. 3 Phil. Theol., II, p. 100.
4 This is the second essay in Essays Catholic and Critical. It is only right to add that
Taylor does not claim demonstraiive certainty for his arguments, and that he intro-
duces “‘faith” in much the same way as Tennant does (op. cit., p. 32).
182 HE WHO IS
that asserts as the ground of the cosmos a “‘God”’ who is finite or
to whom the cosmos is necessary for the full expression of his
activity, the response which the moral consciousness demands
being directed towards the object thus postulated, religion will
tend to lay its main emphasis upon fellowship and co-operation
with the Deity; this trend is strongly marked in Whitehead’s
system and less strongly in Tennant’s and Ward’s. If God is
conceived as purely immanent, man’s effort will be directed
inwards in a quietistic self-stultification, as in the religions of
India. But if God is conceived as strictly infinite and the world as
adding nothing to his inherent perfection, the dominant note will
be one of adoration and self-abandonment, and of supplication
for a union with God which only God can confer; this is the
position of historical Christianity.
A few words must be added about the problem of evil, if only
because it provides the objection which the modern man or woman
nearly always falls back upon as his or her conclusive reason for
rejecting the theistic position, however convincingly this may
seem to have been stated. It is commonly alleged that academic
theologians are remarkably insensitive to its gravity; with the
implication that this is due to the seclusion of their lives from the
hard realities of the contemporary world. To some extent this
accusation may be admitted; the system of McTaggart was
woefully deficient in any real understanding of the problem,
Tennant deals with it better, though still unsatisfactorily,? and
Taylor better still. But one of the most profound discussions of it
in recent years is due to one who is not a professional philosopher
or theologian at all, namely, Mr. C. S. Lewis, in his short book,
The Problem of Pain. His discussion is not merely philosophical, but
theological ;it is based upon a rigidly orthodox, but by no means
cut-and-dried Christian position. It is indeed doubtful whether
one can get very far towards a solution by purely philosophical
considerations ; the theist will, in the last resort, go on repeating,
“T have proved that God exists and is both good and almighty,
therefore the problem of evil must be somehow soluble, although
I can’t solve it,” while the objector will as often reply, ‘‘ But the
problem is obviously insoluble, therefore there must be some flaw
in your arguments about God, although I can’t see where it is.”’ o
1 On the other hand, it is right to remark that very few ordinary men and women
have so gloomy an outlook upon life as the radically pessimistic philosopher.
2 Phil. Theol., 11, ch. vii. 3 Faith ofaMoralist, I, ch. v.
THE MORAL ARGUMENT 183
Only one or two remarks which seem relevant will be offered
here.
In the firstplace, the problem is one which religion creates for
itself, not one which it finds awaiting a solution. If there is no
God, then there is no problem of reconciling the existence of pain
and xa with his love and power; and, while the atheist may with
reason urge against theism that it has set itself a problem which it
cannot solve, he has no business to feel evil as constituting a prob-
lem for him, except in the purely intellectual sense of causing him
to wonder where it came from. In actual fact, however, very
many atheists appear to labour under a gigantic sense of resent-
ment for the existence of evil, although, on their own hypothesis,
there is no one for them to be resentful against; this note is, for
example, very marked in Mr. Bertrand Russell’s famous essay on
“A Free Man’s Worship.”’! This is very mysterious, and almost
leads one to suspect that the atheists have been indulging in a
little surreptitious theism on the quiet.
Secondly, evil, at any rate in the form of pain (which is the aoa
form in which it troubles most people), is only a problem because |
people rebel against it. No one, so far as is known, has ever lost |
his faith as a result of the muscular and dermatic pain involved in
rowing for Cambridge in the Boat Race, though, if the mere
existence of pain as such is incompatible with theism, it should
provide the material for as strong an argument against religion
as is furnished by all the suffering involved in a world-war. The
fact is that suffering is rebelled against only if no end can be seen
which appears to be proportionate to it, In the case mentioned,
the suffering is believed to be amply compensated for by the hope
of arriving at Mortlake Brewery before the Oxford Boat. This
suggests that the problem of pain presents itself to us largely
because of the limitations of our knowledge and imagination. If
we could see as God sees, it would, for all we know, be trans-
parently obvious that the sufferings of the present time are not
worthy to be compared with the glory that shall be revealed in us,
and indeed that the sufferings were instrumental to the glory.
This is perhaps as far as one can get towards a solution from
philosophical considerations; they are almost entirely negative in
their scope. But the Christian religion, with its teaching of union
with Christ and of reparation for sin, provides a positive tech-
nique for dealing with evil, both moral and physical. “Consider
1 Mysticism and Logic, ch. iii.
184 HE WHO IS
the fact of suffering,” writes Mr. F. J. Sheed, “Christianity
answers... that suffering would be altogether intolerable if |
there were no God, but can be turned to the highest uses of man
ifthere be a God. Atheism answers that the fact of suffering proves
‘that there is no God. But this does not reduce the world’s suffer-
ings by one hair-breadth, it only takes away hope.”! “The
Christian idea of God,” writes the Abbé Nédoncelle, “has lighted
up the ugliness of evil, and has shown itself to be the best and most
practical weapon with which to fight suffering and sin. Christian-
__ity has done something better than expound the nature of evil:
_it has provided mankind with a ferment which dissolves it. It has
produced the beatitude of the Saints and the radiance of their
healing influence. Christ has shown us, once and for all, how
suffering should be faced, and we have only to imitate him. He
suffered in achieving the impossible: in turning suffering—some-
thing which is hateful and bad—into action, into the most fruitful
of all actions, into one which clothes us in humility, and establishes
us in an utter disinterestedness.”’?
ADDITIONAL Note D To CHAPTER XII
NATURAL THEOLOGY AND THE DOCTRINE OF THE TRINITY
As we have seen, traditional theism asserts that the considera-
tion of finite beings, while it can assure us of the existence of God,
is powerless to give us any quidditative knowledge of his interior
essence ; it follows from this that the Doctrine of the Trinity falls
outside the proper sphere of natural theology. This conclusion is
in no way contradicted by the fact that St. Augustine has seen an
analogue of the three divine Persons in the natural structure of the
human soul,? or by the universal analogism of St. Bonaventure,
in virtue of which the latter saint can write that ‘‘a creature of the
world is, as it were, a book in which there shines forth, is depicted
and can be read the Trinity that made it” ;* for, as the very
1 Communism and Man, p. 187.
2 Baron Friedrich von Hiigel, p. 120. 3 De Trin., ix, x, xiv.
*“Creatura mundi est quasi quidam liber in quo relucet, repraesentatur et legitur Trinitas
oe (Breviloquium, 11, 12, 1; quoted by Gilson, Phil. of St. Bonaventure, E.T.,
p. 214.
DOCTRINE OF THE TRINITY 185, -
reluctance of these two great Christian thinkers to separate natural
from revealed theology would suggest, this discernment of a
trinitarian structure in creation is not the product of natural
theology at all, but arises from a deliberate reflection upon the
created world from the standpoint of the Christian revelation.
This ex post facto character is even more clear in such a case as that
_ of Hegel, with his extraordinary identification of the distinctions
_~of his secondary triad with the Persons of the Trinity.t And it
seems to be generally agreed that the parallels to the Christian
Trinity which can be found in certain non-Christian religions are
very remote indeed.? We need not, however, deny that, once the
doctrine has been accepted, we may then be able to see how the
three Persons each participate in their several modes in the work
of creation. Thus, St. Thomas himself, after vigorously asserting
that “‘all things caused are the common work of the whole God-
head”’ and hence that “‘to create is not proper to any one Person,
but is common to the whole Trinity,” immediately adds that, just
as “the craftsman works through the word conceived in his mind,
and through the love of his will regarding some object,”’ so also
“God the Father made the creature through his Word, which is
his Son, and through his Love, which is the Holy Ghost.”’? Indeed,
he goes on to affirm with St. Augustine’ that in creatures there is
necessarily found a trace (vestigium) of the Trinity. ‘‘In rational
creatures,” he tells us, “‘there is found the representation of the
| Trinity by way of zmage,” since the processions of the Son and the
| Spirit are analogous to the rational operations of intellect and
. will, but ‘‘in all creatures there is found the trace of the Trinity,”
since as substances they represent the cause and principle which
is the Father, as possessing form and species they represent the
Son, and as having relation of order they represent the Holy
Ghost.* And he refers back to an earlier article in which, after
denying (since the creative power of God belongs to the unity of
the essence and not to the distinction of the Persons) that God can
be known as Trinity from the consideration of created things, he
has nevertheless admitted that reason can confirm that God is
Trinity, when this has already been accepted on other grounds.®
And he admits that some of the Greek philosophers had an obscure
awareness of some kind of plurality in the Godhead, though he
1 Cf. McTaggart, Studies in Hegelian Cosmology, ch. viii.
2 See, e.g., Dr. Berriedale Keith in E.R.E., XII, s.v. “Trimirti.”
3. Theol, I, xlv, 6, sed contra et resp. 4 De Trin, vi, 10.
5 §. Theol., I, xlv, 7c. 6 §. Theol., I, xxii, 2,¢ et ad 2.
186 HE WHO IS
adds that it was partial and inaccurate. We may not unreasonably
suspect that, had he known about it, he would have made a similar
admission about the Hindu Trimarii.
From a similar ex post facto standpoint it can be seen that the
doctrine of the Trinity provides the great bulwark against the
tendency, which we have seen to be so prominent in modern
theistic philosophy, to view the world as necessary to God in order
that God shall have an arena for his self-expression and an object
for his love. In theory, of course, we can assert simply that in the
contemplation of his own perfection and the love of his own good-
ness God has an entirely adequate sphere for the exercise of his
supremely full and vital activity. But, in practice, unless we believe
in the eternal generation of the divine Word and the eternal
spiration of the Spirit of love, we shall probably find ourselves
saying with Dr. Tennant that “God without a world, or a Real
other, is not God, but an abstraction.’’! We shall then argue with
him that, while ‘“‘a self or subject does not depend, gud an existent
or as to its being, upon any object to which it stands in the relation
of presentation and attention,” yet “relation or rapport is essential
for any existent’s assuming the réle of, and being, a subject,’”?
without realizing that this is exactly the kind of argument which,
in conjunction with the notion of internal relations, a trinitarian
will use in support of the doctrine of the Trinity. We may add
that Tennant’s criticism of this cardinal doctrine of the Christian
Faith does not seem very convincing. According to him, it claims
for the three persons a status between the substantival and the
adjectival which is inconceivable.* But here he has surely missed
the point. The three Persons are distinct and substantival (not,
be it observed, three substances, but three substantives, three hypo-
stases), but this does not involve that they are three Gods, simply
because the processions within the being of God are necessary. Of
course, if the creation of the world was necessary too, this would
place the Second and Third Persons on the same level as the world
itself, and we should be involved in something rather akin to
Arianism. But the traditional view of creation stands or falls with
the contingency of the world. It is because the creation of the
world is contingent, whereas the processions of the Persons are
necessary, that we can at the same time assert that God without the
world is still God, and that the three Persons are not three Gods.
1 Phil. Theol., Il,p. 168. @Vbid.; Il, p. 167. 3 Cf, p. 177, n. 1 supra.
4 Phil. Theol., IL, Appendix: Note D, p. 267 f.
DOCTRINE OF THE TRINITY 187
In co clusion, some very tentative suggestions might be made
here as to the way in which the “trace”’ of the Trinity is to be
observed in the created world.
There are in the world two great contrasts which force them-
selves upon our attention. The first is the contrast between the
actual and the possible; the second the contrast between. per-
manence and change.
(a) The contrast between the actual and the possible has come
in recent years very much to the front in modern philosophy,
although the distinction between a realm of experience and a
realm of ideas is at least as old as Plato. The materialistic-
scientific philosophy of the last century tried to obliterate the
distinction completely. It was claimed that a sufficient account
could be given of the universe by supposing it to consist of an
aggregate of material particles moving under the direction of a
few simply defined mathematical laws. As a matter of fact, even
if this claim were true, it would still be necessary to account for
the initial configuration of the particles at some arbitrarily chosen
zero of time, and to explain why the laws governing their motions
should be those that were actually observed to hold rather than
any of the multiplicity of other laws which could, without any
violation of logic, be equally easily imagined. But Victorian
science, under the baneful shadow of Herbert Spencer, was con-
tent to ignore these awkward questions and to take the world of
experience as a self-obvious axiomatic fact. In recent years,
however, the breakdown of the materialistic theory has forced
scientific philosophers to recognize the essential contingency of the
universe. Different scientists at the present day give us different
models of the ultimate constitution of the physical world; but
whether they correlate phenomena by the concepts of g-numbers,
probability-waves, matrices or what_not,-it-is-clear that the uni-
verse which we experience possesses no logical necessity. (We have
commented on the unconvincingness of Eddington’s attempt to
show that the actual world is inevitable.1) The fundamental thesis
of Whitehead’s cosmology is, as we have seen, the distinction
between-the-realm of possibility and that of actuality, between
‘eternal objects”? and the ‘“‘actual occasions’”’ into which these
are ingredient ;and the ultimate problem, which he tries to over-
come by the two notions of actual entities as units of creativity and
of God as the ideal locus of all eternal objects, is to explain why
1 See p. 51 supra.
188° HE WHO Is
© the eternal objects are ingredient into the actual occasions in the
{particular modes in which they are ingredient rather than in
— “others which are equally possible. —
—(b) The second contrast is that between permanence and
nikechange, and it is well to note that these two notions of permanence
‘and change, though they are often felt to be contradictory, are
both necessarily involved if there is to be any experience of a
world at all. I can only say ‘‘A is permanent”? if there is some-
thing by whose change I measure the passage of time. I can only
say ““B has changed” if, in some sense, B has remained the same;
otherwise instead of B having changed, we shall have two totally
different entities, a B, which is annihilated and a B, which
appears in its place. silce the first one, this contrast has come to
be recognized as fundamental in modern philosophy. —
out of account the wilder forms of “‘philosophy of change,” }
need again only think of Whitehead’s self-creative actual occa-
sions. Incidentally, this same contrast lies behind theAristotelian,
Az
and Thomist doctrine of potentiality and act. wn?
In consequence, therefore, of these two contrasts we are
_ presented with three fundamental and irreducible, yet mutually
implicated, facts of experience which can be summarized as
—
(i) There is an infinite realm of abstract possibility.
(ii) There is, selected from this realm, a limited realm of
actual concrete occurrence.
(iii) In contrast with the former realm, there is in this latter
realm_a process of change.
The suggestion which is here made is that these three facts are to
be correlated (in Thomist language, ‘‘appropriated’’) to the three
Persons of the Trinity, as, in virtue of their mutual coinherence,
they all concur in the unity of the creative act.
God the-Father, the great Catholic Creed asserts, is the Maker
(Factor, Tloinths) of all things visible and invisible, while
through God the-Son all things “came into being”?((facta sunt,
éyéveto). ‘This is the obvious force of ééyéveTo, and it is perhaps
not without point to remark that facta sunt is at least as much the
active of fio as it is the passive of facio. Thus the Creed affirms
that, while in the last resort everything derives from the Father,
who is the ground of all possibility, the concrete existence of the
actual world is the outcome of the peculiar mode of action of the
DOCTRINE OF THE TRINITY 189,
on. (We are tempted to identify Whitehead’s principle of “‘con-
cretion” or “‘decision’’.with the Son, but in view of Whitehead’s
radically unorthodox doctrine of actual entities as self-creative,
e could do so only with many reservations.) If we remember™
that the Son is also the Father’s Word, the complete expression
and inner reduplication of himself, the Godfrom.God who is the
Father’s Image, this will not seem fantastic. In the perfect intel-
lectual act which is the eternal generation of the Word, the whole
realm of possibility which is rooted in the divine fecundity will
receive its ideal expression; as St. Thomas tells us, the ideas of all
created things are, in a certain way, in God.t They are, we
might say, uttered by the Father to himself in the generation of
the Son as his Image. But the selection and expression of ideas
in the created realm will then fall to the Son “by whom all things
come into being.” Thus, as St. Paul writes, ‘‘‘To us there is one
God the Father, of whom are all things (€€ oW ta tTdvta)...
and one Lord, Jesus Christ, by whom are all things (81 o0 T&
Tavta) ...”? The subtle distinction between & and 51& gives
the exact shade of meaning required to bring out the difference
between the réles played by the Father and the Son in the ~
creative act.®
Can we now associate the third fact, that of change, with the
Holy-Spirit? Can we ‘‘appropriate”’ to him the “‘ becomingness’”’
of the universe? The Nicene Creed tells us that the Spirit is “the
Lord, the Giver of Life’? (Dominum et vivificantem, tO Kupiov to
zworroidv), and the essenceof life inthematerialrealmis movement,
adaptability and change. It is surely not fanciful to see the
peculiar action of the Holy Spirit, no less than that of the Creative
Word, in every event of the world’s history. There is a variant
reading of a text in the first chapter of St. John’s Gospel, which
tells us that that which was made was life in the divine Word,
and if we can take this in conjunction with the truth that the Spirit
is the Giver of Life, we can then recognize that at the heart of the
world’s process is the peculiar work of the Holy Ghost. St. Thomas
tells us that “seeing that the Holy Ghost proceeds by way of love,
and that love is an impelling and moving force, any movement
LS. Theol., V5 xv. S.¢:G. 5.1, li: 2 1 Cor. viii, 6.
8 It may be remarked that Bulgakov sees the divine Essence rather than the Logos
as the locus of the ideas to be realized in creation, apparently on the““ Bonaventuran”
ground that creation reflects not just the F ather but the tripersonal Godhead. See
The Wisdom of God, ch. ii. Cf. also p. 140, n. 1 supra. The two views do not appear
irreconcilable.
190 HE WHO IS
that God causes in things is rightly appropriated to the Holy
Ghost,’’! and he interprets this, not only of the movement of living
creatures, but also, on the strength of Genesis i, 2, of the initial
origination of created being.
While emphasizing again the speculative nature of this exposi-
tion, we may sum it up as follows. Creation is not, as the deists
conceived it, an isolated occurrence which happened once for all
in the remote past; it is the act of love and power, continuous from
our standpoint but timeless from God’s, by which the triune Deity
sustains the world in being. In this creative act, all the three
Persons take part; and, just as they are not three Gods but three
Persons united in one indivisible Godhead, so there are no: three
acts but three elements indivisibly combined in one act. At the
root of all created being there is the infinite potentiality_of all
possible existence, grounded in God the Father and imaged in the
eternal generation of the Word; superimposed. upon this there is
the contribution of the Divine Word, the principle of decision and
concretion, resolving pure potentiality into the actuality of
occurrence; while the full character of the being as embodied in
the world-process is completed by the inbreathing of the Holy
Spirit, who confers upon it the character of temporality and
becomingness which makes it an actual member of the finite,
temporal world.
Thus, it may be suggested, the doctrine of the Trinity may be
seen to be directly relevant to cosmology, although it is not itself
to be deduced from the consideration of the created realm.
1 $.c.G., IV, xx.
ar’ CHAPTER ZI
CONCLUSION
OMPARATIVELY little need be added in bringing this
book to its end. It may, however, be well to summarize
briefly the course that the argument has followed.
_ We began, m the Introduction, by distinguishing between the
| ordo exsendi, the order in which things ultimately exist, and the wa
ordo cognoscondi, the order in which we come to know them, and
remarked that from the former, though not from the latter, stand-
“point the doctrine ofGod is the primary doctrine ofthe Christian
Religion. Observing in passing the universal belief of mankind
in 4 Being who, however dimly he may be apprehended, is the
Being whom the theist identifies as God, we then saw how Christian
theism, with its two main sources of Judzo-Christian and Graeco-
Roman origin, respectively, recdaved a coherent, and for Western
Christendom, a normative formulation in the transformation
which Aristotelianism
underwent in the hands of St. Thomas
In the Second Chapter, as a preliminary to investigating the
question of the existence of God, we endeavoured to elucidate the
precise meaning of he word “God” in Christian usage. Definition
\ by genus and difference was seen to be imposible, for the reason ~
that God does not bdlong to any genus, We then examined the
Ansdtiian definition of God as aliquid quo majus nihil cogitori potest,
with its implication of infinity as the formal constituent of the ae
nature of God, and noted that, in contrast, the Thomist conception
of God as ipsum esse subsistens, while practically equivalent to it,
gives this place to the notion of being.
The Third’ Cc began by asking the question: Why do we
believe that exists?, and it was seen that in all probability no
two persons would answer this question in the same way. Reject-
ing the view that there is a natural human faculty by which we ~
could apprehend God with the same directness as we apprehend
material objects, we went on to divide the grounds on which.
belief in God has been alleged to be justified into three: religious ~
experience, revelation, and reason. Our judgment on religious
experience was that, while it might be overwhelmingly convincing
Ig
192 | HE WHO IS
to the person who was its subject, it was far too inexpressible to
provide by itself the basis of a formulated argument ;we observed
also that Catholic theology had beensina reluctant to admit, even
in genuine mystical” “experience, » any direct apprehension or
immediate knowledge. of the essence of God. We then glanced
at the contention off many neo- Protestants that the only knowledge
of God that is not hopelessly perverted is that given in the revela-
tion of Jesus Christ, and concluded with some discussion of the
relation between natural and revealed theology.
The next three chapters were devoted to an examination of the
traditional approach to the problem of God’s existence. In
Chapter Four the Ontological Argument of St. Anselm was seen
to have three vulnerable points; in particular in its attempt to
pass from the realm of logic to that of concrete existence. Atten-
tion was then turned to arguments from the existence of finite
being, and especially to the one general argument of which,
according to Fr. Garrigou-Lagrange, the five classical arguments
of St. Thomas are refinements.
In the Fifth Chapter the Thomist “Five Ways” were con-
sidered in detail, while in the Sixth Chapter 2an attempt was made
to estimate their significance and validity. The criticisms brought
by Professor Broad against all arguments of the cosmological type
were seen to be inconclusive, and indeed to rest upon the attribu-
tion to traditional theism of a view which it vigorously repudiates.
It was seen that the arguments depend upon the recognition of
finite beings as possessing both being and finitude. They are argu-
ments from finite being as such, and not from any special kind of
finite being; and, while they cannot give us a knowledge of what
God is in his inner essence, they can tell us a great deal about him
from the fact that he is the ultimate source of all being other than
his own. The question as to whether the Five Ways all necessarily
pointed to the same ultimate being as one another led us to
consider how far the syllogistic form really represented the bearing
of the arguments in their essential nature. We came to the con-
clusion that syllogistic statement was primarily a device for
persuading our minds to apprehend finite beings in their radical
finitude and thus, in so apprehending them, to apprehend the
existence of the God who is their Creator. The implications of
this conclusion were developed at some length, and an explanation
was offered of the difficulty experienced by modern people in
accepting the validity of the theistic arguments.
CONCLUSION 193
This led us, in Chapter Seven, to investigate the relation
between intellect and intuition, since the position arrived_at in
the previous chapter oobviously involved the attribution to the
human mind of the capacity to achieve some kind of intuition of
the essence of the beings which it experiences. Against all
epistemological theories of sensationalist type it was asserted that
the intellect is by its very nature equipped to abstract from finite
beings their intelligible content and to grasp, however partially
and obscurely, their inner essence. We saw that this doctrine was
at the heart of Fr. D’Arcy’s extremely impressive argument for
the theistic interpretation of the world; and we also saw that it, or
something very like it, was absolutely necessary if any answer was
to be found to one of the most important problems of philosophy,
which sensationalist systems were quite powerless to solve,
namely, the problem of induction. A note was added on the inter-
relations of Certainty and Certitude.
The Eighth Chapter took as its subject God and the World, and
developed the doctrine of analogia entis. In it we maintained that
the whole course of the preceding argument compelled us to
believe, on the one hand, that the world is entirely dependent
upon God, while, on the other, God is in no way dependent on
the world, so that the relation between God and the world is an
entirely one-sided one. On the part of God, therefore, creation is
an extra-temporal act and is not really distinct from preservation.
Two great_questions were then dealt with. The-first was how, if
God is the fullness of being, anything can exist other than God;
the second, why, granted that God can create a world, he should
decide to do so.
The answer to the first question was seen to lie in the fact that,
since God isstrictly infinite, God and the world do not “add up.”
The second-we admitted to enshrine an ultimate mystery; but we
pointed out that the very impossibility of answering it is a direct
consequence of the position that we have adopted. No answer,
other than the freedom of the divine will, can be given either to
the question why God should create a world at all, or to the
question why, if he does create one, it should be this one rather
than any other. Any world that God makes must be good, but no
world can be the best of all possible worlds, for any world that
God could make would fall infinitely short of his own perfection
and therefore leave room for an infinite number of better ones.
We concluded this chapter with some criticisms of Dr. W. R.
194 HE WHO IS 3
Matthews, and added a note on the teaching, in many respects
similar, of the late James Ward. an
In Chapter Nine we discussed the Divin é Attributes, so far as.
they were relevant to our main line of argument. The considera-
tion of freedom and omnipotence led on to some brief remarks on
miracles.
In the Tenth Chapter we considered at some length ve
Transcendence and Immanence_ of-God. The two elements were
seen to be mutually related, ‘and it was asserted that to lay stress
‘on one to the exclusion of the other leads to religious interpreta-
tions which are not only erroneous but also unstable. As instances
of transcendentalist _ error we mentioned Deism and Moham-
medanism. The former was seen to pass easily over into a
formalistic immanentism; while in the latter a compensating
factor for its extreme eciemdicue was found in the development
of Sufi mysticism. An obvious example of modern transcen-
dentalism within the sphere of Christianity appeared in the guise
of the neo-Protestant _ revival, with its strongly marked anti-
mystical trend.
Typical instances of unbalanced immanentism were found in
the religions of Asia: Brahmanism, Buddhism and Taoism. As
formulated European systems of immanentist type we considered
the writings of two modern Eastern Orthodox thinkers, Professor
Nicolas Berdyaev and Fr. Alexis van der Mensbrugghe. We ended
this chapter by pointing out how, both in sociology and in mystical
theology, it is essential to keep the balance between immanence
and transcendence. The mystical doctrine of Gardeil was sum-
marized, and it was seen, in consequence, that the Catholic
notion of the analogia entis between God and the world made it
possible to construct a mystical theology which was not vulnerable
to the attacks of the neo-Protestants.
In the following two chapters we discussed, in the light of the
conclusions already drawn, the approachesto theism of two of the
most typical and distinguished of modern natural theologians,
Professor A. N. Whitehead and Dr. F. R. Tennant. In both of
them we discovered certain defects. ~Whitehead’s adoption of
“creativity”’ as the fundamental characteristic of finite beings was
seen to have diverted his attention away from the existence of
things to their behaviour, and so to have caused him to be satisfied
with the notion of a purely immanent deity, while his neglect of
_ GONCLUSION 195
the concept of pilogia entis left him at last with only a finite God.
Tennant’s position-was far more satisfactory. His arguments for
the reality of the soul and of the extra-mental world were recog-
nized as of the highest value. But a certain hesitation to follow
these arguments to their natural conclusion and an unfortunate
introduction of ‘‘faith,” rather than of some kind of intuition, in
order to solve the problem of induction led him to a final position
which, while it is indeed far more acceptable than that of
Whitehead, falls short of that of traditional theism. Two notes
were added in which a brief discussion was made of two important
questions falling outside the strict limits of this book: namely, the
Moral Argument, with the closely related Problem of Evil, and
the connections between Natural Theology and the revealed
doctrine of the Holy Trinity.
Whether the attempt that has been made in this book to vindi-
cate the traditional doctrine of God and of his relation to the world
be judged as a success or as a failure, we cannot leave the subject
without emphasizing its supreme importance. For, on the one
hand, unless God is, in his essential nature, altogether exalted
above his world and independent of it, we shall have no absolute
standard-by which human exploits are to be judged; the logical
outcome of unbalanced immanentism is the moral relativism
which, both in personal life and in world affairs, is so evident at the
present day. On the other hand, unless, at the very heart of its
being, the world is in constant and intimate contact with God,
God will be so disconnected from the world as to be irrelevant to
it; and the logical outcome of unbalanced transcendentalism is a
doctrine of God as unknowable by man which can be counter-
balanced only by an anti-rational revelationism. Dr. Demant has
written, in words which we have quoted before, that if the faith
has steadily lost its transforming power over the last three cen-
turies, “the reason is to be found in the Christian mind having
split, in its innermost outlook, into cosmic interpretations which
bring God within the world process and purely redemptive
theologies. which take man as religious out of it, leaving his actual
existence at the mercy of its floods.”! Ifin this book we have been
concerned more with rebutting the immanentist than the transcen-
dentalist error, the reasonis that it is the former towards which
1 The Religious Prospect, p. 214. See p. 142 supra.
196 HE WHO IS Z
natural theology has tended in recent years ;the transcendentalist
| extreme is almost entirely represented by a certain recent but very
| influential school of theologians to whom the very idea of natural
| theology is anathema.
' In English-speaking countries it is still unfashionable, though
less so than formerly, to avow oneself to be an atheist; there is
thus a temptation to those who do not believe in the God of
traditional Christian theology to apply the word “‘God”’ to what-
ever entity or influence or principle they may suppose to underlie
the phenomenal world and then to ask what God is like. The
result is that man’s conception of God changes with every fashion
of thought. In contrast, the traditional approach of Christian
philosophy has been first to formulate the Christian conception
of God, and then to inquire whether he exists or not. In other
words, the traditional attitude has considered the primary problem
for investigation to be that of the existence of God, while the
“‘modern”’ attitude has considered the primary problem to be
that of his nature. This is more than just a difference of method.
/At touches the very nerve of the Christian life, for while the
/ traditional approach judges man by the measure of God, the
| “modern”? approach at least tends to judge God by the measure
| of man. The famous Aberdonian epitaph—
‘Here lie I, Martin Elginbrodde;
Ha’e mercy o’ my soul, Lord God,
As I would do, were I Lord God
And ye were Martin Elginbrodde”—
puts in a nutshell the anthropocentrism of this type of thought.
How different in effect is the ejaculation attributed to St. Augus-
tine: ““O my God, if I were God and thou Augustine, I would wish
that thou wert God and I Augustine!”’
For only a God who is, in St. Anselm’s phrase, “something than
which nothing greater can be thought,” or, as St. Thomas puts it,
‘‘subsistent Being itself,’ can be the adequate object of Christian
devotion, which has always believed itself to reach its climax in
the sheer adoration in which the creature, knowing its own entire
insufficiency and its relative nothingness, casts itself in complete
abasement before the majesty and holiness of a God whom it
recognizes. as being altogether complete without it and yet as
conferring upon it its very existence and tending it with the most
gentle and intimate love. “Religion is adoration,” wrote Baron
CONCLUSION 197
von Hiigel in a never-to-be-forgotten phrase,! and the name of
its Object
is Jam. If there is nothing in reality corresponding to
this concept, then the Christian life is based upon an utter and
tragic error, and the whole structure of Christian devotion falls
into ruins. The Catholic Christian, if he is convinced that no
such being exists, will not say that, after all, God is not so great
as he thought; he will say quite simply that, after all, there is no
God. Dicet in corde suo, Non est Deus. God may or may not exist,
but the Catholic will either believe in a God who is the self-
existent Being than which nothing greater can be thought, or he
will cease to claim that, in any sense in which he is interested in
the word, there is a God at all. Aut Deus aut nihil will be his motto;
Nc,
“and whether God exists or not, his glory he will not give to another.
To adore any being less than one who comprises in himself all
possible perfection would be to the Catholic a kind of conceptual
idolatry. The point at issue is well brought out in the dialogue
between St. Augustine and Evodius which Fr. M. C. D’Arcy
quotes in his essay on “The Philosophy of St. Augustine.” “If,”
the Saint says to Evodius, “we can find something indubitably
superior to our reason, would you hesitate to call that, whatever
it be, God?”’ And the answer of Evodius, with which Augustine
agrees, is, “I would not straightway ... call that God. For it is |
not one to whom my reason is inferior whom I would call God, but
one who has no superior.’’? And from the'religious point of view,
the reason why Professor A. E. Taylor’s conclusion is so pro-
foundly satisfactory in comparison with that of Whitehead or
Tennant is to be found in the place which it leaves for adoration.®
It is true that he finds it not primarily in the metaphysical realm,
but in that realm of religious experience which we have con-
sidered as being, when taken in isolation, of doubtful evidential
value. On the other hand, it must be recognized that he does not
take it in isolation, but as the culmination of a threefold argument
whose primary steps are provided by cosmological and moral
considerations. And we have also agreed with Fr. D’Arcy, on
the ground that the classical arguments lead us not merely to
intellectual acquiescence in a proposition but to a genuine appre-
1 Cf. Essays and Addresses, 11, p. 224. ‘‘The most fundamental need, duty, honour
and happiness of man, is not petition, nor even contrition, nor again even thanks-
giving; these three kinds of prayer which, indeed, must never disappear out of our
spiritual lives; but adoration.”” Cf. also I, p. 90; II, p. 233.
2 A Monument to St. Augustine, p. 167.
3 I do not, of course, suggest that this is an argument for its truth.
198 HE WHO Is
hension of God as present by immensity at the ontological root o
finite beings, that, while the argument from religious experience
must not be taken in isolation, there is a presentation of it in which
it appears as an interior view of the argument from contingency,?
and therefore has a real, though subsidiary, validity. In any case,
our conclusion remains, that the God of traditional theism is
identical with the God of Catholic devotion. Reason, as St.
Thomas says, tells us that God exists, rather than “what, in his
inner-nature,-he.is;. but reason is supplemented by revelation and,
although all Eibwiedee of God in this life must be through his
effects, when those effects are the unfolding of his secrets to the
soul as it lives in the Revelation itself, as it advances in its sharing
of the divine nature through the deepening of its sacramental
union with the humanity of the Incarnate Word, we can achieve
here upon earth a knowledge of God by grace as not merely the
God of the philosophers and scholars, nor even as just the God of
Abraham, Isaac and Jacob, but as the God and Father of our
Lord Jesus Christ, who is the effulgence of the Father’s glory and
the very image of his substance: a knowledge in comparison with
which all that the greatest philosophers can tell us fades into
insignificance. _ Nevertheless, \grace does not destroy nature, but
perfects it; reason is not annihilated by revelation, but supple-
mented and transformed by it. And, while the Angelic Doctor
himself left his great theological work unfinished on the ground
that he had seen something in comparison with which all that he
had written was as straw, he did not wish, like Faust, to burn his
books. To be the handmaid to theology is, for philosophy, a great
dignity, which confirms, instead of destroying, philosophy’s
autonomy in her own sphere. Philosophy comes to theology like
the Queen of Sheba when she visited King Solomon, bearing with
her treasures of inestimable price and yet awed into silence by the
glory that confronted her. Dominus meus et Deus meus! cried the
Apostle Thomas as he fell at the feet of the Risen Lord. Deus meus
et omnia! repeated St. Francis of Assisi through his night of prayer.
And Anselm and Aquinas are but echoing their worship when they
write that God is Aliquid quo majus nihil cogitari potest and Ipsum
esse subsistens.
1 See p. 92 supra.
CONCLUSION 199
And when the Queen of Sheba heard of the fame of Solomon
concerning the Name of the Lord, she came to pra him
with hard questions.
And she came to Jerusalem with a very great train, with camels
that bare spices, and very much gold, and precious stones:
and when she was come to Solomon, she communed with him
of all that was in her heart.
And Solomon told her all her questions: there was not anything
hid from the King, which he told her not.
And when the Queen of Sheba had seen all Solomon’s wisdom,
and the house that he had built,
And the meat of his table, and the sitting of his servants, and
the attendance of his ministers, and their apparel, and his
cupbearers, and his ascent by which he went up unto the
House of the Lord; there was no more spirit in her.
And she said to the King, It was a true report that I heard in
mine own land of thy acts and of thy wisdom.
Howbeit I believed not the words, until I came, and mine eyes
had seen it : and behold the half was not told me: thy wisdom
and prosperity exceedeth the fame which I heard.
Happy are thy men, happy are these thy servants, which stand
continually before thee, and that hear thy wisdom.
Go forth, O ye daughters of Zion, and behold King Solomon
with the crown wherewith his Mother crowned him in the
day of his espousals, and in the day of the gladness of his heart.
The Queen of the South shall rise up in the judgment with this
generation and shall condemn it: for she came from the ,
uttermost parts of the earth to hear the wisdom of Solomon :
and, behold, a greater than Solomon is here.
BIBLIOGRAPHY
(In each case the edition mentioned is the one that has been made
use of in this book; other editions are referred to in parentheses where
such mention seemed relevant. The dates given are normally those
on the title-pages.)
ANDREWES, Lancelot. A Pattern of Catechistical Doctrine and other Minor
Works. Oxford: Parker. 1846. :
Anglicanism. The Thought and Practice of the Church of England, illustrated
from the Religious Literature of the Seventeenth Century. Compiled and
~
edited by P. E. More and F. L. Cross. London: $.P.C.K. 1935.
ANSELM, St. The Devotions of St. Anselm. Ed. by CG. C. J. Webb. Lon-
don: Methuen. 1903.
AUGUSTINE, St. Confessions. E.T. by E. B. Pusey. Oxford: Parker.
1843. :
AyeER, A. J. The Foundations of Empirical Knowledge. London: Mac-
millan. 1940.
Language, Truth and Logic. London: Gollancz. 1936.
BarLur, John. Our Knowledge of God. Oxford Univ. Press. 1939.
Barnes, E. W. Scientific Theory and Religion. The World described by
Scvence and its Spiritual Interpretation. Cambridge Univ. Press. 1933.
BarTH, Karl. Credo. A Presentation of the Chief Problems of Dogmatics
with reference to the Apostles’ Creed. E.T. by J. S. McNab. London:
Hodder & Stoughton. 1936.
The Knowledge of God and the Service of God according to the Teaching of
the Reformation. E.T. by J. L. M. Haire and I. Henderson. Lon-
don: Hodder & Stoughton. 1938.
BERDYAEV, Nicolas. Freedom and the Spirit. E.T. by O. F. Clarke.
London: Geoffrey Bles. 1935.
Solitude and Society. E.T. by G. Reavey. London: Geoffrey Bles.
1938.
Bevan, Edwyn. Symbolism and Belief. London: Allen & Unwin. 1938.
BEVAN Jones, L. The People of the Mosque. An Introduction to the Study of
Islam with special reference to India. London: §.C.M. Press. 1932.
BEVERIDGE, William. Theological Works. Oxford: Parker. Vols. VII-
TX. 1845-7.
Bible of the World, The. Ed. by R. O. Ballou. London: Kegan Paul.
1940.
Brack, M. The Nature of Mathematics. A Critical Survey. London:
Kegan Paul. 1933.
Biosius, L. A Book of Spiritual Instruction (Institutio Spiritualis). E.T. by
200
BIBLIOGRAPHY 201
B. A. Wilberforce. (Vol. I of the 7-vol. ed. of The Works of Louis
de Blois.) London: Burns, Oates. 1925.
Box, H. S. The World and God. The Scholastic Approach to Theism.
_ London: S.P.0.K. 1934.
Brasant, F. H. Time and Eternity in Christian Thought. London:
Longmans. 1937.
Brav.ey, F. H. Appearance and Reality. A Metaphysical Essay. London:
Allen & Unwin. end ed., 1897. (1st ed., 1893.)
BreMonD, Henri. A Literary History of Religious Thought in France from
the Wars of Religion down to our own times. 3 vols. so far published.
London: $.P.C.K. 1928-36. (French original: Histoire littéraire
du sentiment religieux en France depuis la fin des guerres de religion jusqu’d
nos jours. Paris: Bloud et Gay. 1916-1932.)
Broan, C. D. Examination of McTaggart’s Philosophy. Cambadse Univ.
Press. 2 vols. in 3 parts. 1933, 1938.
BRUNNER, Emil. The Mediator. A Study of the Central Doctrine of the
Christian Faith. E.T. by Olive Wyon. London: Lutterworth
Press. 1924. (German original: Der Mittler. Tiibingen: Mohr.
1927.)
Buteakov, Sergius. The Wisdom of God. A Brief Summary of Sophiology.
E.T. by P. Thompson, O. F. Clarke and X. Braikevitch. London:
Williams & Norgate. 1937.
Burnasy, John. Amor Dei. A Study of the Religion of St. Augustine.
London: Hodder & Stoughton. 1938.
BuTLerR, Cuthbert. Western Mysticism. The Teaching of SS. Augustine,
Gregory and Bernard on Contemplation and the Contemplative Life.
Neglected Chapters in the History of Religion. London: Constable.
end ed., 1927. (1st ed., 1922.)
Butter, Joseph. Works. Ed. by W. E. Gladstone. 2 vols. Vol. I:
The Analogy, etc. Oxford: Clarendon Press. 1897.
CarnaP, Rudolf. Philosophy and Logical Syntax. London: Kegan Paul.
1935.
The Unity of Science. E.T. by M. Black. London: Kegan Paul. 1934.
CuapMaN, H. John. The Spiritual Letters of Dom John Chapman. Ed. by
R. Hudleston. London: Sheed & Ward. and ed., 1935. (1st ed.,
1935.)
Cartes, P. Prayer for All Times. E.T. by Maud Monahan. London:
Sands. 3 vols. 1929, 1930. (French original: La Priére de toutes les
heures. Paris: Desclée de Brouwer. 1936.)
CHESTERTON, G. K. The Everlasting Man. London: Hodder &
Stoughton. People’s Library ed. 1927. (1st ed., 1925).
Conrucius. The Analects, or the Conversations of Confucius with his Disciples
and certain others. E.T. by W. E. Soothill. Ed. by Lady Hosie.
Oxford Univ. Press. 1937.
202 'HE WHO IS
Coomaraswamy, Ananda. Buddha and the Gospel of Buddhism. London:
Harrap. 1916.
D’Arcy, M. C. The Nature of Belief. London: Sheed & Ward. 1931.
Thomas Aquinas. London: Benn. Oxford Univ. Press. 1930. ;
Dawson, Christopher. Enquiries into Religion and Culture. London:
Sheed & Ward. 1933.
DemantT, V. A. The Religious Prospect. London: Muller. 1939.
DeNzINGER, H. Lnchiridion Symbolorum Definitionum et Declarationum de
rebus fidet et morum. 11th edition, ed. by C. Bannwart. Friburgi
Brisgoviae: Herder. 1911.
Dewar, Lindsay. Man and God. An Essay in the Psychology and Philosophy
of Religious Experience. London: S.P.C.K. 1935.
Dogma in History and Thought. Studies by various writers. London:
Nisbet. 1929.
Eppincton, A. S. The Nature of the Physical World. Cambridge Univ.
Press. 1928.
New Pathways in Science. Cambridge Univ. Press. 1935.
The Philosophy of Physical Science.. Cambridge Univ. Press. 1939.
Ewin, Verrier. Christian Dhyana or Prayer of Loving Regard. A Study of
“ The Cloud of Unknowing.”? London: 8.P.C.K. 1930.
Emmet, D. M. Whitehead’s Philosophy of Organism. London: Macmillan.
1932.
Encyclopedia Britannica. London: Encyc. Brit. Co. 14th ed. 1929.
Encyclopedia of Religion and Ethics. Ed. by J. Hastings. Edinburgh:
T. & T. Clark. 13 vols. 1908-26. (Vol. XII, 1921.)
Essays Catholic and Critical. By Members of the Anglican Communion.
Ed. by E. G. Selwyn. London: $.P.C.K. grd ed, 1929. (1st ed.,
1926.) :
FRANGOIS DE SALES, St. Introduction a la Vie dévote. Paris: Nelson. n.d.
GasrRiEL OF ST. Mary Macpateng, Fr. St. John of the Cross, Doctor of
Divine Love. E.T. by a Benedictine of Stanbrook. London:
Thomas Baker. 1940.
GarpveiL, A. La Crédibilité et ?Apologétique. Paris: Gabalda. end ed.,
IQI2.
La Structure de ’Ame et V Expérience mystique. Paris: Gabalda. 2 vols.
grd ed., 1927. (ist ed., 1926.)
GARRIGOU-LAGRANGE, R. Dieu, son Existence et sa Nature. Paris:
Beauchesne. 1st ed., 1915. (E.T. by Bede Rose, God, his Existence
and Nature. St. Louis, Mo.: Herder. 2 vols. 1934.)
Perfection chrétienne et Contemplation, selon S. Thomas d’Aquin et S. Fean
de la Croix. Paris: Desclée. 2 vols. 7th ed., 1923.
Le Sens du Mystere et le clair-obscur intellectuel. Paris: Desclée de
Brouwer. 1934.
BIBLIOGRAPHY 203
Gmson, Etienne. God and Philosophy. New Haven: Yale Univ. Press;
and London: Milford. 1941.
The Mystical Theology of St. Bernard. E.T. by A. H. C. Downes.
London: Sheed & Ward. 1940. (French original: La Théologie
mystique de saint Bernard. Paris: Vrin. 1934.)
The Philosophy of St. Bonaventure. E..T. by I. Trethowan and F. J.
Sheed. London: Sheed & Ward. 1938. (French original: La
Philosophie de saint Bonaventure. Paris: Vrin. 1925.)
The Philosophy of St. Thomas Aquinas. E.T. by E. Bullough. Cam-
bridge: Heffer. 2nd ed., 1929. (French original: Le Thomisme.
Introduction au Systéme de saint Thomas d’Aquin. Paris: Vrin. 3rd ed.,
1927.)
Reason and Revelation in the Middle Ages. New York and London.
Scribner. 1939.
The Spirit of Medieval Philosophy. E.T. by A. H. C. Downes. London:
Sheed & Ward. 1936. (French original: L’ Esprit de la Philosophie
médiévale. Paris: Vrin. 2 vols. 1932.)
The Unity of Philosophical Experience. London: Sheed & Ward. 1938.
Christianisme et Philosophie. Paris: Vrin. 1936.
La Philosophie au Moyen Age de Scot Erigéne a G. d’Occam. Paris: Payot.
1930.
Réalisme thomiste et Critique de la connaissance. Paris. Vrin. 1939.
Gonsetu, F. Les Fondements des muthématiques. De la Géometrie d’ Euclide
a la Relativité générale et a VIntuitionisme. Paris: Blanchard. 1926.
Harris, C.R.S. Duns Scotus. Oxford: Clarendon Press. 2 vols. 1927.
HERBERT, A. P. Misleading Cases in the Common Law. London: Methuen.
1927.
Horton, W. M. Contemporary Continental Theology. An Interpretation for
Anglo-Saxons. London: 8.C.M. Press. 1938.
Hume, T. E. Speculations. Essays on Humanism and the Philosophy of Art.
Ed. by Herbert Read. London: Kegan Paul. 1924.
Houxtuey, Aldous. Ends and Means. London: Chatto & Windus. 1937.
Grey Eminence. A Study in Religion and Politics. London: Chatto &
Windus. 1941.
Incg, W. R. Christian Mysticism. London: Methuen. 1899.
Jackson, Thomas. Works. Vol. VY. Oxford Univ. Press. 1844.
Joun or THE Cross, St. Works. E.T. by E. Allison Peers. London:
Burns, Oates. 3 vols. 1934. (Spanish original is the critical
edition of P. Silverio de Santa Teresa. Burgos. 1929-31.)
Josrepu, H. W. B. An Introduction to Logic. Oxford: Clarendon Press.
and ed., 1916.
Joutan or Norwicu. Revelations of Divine Love. Ed. by G. Warrack.
London: Methuen. 1901.
204, HE WHO Is -
Karrer, O. Religions of Mankind. E.T. by E. I. Watkin. London:
Sheed & Ward. 1936.
Keynes, J. M. A Treatise on Probability. London: Macmillan. 1921.
Lairp, J. Theism and Cosmology. London: Allen & Unwin. 1940.
Mind and Deity. London: Allen & Unwin. 1941.
Lerpniz, G. W. Philosophical Writings. London: Dent. Everyman ed,
1934.
Lewis, C.S. The Problem of Pain. London: Centenary Press. 1940.
Macmurray, John. Creative Society. London: S.C.M. Press. 1935.
The Structure of Religious Experience. London: Faber. 1936.
McTaccart, J. McT. E. Studies in Hegelian Cosmology. Cambridge
Univ. Press. 2nd ed., 1918. (ist. ed., 1901.)
Malvern, 1941. The Life of the Church and the Order of Society. Being the
Proceedings of the Archbishop of Yorks Conference. London: Long-
mans. I94I.
Mar&EcHAL, Joseph. Studies in the Psychology of the Mystics. E.T. by
Algar Thorold. London: Burns, Oates. 1927. (Selected from
French original: Etudes sur la Psychologie des mystiques. Paris:
Desclée de Brouwer. 2 vols. 1924, 1937.)
Marirain, Jacques. The Degrees of Knowledge. E.T. by Bernard Wall
and Margot R. Adamson. London: Geoffrey Bles. 1937. (French
original :Distinguer pour unir. Paris: Desclée de Brouwer. 2nd ed.,
1932.)
Freedom in the
oa
Modern World. E.T. by R. O’Sullivan. London:
Sheed & Ward. 1935. (French original: Du Régime temporel et de
la Liberté. Paris: Desclée de Brouwer.)
An Introduction to Philosophy. E.T. by E. I. Watkin. London: Sheed &
Ward. 1930. (French original :Introduction générale a la Philosophie.
Paris: Téqui).
Religion and Culture. E.T. by J. F. Scanlan. London: Sheed & Ward.
1931. (French original: Religion et culture. Paris: Desclée de
Brouwer. n.d.)
St. Thomas Aquinas. E.T. by J. F. Scanlan. London: Sheed & Ward.
1938. (First English impression, 1931.) (French original: Le
Docteur Angélique. Paris: Desclée de Brouwer. 1929.)
True Humanism. E.T. by M. R. Adamson. London: Geoffrey Bles.
1938. (French original: Humanisme intégral. Paris. : Aubier. 1936.)
De la Philosophie chrétienne. Paris: Desclée de Brouwer. 1933.
Quatre Essats sur l’Esprit dans sa condition charnelle. Paris: Desclée de
Brouwer. 1939. .
Réflexions sur l’Intelligence et sur sa vie propre. -Paris: Desclée de Brouwer.
end ed., 1926. (1st ed., 1924.)
Mattuews, W. R. God in Christian Thought and Experience. London:
Nisbet. 1930.
BIBLIOGRAPHY 205
The Biicin of God. London: Nisbet. 1935.
Monument to St. Augustine, A. Essays on some aspects of his Thought written
in commemoration of his 15th Centenary. London: Sheed & Ward.
1930.
NEDONCELLE, M. Baron Friedrich von Htigel. A Study of his Life and
Thought. E.T. by Marjorie Vernon. London: Longmans. 1937.
Newman, J. H. An Essay in aid of a Grammar of Assent. London: Burns,
Oates. 4th ed., 1874. (1st ed., 1870.)
Niesunr, Reinhold. The Nature and Destiny of Man. A Christian Inter-
pretation. Vol. 1, Human Nature. London: Nisbet. 1941.
NyGren, A. Agape and Eros. A Study of the Christian Idea of Love. Part I.
E.T. by A. G. Hebert. London: S.P.C.K. 1932. Part II in
2 vols. E.T. by P. S. Watson. London: S.P.C.K. 1938, 1939.
(Swedish original: Den kristna kdrlekstanken. Stockholm: Svenska
Kyrkans Diakonistyrelses Bokf6rlag. 1930, 1932.)
Orro, R. The Idea of the Holy. An Inquiry into the non-rational factor in
the idea of the Divine and its relation to the rational. E.T. by J. W.
Harvey. Oxford Univ. Press. 3rd. imp., with additions, 1925.
‘(1st ed. 1923.) (German original: Das Heilige. 1st ed., 1917.)
PascaL, B. Thoughts. E.T. by W. F. Trotter. London: Dent. 1904.
(From French ed. of L. Brunschvicg. Paris: Hachette.)
Patterson, R. L. The Conception of God in the Philosophy of Aquinas.
London: Allen & Unwin. 1933.
Peck, W. G. The Salvation of Modern Man. London: Centenary Press.
1938.
Peers, E. Allison. Spanish Mysticism. A Preliminary Survey. London:
Methuen. 1924.
Peniwo, M. T-L. Le Réle de l’Analogie dans la Théologie dogmatique. Paris:
Wier. 293r.
PriecER, Karl. Wrestlers with Christ. E.T. by E. I. Watkin. London:
Sheed & Ward. 1936. (German original: Getster die um Christus
ringen. Pustet.)
Priturs, R. P. Modern Thomistic Philosophy. London: Burns, Oates.
2 vols. 1934, 1935.
Poutain, A. The Graces of Interior Prayer. A Treatise on Mystical Theology.
E.T. by L. L. Yorke Smith. London: Kegan Paul. 1910. (French
original: Des Graces d’Oraison. 6th ed., 1907.)
PrINcLE-Pattison, A. S. The Idea of God in the light of Recent Philosophy.
Oxford Univ. Press. 2nd ed., 1920. (1st ed., 1917.)
PrzywarA, E. Polarity. A German Catholic’s Interpretation of Religion.
E.T. by A. C. Bouquet. Oxford Univ. Press. 1935. (The German
original is an essay in Vol. II of Handbuch der Philosophie, ed.
Batimber and Schréter. Munich. 1927-30.)
Quick, O. C. The Christian Sacraments. London: Nisbet. 1928.
206 HE WHO IS
Reape, W.H.V. The Problem of Inference. Oxford: Clarendon Press.
1938.
Revelation Ed. by J. Baillie and H. Martin. London: Faber. 1937.
Ruys Davips, Mrs. Buddhism. A Study of the Buddhist Norm. London:
Williams & Norgate. 1912.
RickaBy, J. Studies on God and his Creatures. London: Longmans. 1924.
RousseLot, P. The Intellectualism of St. Thomas. E.T. by J. E.
O’Mahoney. London: Sheed & Ward. 1935. (French original:
L’Iniellectualisme de saint Thomas. 2nd ed.)
RussEtt, Bertrand. The Analysis of Mind. London: Allen & Unwin. —
1921.
An Inquiry into Meaning and Truth. London: Allen & Unwin. 1940.
An Introduction to Mathematical Philosophy. London: Allen & Unwin.
and ed., 1920. (1st ed., 1919.) -
Mysticism and Logic, and other Essays. London: Allen & Unwin. 1917.
An Outline of Philosophy. London: Allen & Unwin. 1927.
(See also under “Whitehead and Russell.’’)
Scumipt, W. The Origin and Growth of Religion. E.T. by H. J. Rose.
London: Methuen. end ed., 1935. (ist ed., 1931.)
SERTILLANGES, A. D. Le Christianisme et les Philosophies. Vol. I. Paris:
Aubier. n.d., but (?) 1939.
Foundations of Thomistic Philosophy. E.T. by G. Anstruther. London:
Sands. 1931. (French original: Les Grandes Théses de la Philosophie
thomiste. Paris: Bloud et Gay. 1928.)
SHEED, F. J. Communism and Man. London: Sheed & Ward, 1938.
SmiTH, Margaret. An Early Mystic of Baghdad. A Study of the Life and
Teaching of Harith B. Asad al-Muhasibi. London: Sheldon Press.
1935-
Studies in Early Mysticism in the Near and Middle East. London:
Sheldon Press. 1931.
SotovyEv. God, Man and the Church. The Spiritual Foundations of Life.
E.T. by Donald Attwater from the French version of G. Tsebrikov
and A. Martin. London: James Clarke. n.d. (Russian original:
Religiosnya Osnovy Zhizni, written in 1882-4.)
Sortey, W. R. Moral Values and the Idea of God. Cambridge Univ.
Press. 2nd ed., 1921. (1st ed., 1918, but lectures delivered
1914-15.)
Spens, W. Belief and Practice. London: Longmans. 2nd ed., 1917.
STEBBING, L. S. Philosophy and the Physicists. London: Methuen. 1937.
Studies in Comparative Religion. General editor: E. C. Messenger. Lon-
don: Catholic Truth Society. 5 vols. n.d.
Tawney, R. H. The Acquisitive Society. London: G. Bell. 1924.
Taytor, A. E. The Faith of a Moralist. London: Macmillan. 2 vols.
1930.
BIBLIOGRAPHY 207
Tempe, W. Nature, Man and God. London: Macmillan. 1934.
TENNANT, F. R. Miracle and its Philosophical Presuppositions. Cambridge
Univ. Press. 1925.
Philosophical Theology. Cambridge Univ. Press. 2 vols. 1928, 1930.
Philosophy of the Sciences, or The Relations between the Departments of
Knowledge. Cambridge Univ. Press. 1932.
Tuomas Aguinas, St. De Potentia Dei. E.T. by the English Dominican
Fathers. London: Burns, Oates. 1932-4.
Summa contra Gentiles. E.T. by the English Dominican Fathers.
London: Burns, Oates. 1924-9.
Summa Theologica. E.T. by the English Dominican Fathers. London:
Burns, Oates. 1920-5.
Turner, W. History of Philosophy. New York: Ginn. 1903.
VAN DER MENSBRUGGHE, Alexis. From Dyad to Triad. A Plea for Duality
against Dualism and an Essay towards the Synthesis of Orthodoxy. Lon-
don: Faith Press. 1935.
VERRIELE, A. Le Surnaturel en nous et le Péché originel. Paris: Bloud et |
Gay. New ed., 1934.
von Huce., F. Essays and Addresses on the Philosophy of Religion. London:
Dent. 2 vols. 1921, 1926.
The Mystical Element of Religion as studied in Saint Catherine of Genoa
and her Friends. London: Dent. 2 vols. 2nd ed., 1923. (1st ed.,
1908.)
Warp, James. Essays in Philosophy. Cambridge Univ. Press. 1927.
Naturalism and Agnosticism. London: Black. 4th ed., 1915. (1st ed.,
1899.)
The Realm of Ends, or Pluralism and Theism. Cambridge Univ. Press.
and ed., 1912. (1st ed., 1911.)
Watkin, E. I. The Philosophy of Mysticism. London: Grant Richards.
1920.
WEINBERG, J. R. An Examination of Logical Positivism. London: Kegan
Paul. 1936.
WHirEHEAD, A. N. Adventures of Ideas. Cambridge Univ. Press. 1933.
The Concept of Nature. Cambridge Univ. Press. 1920.
An Enquiry concerning the Principles of Natural Knowledge. Cambridge
Univ. Press. 2nd ed., 1925. (1st ed., 1919.)
Modes of Thought. Cambridge Univ. Press. 1938.
Process and Reality. An Essay in Cosmology. Cambridge Univ. Press.
1929.
Religion in the Making. Cambridge Univ. Press. 1926. (There is an
edition of 1927 with different pagination.)
Science and the Modern World. Cambridge Univ. Press. 1926. 84 in.
by 5% in. (There is another edition of the same date and from the
same press on smaller pages with different pagination.) Also in
208 HE WHO IS
Pelican Books, 1938. _ (References to this edition are given in
parentheses. )
Symbolism, its Meaning and Effect. Cambridge Univ. Press. 1928.
WuireHeEap, A. N. and RussELL, B. Principia Mathematica. Cambridge ©
Univ. Press. 1st ed., 1910-13. (2nd ed., 1925.)
WicksTEED, P. H. The Reactions between Dogma and Philosophy illustrated
Srom the Works of s. Thomas Aquinas. London: Williams & Norgate.
1920.
WiuiaAMs, Charles. He came down from Heaven. London: Heinemann.
1938.
INDEX
[Where a number of references occur under one name the most important are
indicated by heavy type.]
Abelard, Peter, 108 ~ Bossuet, J. B., 108
Abraham, Isaac and Jacob, 198 Bouquet, A. C., 126
Addison, Joseph, 127 Box, H. S., 57
Albertus Magnus, St., x Brabant, F. H., 101
al-Ghazali, 130 Bradley, F. H., 66, 78
al-Hallaj, 22, 130 Braithwaite, R. B., 52
Amos, 5 Bremond, Henri, 105
Andrewes, Lancelot, 27 Broad, C. D., 33 f., 52, 57 f., 67-8, 165,
Angelus Silesius, 148-9 175, 192
Anselm, St., 11 f., go f., 51, 67, 70, 113, Brunner, Emil, 24, 131, 148-9
192, 198 Bulgakov, Sergius, 140, 189
Arendzen, J. P., 140 Buridan, x oe
Aristotle, vili, 7, 9, 25-6, 37, 40, 44, 50, Burnaby, John, 110
79> 157 ; Butler, Cuthbert, 131
Augustine, St., viii, 10, 11, 54, 62, 76, Butler, Joseph, 26, 28
100, I10, 184-5, 196-7
Averroés, 40 Cajetan, St., 119
Avicenna, 40 Caldin, E. F., 51
Ayer, A. J., 66 f., 90, 121 Carnap, Rudolf, 66
Cecilia of the Nativity, Mother, 149
Baillie, John, 93 Chapman, H. John, 22, 26, 93, 110
Barnes, E. W., 17 f., 22 Charles, Pierre, 109
Barth, Karl, 1, 23 f., 28, 131, 135, 148 Charlier, L., x
Bellarmine, St. Robert, 95 Chekhov, Anton, 67
Berdyaev, Nicolas, 10, 82, 103, 135 f., Chesterton, G. K., 4, 117
140, 149, 194 Cleopatra, 72
Bergson, Henri, 42 Coleridge, 8S. T., 97
Berkeley, George, 117, 122 Confucius, 132
Bernard, St., 11, 108 Congar, M. J., x
Bevan, Edwyn, 119, 148 Coomaraswamy, Ananda, 132, 134
Bevan Jones, L., 127, 130 Crusoe, Robinson, 69
Beveridge, William, 64, 76
Black, M., 33, 48 D’Arcy, M. C., 10, 21, 85 f., 165, 193,
Blosius, Ludovicus, 142 197
Boehme, Jacob, 136, 148 David of Dinant, 138
Bonaventure, St., vili, 11, 30, 44, 105, 184 Dawson, Christopher, 22, 130
INDEX 209
de Lubac, H., x Jackson, Thomas, xi, 13, 63, 97, 119, 120,
Demant, V. A. , Vii, 141f., 195 124
Denzinger, eee "28, 97 James, William, 17
Descartes, René, vill, 32, 67, 113, 122, Jefferies, Richard, 18, 22
129, 151, 163 Jeremiah,
Dewar, Lindsay, 16 John of the Cross, St., 11, 63, 143, 147-9
Duns Scotus, viii, 10, 13, 32, 119, 122 John of Damascus, St., viii, 55
John the Evangelist, St.rg; L8G
Eckhart, Meister, 136, 148-9 Joseph, H. W. B., 87
Sil A. S., 51, 68, 99, 100, 124, Julian of Norwich, 1o1, 126
I Julius Caesar, 60-1
Elginbrodde, Martin, 196
Elijah, 5 Kant, I., ix, 32, 44, 57, 67, 90, 122, 151,
Elwin, Verrier, 22 156, 164-5, 169, 17
Emmet, Dorothy M., 150, 156 Karrer, Otto, 4
Evodius, 197 Keith, A. Berriedale, 185
Ezekiel, 5 Kelly, Bernard, 119
Keynes, J. M., 89 f., 164
Faithfull, C. E., 74 Krivoshein, Basil, 135, 141.
Farges, Albert, 148
Fénelon, Frangois de S. de la M., 108 Laird, John, 112, 145, 165
Fichte, J. G., 107, 122 Lao-Tse, 132
Florensky, Paul, 140 Leibniz, G. W., 32, 44, 105
Foster, Kenelm, 54 Lewis, C. S., 182
Francis of Assisi, St., 198 Locke, John, 163
Francis of Sales, St., 142 Lunn, Arnold, 80
Friday, Man, 69 Luther, Martin, 131
Frisbee, Graham, 178
MacKinnon, D. M., 67, 104
Gabriel of St. Mary Magdalene, Father, Macmurray, John, 110
22 McTaggart, J. McT. E., 37, 61, 67, 182
Gardeil, A., x, 122, 144 f. Malebranche, Nicolas, viii, 122
Garrigou-Lagrange, R., x, xi, 10, 22, 34, Mansel, H. L., 119
37f., 44, 47, 49, 57, 71, 80, 90, 93, 99, March, W. W.S., 117
105, 109, 117, 119-20, 148, 176, 192 Maréchal, Joseph, 21, 22, 68, 130, 131
Gaunilo, 32 Maritain, Jacques, ix, x, 6, 10, 22, 25, 28,
Gautama Buddha, 134 63, 70, 76, 78-9, 82, 100, 105, 120-1,
George IV, King, 8 124, 129, 131, 134, 141
Gill, Eric, 54 Marrow, Mr. Justice, 14
Gilson, Etienne, viii—x, 5, 17, 25, 28, 30, Massignon, Louis, 130
32, 35, 40, 47, 49, 67, 85, 96-7, 108, Matthews, W. R., 105 f., 156, 172, 178,
122, 175, 184 194
Goebbels, J., 72 Mercury, 53
Gonseth, F., 33, 60, 74 Mersch, Emile,x
Gosse, Philip, 2 Meyerson, 50
Gregory Palamas, St., 135 Michael Angelo, 98
Mill, John Stuart, 87-8
Hanson, Richard, 57, 96 Mohammed, 130 ;
Harris, C. R. S., 32 Mohasibi, 130
Hegel, G. W. F., 97, 107, 113, 185 Molina, Luis de, 119
Herbert, A. P., 14 Moses, 5, 36
Hooper, S. E., 150 Mozart, W. A., 98
Horton, W. M., 126
Hosea, Nédoncelle, Maurice, 184
Hulme, T. E., 157-8 Nelson, Horatio, 103
Hume, David, 123 Neurath, Otto, 66
Huxley, Aldous, 22, 143 Newman, J. H., 85 f.
Newton, Isaac, 42 f., 100, 127, 166
Inge, W. R., 148 Nicholas of Autrecourt, x
Trenaeus, St., 109 Nicholson, R. A., 130
Isaiah, 5 Nicod, J., 89
210 HE WHO IS
Niebuhr, Reinhold, 148-9 Smith, Margaret, 130
Noél, L., 68 Solomon, King, 198
Noyes, Alfred, 18 Solovyev, yoo 18, 111, 135, 140-1
Nygren, A., 110, 131, 148 Sorley, W. R., 57, 181
Spencer, Herbert, 187
Ossuna, 105 Spens, Will, 67
Otto, R., 16 Spinoza, B., 160
Sri Ramakrishna, 134
Paley, William, 54 Stebbing, L. Susan, 52, 67, 99, 158
Pascal, Blaise, 75
Patterson, R. L., 47, 57, 70 Tawney, R. H., 128
Paul, St., 9, 149, 189 Taylor, A. E., ix, 21, 32, 33, 48, 50, 57-8,
Peck, W. G., 3 104, 160, 181 f., 197
Peers, E. Allison, 144 Temple, William,
97, 106, 181
Penido, M. T-L., xi, 103, 117 Tennant, F. R., 19 f., 83, 94, 97, 125,
Pfleger, Karl, 18 150, 161 f., 179 f.,a 194, 197
Phillips, R. P.
-» 10, 34, 48-9, 55, 70, 116 f. Teresa of Avila, St.,
Picard, G., 68 Thomas the Apostle, =25 100
Plato, 7, 187 Thomas Aquinas, St., viii, x, Xi, 4, 7, 9,
Pope, Alexander, 128 12, 13, 25-7, 30f., gof., 57, 62 f., 83,
Poulain, A., 21, 148 85, 93> 95» 99-IOI, 10g, 112, 114, I 16—
Pringle-Pattison, A. S., 97, 139, 172 17, 121, 138, 149, 160, 167, 172-3, 185,
Przywara, Erich, xi, 126, 143-4 189, 191, 196, 198
Thomas, Ivo, 93.
Quick, O. C., 67 Troeltsch, Ernst, 131 .
Turner, William, 7
Rashdall, Hastings, 169
Reade, W. H. V., 90 van der Mensbrugghe, Alexis, 135, 137 f.,
Rhys Davids, Mrs., 134 194
Rickaby, J., 40 Verriéle, A., 125
Roland-Gosselin, M. D., 68 von Hiigel, Friedrich, 108, 184, 197
Rousselot, Pierre, 78, 84, 85 f.
Russell, Bertrand, 33, 48, 66 f., 83, 87 f., Ward, James, 97, 113 f., 171-2, 182, 194
150, 152, 162, 183 Warrack, Grace, 102
Watkin, E. I., 18, 22, 143, 149
Sanderson, Robert, 63 Weinberg, J. R., 66, 89
Sankara, 149 Wesseling, Theodore, 141-
Saudreau, A., 148 White, Victor,-47
Scheeben, 38 Whitehead, A. N., x, xii, I, 42, 51, 90, 97,
Schelling, F. W. J. von, 122 100, I2I, 124, 127, 150f., 162, 171,
Schlick, M., 66 175, 178, 179-80, 182, 187 f., 194, 197
Schmidt, W., 4 Wicksteed, P. H., 40
Schopenhauer, A., 107 William of Ockham, viii, x
Sertillanges, A. D., x, 25, 105 Williams, Charles, 112
Sheba, Queen of, 198 Wittgenstein, L., 65-6
Sheed, F. J., 184 Wren, Christopher, 127
THEOLOGY LIBRARY
CLAREWONT, CALIF.
oFir by ad
eo Y: yO
Mascall, Eric Lionel, 1905-
200 He who is; a study in traditional theism, by E. L.
| M32 London, New York ,etc.; Longmans, Green and «
| 1943 MLL, hy, 210 p. 225 cm:
“First published 1943.”
Bibliography : p. 200-208.
Includes index.
1. Theism. I. Title,
BL200.M32 211 CCSC/1t
Library of Congress “ 59h1,
siete cheetah
petal
Ontos
ielslterenet
eat
ey
nea
es
fetels,¢
ebartecartece r
Peters
Piet
Bois puss
cist ee
pals
Hi iat
t
ie
itt
es
Se
n=}
nites
eae= Sate
Seisan =
ice
seas
petstita
sist
He
tS)
stat
tL
artnet
Tete
suet rit
it
seaet
ag
- eiepeitrty f
‘rf‘sryaeh oe
Tyan se erates
ester
ee
o}
: Bek:Faitetet aes
Side phere : adie
- a5 tga
is
et eiebrettrnmetrntteer iY ; sates
; i¢
iaieity
solehle
ases
i2is
erin
sit ieegit
2esaoee
ee ir tin eran ; acne e ald ies He if : s f saa itr +Bi
Pash
Css
i‘
si
eet
; Ns
n onipreeree
nadmaaatta fe s Soe cpa ater:
neta:
retelrct
BH st raSeeoe
i ects a se Te le ei
Hane
iatoracere rael eueee a ; ; ert
i i piers ae uyie
RATE stress
pie
ASeIpEgeateatlsat Neen sera
Beste
seca
ers
j hoes i a sa
pistes
aie5
Hata
toneeterare
uewearretreseirehe
raga
ae
aint
Ss
y ss
saci
ia
hee
Piste
naan
; za2 &
Paneer
g
nie
ana
Be
hehe
se
ator
th
Ht,‘Ses
eeeater
Srnaounnieebtt
;
ae
;
gentry
aera
et
Bsa
oe
Ba
eeset
fret
Ie
tat
:
Sieatte
itty as mise 3 it
seltaseett f
ee ae Blea ite arate
alee Seape esate
ie acre
Seaecohese
a
Bs
ISR
;
:
Sarit eteperat
cas
aie
: “ z
f resiarettrealenn niett seeeeatseemt titrae arte %
BhLEcok ee ple acti irate tetra tf
sakes
=
Sis a 4 ee ests sae
theta
Beet ;
eletien
a
cae
renee
easy
ii
terre
Perris
shibchies 7 z iret tte! ;
53
aas:
3
‘4
tr rebrialtstra rotates
Beth
Hey peat
sae
ae
Fi
Be
tebe
a
s
ie
SS Wak
ititesty
a