Introduction of philosophy
UNIVERSITY 
OF FLORIDA 
LIBRARIES
Digitized by the Internet Archive 
in 2010 with funding from 
Lyrasis IVIembers and Sloan Foundation 
http://www.archive.org/details/introductiontophOOmari
Introduction of philosophy
Introduction of philosophy
Introduction of philosophy
AN INTRODUCTION TO 
PHILOSOPHY
BT THE SAME AUTHOR 
THREE REFORMERS 
LUTHER—DESCARTES—ROUSSEAU 
PRAYER AND INTELLIGENCE 
TRANSLATED BY ALOAR THOROLD 
ART AND SCHOLASTICISM 
TRANSLATED BY J. F. SCANLAN 
THE THINGS THAT ARE NOT 
CAESAR'S 
TRANSLATED BY J. F. SCANLAN 
ST. THOMAS AQUINAS—ANGEL 
OF THE SCHOOLS 
TRANSLATED BY J. F. SCANLAN 
RELIGION AND CULTURE 
TRANSLATED BY J. F. SCANLAN 
FREEDOM OF THE INTELLECT 
TRANSLATED BY F. J. SHEED 
FREEDOM IN THE MODERN WORLD 
TRANSLATED BY R. o'sULLIVAN
AN INTRODUCTION TO 
PHILOSOPHY 
BY 
JACQUES MARITAIN 
TRANSLATED BY 
E. I. WATKIN 
NEW YORK 
SHEED & WARD, INC.
Published by 
Sheed & Ward, 31 Paternoster Row, London, E.C.4 
and by 
Sheed & Ward, Inc., 63 Fifth Avenue, New York 
First printing, March, igjy 
Twelfth printing, April, ig4'j 
Printed in the United Stales ofAmerica
PUBLISHER'S NOTE 
The French edition of this work, under the title Introduction 
G^ndrale a la Philosophic, appears as the first volume of 
seven, which deal with Formal Logic, Theories of Knowledge, 
Cosmology, Psychology, Metaphysics, Ethics, ^Esthetics, and the 
History of Philosophy. But, since six out of the seven volumes 
remain to be written, it has been thought better to issue the present 
volume quite independently. 
The series as a whole is intended to provide text-books for a 
regular university course as it is found in France, and with that 
particular end in view prints in larger type those paragraphs 
which the student should read first, and in smaller type those 
paragraphs which are merely explanatory or expansive of them. 
This schematising has been abandoned to make the volume serve 
for general readers, no university course being envisaged in England 
or America. 
The translation has been madefrom the eleventh French edition.
Introduction of philosophy
PREFACE 
My chief aim in composing an Elements of 
Philosophy series, to which this book may serve as 
an introduction, is to give a faithful presentation of 
the system of Aristotle and St. Thomas, and in its 
Ught to judge the important systems which have 
followed each other during the last three centuries 
and the principal problems discussed by modern 
philosophy. I have tried to adapt the method of 
exposition to contemporary conditions, and in parti-cular 
have followed of set purpose a progressive order 
of exposition—as far as possible the order of intellectual 
discovery—never appealing to any truth not already 
known and understood, and never introducing a 
new notion or proposition for which the way has not 
been prepared by those which have gone before it 
and led up to it. The method has obliged me to 
depart on several points from the procedure of the 
traditional text-books—above all, considerably to 
magnify the importance and extend the scope of this 
Introduction. Yet, thus, I have but returned to the 
method followed by Aristotle himself. The first three 
books of his Metaphysics are, in fact, nothing but an 
extensive introduction. 
A work of this kind, if it is to be thorough, demands 
the detailed discussion of certain points, without which 
the study it seeks to promote would lose all its value 
as a mental discipline. I should be untrue to tradi-
PREFACE 
tional philosophy if I reduced it to a few main theses 
which have lost their freshness, and a few common-places 
of a spiritualist metaphysic, and neglected to 
bring out its fine intellectual contours and display its 
power of penetrating analysis. 
The present work is intended for beginners. It 
can therefore make no attempt to reproduce the 
depth or the wealth of subtle dialectic to be found in 
treatises written for specialists, and remains strictly 
elementary. It must, however, preserve the scientific 
character proper to a philosophical exposition. 
Some readers may take alarm at scholastic termino-logy. 
Yet no science, no discipline, no form of sport, 
even, or industry, can dispense with a special termino-logy— 
often far more arid and artificial than the 
vocabulary of philosophy. To require that philo-sophers 
should use everyday language implies that 
their science is just an enterprising topic of conversa-tion, 
idle arm-chair speculation for after dinner. On 
the other hand, it may legitimately be demanded 
that no technical term be used until it has been 
clearly defined. 
Finally, I would say that, if the philosophy of 
Aristotle, as revived and enriched by St. Thomas 
and his school, may rightly be called the Christian 
philosophy, both because the Church is never weary 
of putting it forward as the only true philosophy 
and because it harmonises perfectly with the truths of 
faith, nevertheless it is not proposed here for the 
reader's acceptance because it is Christian, but because 
it is demonstrably true. This agreement between a 
8
PREFACE 
philosophic system founded by a pagan and the dogmas 
of revelation is no doubt an external sign, an extra-philosophic 
guarantee of its truth ; but it is not from 
its agreement with the Faith, but from its own 
rational evidence, that it derives its authority as a 
philosophy. 
Nevertheless, reason and faith, while distinct, are 
not separate, and, since I am writing principally for 
Christian readers, I have not denied myself an occa-sional 
reference to knowledge familiar to every 
Catholic, or to certain theological applications of 
philosophic principles, the better to put philosophy 
in its proper place in Christian minds, or to help them 
to maintain the unity of their thought. The fact 
remzdns that in our arguments and in the very structure 
of our exposition of philosophy, it is not faith, but 
reason, and reason alone, which occupies the entire 
ground and holds undivided sway.
Introduction of philosophy
CONTENTS 
INTRODUCTORY 
PACK 
17 
PART ONE 
THE NATURE OF PHILOSOPHY 
I. PHILOSOPraC THOUGHT BEFORE PHILOSOPHY IN 
THE STRIC3T SENSE , . . . .23 
Introduction . . . . -23 
Primitive tradition .... 24 
The Semites and the Egyptians . . 25 
The Indo-Europeans : ... 26 
(a) The Persians .... 26 
(b) The Indians : . . . -27 
(i) Brahmanism ... 28 
(ii) Buddhism • • • 33 
(iii) Other schools . . 36 
(c) The Chinese .... 38 
Limitations of human wisdom . . 42 
The Greeks the chosen people of reason 44 
THE PRE-SOCRATIC PHILOSOPHERS ... 46 
The Sages ..... 46 
The lonians : . . . . -47 
(a) Thales and his successors . . 48 
(b) The great physicists : . . 50 
(i) Heraclitus ... 50 
(ii) Democritus ... 52 
(iii) Anaxagoras • . . 54 
The Italians : Pythagoras • . • 55 
The Eleatics : Parmenides ... 60 
u. 
II
CONTENTS 
m. THE SOPHISTS AND SOCRATES . 
Introduction 
The sophists 
Socrates .... 
(a) Ethics and knowledge 
(b) Irony, maieutic, dialectic 
(c) Moderate intellectualism 
64 
65 
68 
69 
70 
72 
IV. PLATO AND ARISTOTLE . 
The minor Socratics . 
Plato .... 
(a) His theory of ideas . 
(b) His system of philosophy 
(c) Its limitations . 
Aristotle .... 
(a) Corrections of Plato . 
(b) The Aristotelian system 
(c) Aristotle's works 
Aristotle and St. Thomas . 
Philosophia perennis 
74 
75 
75 
78 
81 
82 
83 
87 
92 
97 
99 
V, DEFINITION OF PHILOSOPHY . 
Scientific knowledge . 
Its material object 
Its formal object 
Conclusion I 
Further considerations 
102 
103 
107 
108 
loB 
VI, PHILOSOPHY AND THE SPECIAL SCIENCES 
Philosophy judges the special sciences 
It governs them 
It defends them 
It is pre-eminendy free 
Further observations . 
Conclusion II 
12 
III 
113 
117 
118 
118 
123
CONTENTS 
VII. PHILOSOPHY AND THEOLOGY . . . . I24 
Nature of theology . . . . 1 24 
Theology judges philosophy . .126 
Philosophy submits to theology its 
conclusions, not its premises . 126 
Philosophia ancilla theologiae . . .129 
Further considerations . . .129 
Conclusion III . -132 
VUI. PHILOSOPHY AND COMMON SENSE . . '133 
Unscientific knowledge . . -133 
Philosophy is derived from common 
sense, understood as the natural 
apprehension of first principles . 136 
Common sense may accidentally judge 
philosophy . . . . .138 
Conclusion IV . . 141 
The method of philosophy . . 141 
PART TWO 
THE CLASSIFICATION OF PHILOSOPHY 
I. THE MAIN DIVISIONS OF PHILOSOPHY . 
Logic. Theoretical philosophy. Prac 
tical philosophy .... . 
147 
Their objects 151 
Conclusion V . 153 
II. LOGIC 154 
Correct reasoning 
Ideas and images 
154 
154 
Conclusion VI . 156 
Individual and universal . 157 
Conclusion VII ^59 
147 
13
CONTENTS 
The problem of universals 
(a) Nominalism 
(b) ReaUsm . 
(c) Moderate Realism 
159 
159 
160 
160 
III. THE PHILOSOPHY OF MATHEMATICS AND THE 
PHILOSOPHY OF NATURE 
The term body . 
The philosophy of mathematics 
The philosophy of nature 
(a) Mechanism 
(b) Dynamism 
(c) Hylomorphism 
Psychology 
Problem of the origin of ideas 
Conclusion VIII 
. 163 
• 163 
164 
. 166 
. 166 
169 
. 170 
Abstraction : Problem ofhuman nature 
Conflicting schools .... 
172 
172 
174 
163 
IV. CRITICISM (ePISTEMOLOOY) 
Being qua being 
. 
Criticism .... 
Problem of truth 
Conclusion IX 
Conflicting schools : . 
(a) Scepticism 
(b) Rationalism 
(c) Moderate intellectualism 
Problem of the object of the intellect 
Conclusion X . 
Being and intelligibility 
Conclusion XI . 
14 
178 
178 
179 
181 
181 
182 
182 
183 
185 
187 
187 
187 
178
CONTENTS 
V. ONTOLOGY : ESSENCE 
Problems of ontology 
. 
Essence 
(a) In the wide sense 
{b) In the strict sense 
Characteristics of this essence 
Conclusion XII 
Further observations . 
Our intellect can apprehend essence 
Conclusion XIII 
Further observations . 
Essence is universal in the mind . 
Conclusion XIV 
Individual nature and matter : 
(a) Individual nature 
(b) First matter 
(c) Archetypal being 
(d) Nature, essence, and quiddity 
189 
189 
191 
191 
194 
197 
201 
201 
203 
203 
203 
205 
207 
207 
208 
209 
210 
213 
VI. ONTOLOGY ! SUBSTANCE AND ACQDENT 
Origin of these notions 
Substance .... 
Conclusion XV 
Further observations 
Accident ..... 
Conclusion XVI 
Further observations . 
Conflicting schools 
The individuality of substance . 
(a) Substantia prima, substantia secunda 
(b) Per se, a s«, in se . . . 
217 
217 
222 
224 
224 
227 
227 
228 
230 
233 
233 
236
CONTENTS 
VII. ONTOLOGY : ACT AND POTENTIALITY 
Origin of these notions 
(a) Identity and change . 
(b) Their apparent incompatibility 
(c) Solved by the concept potentiality 
Potency or potentiality 
Act 
Conclusion XVII 
The nature of change 
Act and potentiaHty in things 
Axioms i-vii .... 
Conflicting schools 
Terminology : . 
(a) Material and formal 
(b) Virtual and formal (actual) 
(c) Implicit and explicit 
(d) In express act, in accomplished 
act . . . . . 
239 
239 
239 
241 
242 
242 
244 
245 
246 
246 
248 
250 
251 
252 
254 
255 
255 
VIII. THEODICY (natural THEOLOGY) 
Subsistent being itself 257 
257 
IX. THE PHILOSOPHY OF ART. ETHICS 
Introduction 
The philosophy of art 
Ethics 
Divisions of ethics 
Conflicting schools 
261 
261 
261 
264 
267 
268 
CONCLUSION .... 
Classification of philosophy 271 
271 
16
INTRODUCTORY 
' 
Philosophers were once called wise men. It was 
Pythagoras who first invented the term philosophy 
(91X1* TT^i^ (To^iat;', love of wisdom),^ observing that 
wisdom belongs in the strict sense to God alone, and 
for that reason not wishing to be called a wise man, 
but simply a friend or lover of wisdom. His modesty 
was itself a mark of great wisdom, for the sublimity 
and difficulty of the highest truths, and the weakness 
of our nature "in so many respects enslaved," forbid 
man to acquire " a property right in wisdom " 
such that he can employ it in entire freedom. As a 
result of the many necessities to which he is subject, 
he holds it only by an insecure title, so that he may 
be termed not wise, but far more truly a beggar at 
wisdom's door. Nevertheless philosophy is nothing 
other than wisdom itself so far as it is accessible to 
human nature. 
It is not a wisdom supernaturally infused into our 
souls which man possesses in virtue of a superhuman 
illumination. Neither is it a wisdom wholly sponta-neous 
and unconscious (such as within its limits is 
the prudence of animals, and even the wisdom of 
simple souls), which he possesses in virtue of a natural 
instinct. It is the wisdom of man as man, which he 
' Cicero, Tusc, v. 8 ; cf. Diogenes Laertius, i, 12. 
» Aristotle, Metaph. i, 2, 982 b. St. Thomas, In I Metaph., 1. 3. Cf. 
De Veritate, q., 7, a. 7. 
17
AN INTRODUCTION TO PHILOSOPHY 
acquires by the labour of his intellect, and it is for 
that very reason that his wisdom is gained with 
such difficulty and held so insecurely, and that those 
who seek it should be called philosophers rather than 
wise men. 
Such is the nature of philosophy derived from the 
etymology of the term and its employment in ordinary 
speech. A philosopher is a man humanly wise. 
And the man who devotes himself to philosophy, by 
so doing undertakes to show his fellows the sublimest 
views at which man's understanding can arrive of 
the great problems which solicit the mind of the race. 
The definition of philosophy as " human wisdom " 
is still a superficial definition, and a nominal definition, 
which simply renders agreement possible as to the 
sense of the term. To attain a more profound defini-tion, 
a real definition which reveals the nature of the 
object, we shall study in the sequence of concrete 
history the formation or genesis of what men have 
agreed to call philosophy. 
In so doing we shall follow, so far as it is possible 
in an explanatory text-book, the actual method of 
Aristotle, too often neglected by books which teach 
his conclusions, but apparently ignore his spirit. 
That great reahst advanced nothing a priori and always 
studied the historical development of a problem before 
he proposed his own solution, which thus appeared 
as the natural goal of a process of discovery. Such a 
method will no doubt compel us to undertake a 
considerable digression into the field of history, but 
it is, nevertheless, in our opinion indispensable. 
On the one hand, from the practical and educational 
standpoint, an account of the historical origins of 
i8
INTRODUCTORY 
philosophic thought is the best method of acquainting 
beginners with the problems of philosophy, introducing 
them into the world, entirely new to them, of rational 
speculation, and furnishing them, incidentally, with 
much extremely useful knowledge. Their first requisite 
is to know what they are studying, and to possess a 
sufficiently live and accurate notion of the problems 
of philosophy presented in their simplest form. 
On the other hand, in justice to our subject itself, 
to state straight away, with no previous examination 
or concrete justification, conclusions relating to the 
nature of philosophy, its object, dignity, and so forth, 
would be to present the traditional conception of 
philosophy under an arbitrary and a priori aspect 
wholly alien to it, and to risk enslaving our pupils to 
empty formulae. By beginning, on the contrguy, with 
a brief outHne of the history of ancient philosophy up 
to Aristotle, that is to say until the conclusion of its 
formative period, we display philosophy in its origin 
and construction, and thereby show how the transition 
was effected between the teaching of common sense 
and the scientific knowledge of philosophers, how the 
great philosophic problems arose of themselves, and 
how a particular conception of philosophy, which will 
be put later to the test of discussion, results inevitably 
from this historical inquiry, and naturally forces itself 
upon the mind. We need not fear to insist upon these 
preliminary questions, which we shall have to consider 
again from another angle in criticism. They concern 
the very existence, the nature, and the value of 
philosophy. 
19
Introduction of philosophy
PART ONE 
THE NATURE OF PHILOSOPHY
Introduction of philosophy
PHILOSOPHIC THOUGHT BEFORE PHILOSOPHY IN 
THE STRICT SENSE 
Philosophic speculation, precisely because it is 
the supreme achievement of reason, is unknown to all 
the so-called primitive races. Indeed, even of the 
civilisations of antiquity the greater part either have 
possessed no philosophy or have failed to discover 
its true nature and distinctive character. In any 
case, philosophy only began to exist at a very 
late period about the eighth and especially the sixth 
century B.C., and then found the right path to truth 
by a success which must be regarded as extra-ordinary 
when we consider the multitude of wrong 
roads taken by so many philosophers and philo-sophic 
schools. 
Nevertheless, some of the most elementary truths 
with which philosophy deals were known long before 
philosophy itself had come to birth, and the more 
important of these are to be found in a more or less 
rudimentaiy form and more or less seriously cor-rupted 
among all the peoples of antiquity, even at 
the most remote epochs. But it was not from the 
philosophers that these peoples had learned them ; 
their knowledge was derived in part from that wholly 
spontaneous and instinctive exercise of reason which 
we call common sense, but above all from primitive 
tradition. 
23
AN INTRODUCTION TO PHILOSOPHY 
The most reliable inductions of history combine with 
the conclusions of theology ^ to prove the existence of 
a primitive tradition, common to the different branches 
of the human race and going back to the origin of 
mankind. And even in default of any positive sources 
of information, it is a very reasonable conjecture that 
the first man received from God knowledge together 
with existence, that by education he might complete 
the work of procreation. 
But was it possible that this knowledge, together 
with the primitive religion in which it was incorporated, 
could be transmitted in its integrity by the human 
race ? We have, on the one hand, truths of the 
loftiest sublimity to be handed down from one 
generation to another, yet, on the other, an intelligence 
dominated by the senses and imagination. A dis-proportion 
so extreme inevitably deteriorated the 
tradition received at the outset, as little by little the 
rust of obHvion gathered upon it, error defiled it, 
and it fell a prey to the corruptions of polytheism and 
the more degraded forms of religion (animism, 
totemism, idolatry, magic, etc.). Nevertheless, in 
spite of the changes which it underwent, the primitive 
tradition has preserved for mankind throughout the 
ages a deposit, progressively diminishing no doubt, of 
fundamental truths. In this deposit were included 
many philosophic conceptions—that is to say, concep-tions 
which concerned the most sublime problems 
within the scope of reason. But since they were 
taught only by a religious tradition which corroborated 
the instinctive teachings of common sense, they were 
1 p. Lemonnyer, O.P, (following Schmidt), La Rivilaiion primitive 
et les doTtnSes actuelles de la science. Paris, 191 4. 
24
PRE-PHILOSOPHIG THOUGHT 
; 
known in a pre-philosophic fashion and existed in a 
pre-philosophic state. 
It is not surprising that all peoples in the primitive^ 
stage of history were ignorant of philosophic speculation. 
But it is more astonishing that even certain civilisations 
were devoid of philosophy—for example, the Semitic, 
and the Egyptian, which is, in this respect, in the same 
category as the Semitic. Despite the high level of 
scientific culture reached by the intellectual aristocracy 
of these races, the sole philosophic conceptions, it would 
seem, which the Egyptians and Chaldeans possessed 
were a few very general ideas, implicit in their religion, 
concerning the Deity, the human soul and its state 
after death, and the precepts of morality. These 
truths, which, moreover (as in the case of every race), 
are purer the further back we follow their history, were 
never made the subject of rational study and specula-tion, 
but were simply accepted, as also were their 
scientific beliefs, as part of a sacred tradition. 
Religion took the place of philosophy, and from 
religion these races received certain philosophic truths 
philosophy they had none. In this matter the Jews 
did not differ from their fellow Semites. Scornful 
of human wisdom and the achievements of pure 
reason, and, indeed, without aptitude for such 
investigations, they produced no philosophers (at 
least not before Philo, who was a contemporary of 
* Primitive in respect of a particular brainch of the great human 
tree and so far as our knowledge of the past extends, but not primitive 
in the absolute sense. Far behind what we term the primitive state 
of the peoples known to us lies a long stretch of human history of which 
we know nothing. 
25
AN INTRODUCTION TO PHILOSOPHY 
Jesus Christ), but they possessed the prophets and the 
Law. 
All the great Indo-European civilisations, on the other 
hand, manifest an impulse, which no doubt took 
widely different forms, towards rational and, in the 
strict sense, philosophic speculation. But, except in 
Greece (and to a very partial extent in India), this 
impulse nowhere succeeded in achieving an indepen-dent 
scientific discipline distinct from religion. A 
traditional religion did not in this case take the place 
of philosophy, but philosophy, or, we should rather 
say, human wisdom, penetrated religion and was 
confused with it. The wise man fulfilled a sacred 
function. He was not the head of a philosophic 
school, but the founder of a religious sect, if not of a 
new religion. 
{a) Among the Persians,'^ whose original religion, so 
far as we know it from inscriptions, was a fairly pure 
monotheism, Zoroaster or Zarathustra founded Mazdeism 
or Zoroastrianism (about the eighth or sixth century 
B.C. ?), a powerful achievement of speculation which 
systematised (and incidentally distorted) certain funda-mental 
truths derived from the primitive tradition, 
* In this summary review of the great Aryan religions, we have been 
obliged not only to isolate by a process of abstraction the intellectual 
aspects of those religions with which the philosopher is concerned, 
but, moreover, to simplify considerably and reduce to an artificial 
classification doctrines whose vast and fluctuating complexity (this 
is especially true of Brahmanism and Buddhism), and occasional 
inconsistency, dismay historians. It should be added that the explana-tions 
of Oriental thought given by scholars are still largely conjectural 
and, in all probability, especially as far as philosophy is concerned, 
in many cases extremely inadequate. 
26
PRE-PHILOSOPHIG THOUGHT 
in the attempt to give a rational explanation of the 
vast problem which has faced human thought from 
the outset, the problem of evil. By his failure to 
perceive that God is the sole supreme principle and 
the source of everything which exists, so far as it 
partakes of being, and that evil is mere privation of 
being without positive existence, and therefore that 
no creature is evil by nature, Zoroaster ended in 
dualism and taught the existence of two principles 
imcreated and co-eternal, the principle of Good 
{Ormuzd) and the principle of Evil (Ahriman), who 
share the dominion of the universe and whose un-relenting 
struggle constitutes its history. So far as 
Ahriman is to be identified with the rebel angel of 
primitive tradition, Zoroastrianism tended to make 
the devil a god striving against God. 
(b) Among the peoples of India, whose intellectual 
and religious history is far more complex (since in 
this field no certainty has yet been reached, we 
present the interpretation of their beliefs which seems 
to us most probable), we witness a remarkable pheno-menon. 
When the original religion—the primitive 
reUgion of the Vedas ^—no longer proved sufficient to 
satisfy the intellectual demands or social needs of a 
more advanced civilisation, philosophic notions, which 
seem to have originated as interpretations of sacrifice 
and other sacred ritual, but developed in a spirit 
hostile to the ancient traditions and the cult of the 
^ The most ancient among the religious books of the Hindus ( Veda 
means knowledge), the Rig-Veda, is apparently not older than the 
twelfth century b.c. Vedic religion seems to have been an incoherent 
polytheism coloured by a vague pantheism. 
27
AN INTRODUCTION TO PHILOSOPHY 
gods, found a home among the sacerdotal caste and 
took possession of the priesthood. To reconcile the 
perpetuation of their office with their new opinions, 
the priests, while continuing to perform the traditional 
ceremonies, directed their worship no longer to the old 
gods, but to the undefined and secret forces of the 
universe. 
(i) This resulted, after a period of confusion, in 
the formation of a new system, Brahmanism (or Hindu-ism), 
which is essentially a philosophy, a metaphysic, 
a work of human speculation, but being, so to speak, 
clothed in the ornaments of the sanctuary, was invested 
from the outset with the sanctions and attributes of a 
religion. A divine origin was ascribed to the books 
in which it was taught (the Brahmanas and Upamshads)^ 
and they could be obtained only from the priests. 
Hence Brahmanism may be called a sacred, hieratic 
or theological metaphysic, and already in the eighth 
century B.C. the supremacy of the priestly caste among 
the Hindus seems to have realised in its fashion that 
social and spiritual sovereignty of the philosopher-priest 
and the religion of science which was the dream 
of certain nineteenth-century thinkers. 
It is true that the science which those thinkers wished 
to invest with a sacred character was the science of 
phenomena, or, as it was termed, positive science, which 
is not wisdom, even human, and, as Auguste Gomte 
justly observed, is incapable of producing order in 
any department. The human science to which 
Brahmanism gave a divine character was, on the 
contrary, the science of ultimate realities, metaphysics, 
human wisdom in the strict sense : a powerful effort 
of metaphysical thought (so far as we can judge of it 
28
PRE-PHILOSOPHIG THOUGHT 
from documents whose interpretation is still far from 
certain), but the product of reason still untrained, 
incapable of maidng the necessary distinctions and of 
avoiding internal contradictions, seduced by the 
dream of an intuitive knowledge of the All, angelic 
rather than human, and doomed by its very ambition. 
This system, at least when we consider its pre-dominant 
tendencies, taught that the First Principle 
of the world, named Brahma ^ or Atman,^ constitutes 
in himself the intimate reality of everything which 
truly exists, whence logically follows pantheism, or the 
identification of God with his creation.^ Nevertheless, 
1 From the name of the occult and sacred force which gave ritual 
its efficacy and pervaded all things. Originally regarded as the first 
emanation of the supreme God, it became for the Brahmans the 
unique source of being. The masculine noun, Brahma, designates 
the First Principle as God and Lord, the neuter, Brahman, as the one 
imp>er5onal substance. 
2 From the name of the principle of life (the " self " transcending 
the phenomenal individual), which was regarded as animating man 
and the imiverse. 
' The term pantheism is relatively recent, having been introduced 
into the vocabulary of philosophy by Toland in the eighteenth centiiry. 
But the doctrine it designates is as ancient as the earUest philosophical 
errors. 
For a system to be pantheistic, it need not explicitly identify God 
and creatures (very few pantheists fulfil this condition). It is sufficient 
that its teachings are logically ineconcilable with an absolute distinction 
between God and creatures. 
This observation is particularly important for the study of Oriental 
philosophies, of which pantheism is the original sin. Indeed, it arises 
in their case from the very method of thought they employ, which 
appears to consist primarily in the treatment of analogous concepts 
(realised differently in different objects) as though they existed as 
such outside the mind, which led them to conclude that things which re-main 
the same become on different planes of reality essentially different. 
For example, Atman is both the supreme principle of the universe, 
transcending all multipUcity, and the principle which distingiiishes 
and constitutes every personality. Like the Schoolmen, but for 
29
AN INTRODUCTION TO PHILOSOPHY 
an attempt was made to avoid this conclusion. The 
Supreme Principle, which possesses neither personality 
nor knowledge, to which no attribute can be applied, 
which is absolutely unknowable by any concept, 
however universal, not even by the concept of being, 
so that it must be called Nothing or Non-Being, is 
the sole true reality. Therefore the existence of 
everything multiple or limited, everything we can 
know by our senses or even by our concepts, is as 
such illusion, mere appearance. This is idealism, the 
denial of the reality of the world and of things. But 
the bare existence of this appearance or illusion is an 
evil, indeed evil pure and simple. The existence of 
individual objects and of this cosmic delusion which 
is called Nature (Maya), and which keeps us captives 
of the manifold and the transitory, is essentially evil 
and the source of all suffering. 
The problem of evil, therefore, seems to dominate 
the entire speculation of the Indian metaphysicians, 
as also of the Persian sages. But the Persians, whose 
bent was practical, always considered evil under the 
aspect ofsin, and, obsessed with the differences between 
moral good and evil, which they attempted to use 
as a criterion to divide beings into two metaphysical 
categories, ended in dualism. The Hindus, on the 
different reasons, the Indians distinguish between the personality 
(which is for us the spiritual subsistence of the soul) and the material 
individuality (which arises from the dispositions of the body). 
This mode of thought, which we meet again more or less emphasized 
in every doctrine of theosophic orientation, makes it possible to avoid 
the appearance of pantheism, because its inherent self-contradiction 
permits the affirmation of essential differences between terms which 
should logically be identified. But, precisely because these affirmations 
are only possible in virtue of a fundamental self-contradiction, it 
inevitably involves a real pantheism. 
30
PRE -PHILOSOPHIC THOUGHT 
contrary, exclusively occupied with contemplation, 
regarded evil pre-eminently under the aspect of 
suffering, or rather privation, in the sense in which 
metaphysicians understand the term.^ Led astray 
by a profound realisation of a great truth which they 
were unable to apprehend clearly (for while it is 
very true that it were better for us not to exist than 
to exist without being imited to God, they believed 
it were better for all things not to exist than to exist 
without being God), they ended in a pessimism 
which, though undoubtedly very different from the 
romantic pessimism of a Schopenhauer, was primarily 
the barren renunciation of a proud intellect, and 
attempted to be self-sufl5cient. 
What, then, in their conception did wisdom teach 
man ? It taught him to free himself from suffering 
and illusion, and with that object to rid himself of all 
individual existence. The Brahmans held the doctrine 
of the transmigration of souls, or metempsychosis ; they 
believed that souls, on the death of the organism which 
1 From this point of view Indian speculation may be said to afford 
a prominent example of pure metaphysical intellectualism. Regarding 
things solely from the standpoint of intellectual speculation and the 
universal order, and not from the standpoint of the rectitude of the 
human will and that particular order by which man is ordered to his 
last end, it quickly came to lose sight almost entirely of the notion of 
moral good and evil, and its ethics consists primarily in a metaphysical 
purification, directed exclusively to a particular ideal of intellectual 
knowledge. 
An analogous tendency is observable in every system which confuses 
by an exaggerated intellectualism the moral with the metaphysical 
order (a confusion which is glaring in Spinoza's Ethics, for example) 
and, failing to recognise that God is not only the provisor universalis 
of creation, but also the provisor particularis of the moral life (cf. St. 
Thomas, Sum. TheoL i, q. 103, a. 8, with Cajetan's Commentaty) , ends by 
claiming to transcend the distinction of good and evil and denying 
the existence of moral evil. 
31
AN INTRODUCTION TO PHILOSOPHY 
; 
they had animated, passed into another organism, 
thus living successively in different bodies of men, 
animals, or plants.^ The punishment of the wicked 
and foolish consisted accordingly in continuing to 
undergo in a series of reincarnations the pain of 
individual existence. The soul of the wise man, on 
the contrary, was delivered from the yoke of transmi-gration 
; absorbed or reabsorbed in Atman, it escaped 
the sufferings of the world by losing all distinctive 
individuality. 
The ethics of Brahmanism teaches the means 
whereby this deliverance can be achieved ; the 
wise man progresses towards that goal in this life by 
means of contemplation. Brahmanism understands 
that contemplation is a beginning of beatitude in 
this life ; but, as it mistakes the nature of beatitude, so 
it mistakes the nature of contemplation. The con-templation 
which it claims to teach is, in fact, only 
a metaphysical contemplation, or rather a species of 
supra-rational vision, which it expects to achieve by 
the merely natural powers of the created intellect 
unlike Christian contemplation, it is the product of 
the intellect alone, not of supernatural charity and the 
1 S05 at least, metempsychosis is currently understood. It is not 
unlikely that this interpretation of the doctrine is the popular translation 
of a doctrine less crude, according to which every being passes through 
an indefinite series of states or cycles of existence, each of which is only 
lived once, and our earthly existence is simply one particular state 
among many others. If this be the case, the doctrine of successive 
reincarnations originated in an unintelligent distortion of this theory, 
still further corrupted when it was introduced into the West. (The 
possibility, however, remains that originally the Pythagoreans and 
Orphics understood the transmigration of souls in a symbolic sense.) 
It is also possible, on the contrary, that the theory in question was 
a learned interpretation, elaborated by the Indian metaphysicians, of 
a popular belief in transmigration. 
32
PRE - P HI LO S O P HIC THOUGHT 
infused wisdom which accompanies it. Its aim is 
union with God by knowledge, not by love. Instead 
of admitting an activity overflowing from its own 
superabundance, it withdraws from activity of any 
kind, which it abandons wholly to the inferior powers. 
By this metaphysical contemplation, Brahmanism 
proposes to put us gradually in possession of our last 
end and initiate us into the blessed state of the 
delivered. Since it thus strives to reach by man's 
unaided powers heights which grace alone can attain, 
it results in a pseudo-mysticism of a purely intellectual 
character (in contrast to other, purely emotional, forms 
of false mysticism) in which the wise man, hoping 
not only to be united with God, but to blend with him, 
intoxicates himself not with God, but with his own self-annihilation. 
Hence (apart from those instances of 
genuine spirituality which grace is always free to 
produce) a host of counterfeits of supernatural 
mysticism, also of ascetic exercises and methods, 
including among their baser forms (with the fakirs) 
those tours de force of exaggerated asceticism which 
prove that the mortification of the flesh, when not 
regulated by reason and dictated by love, can be as 
fallacious as pleasure. Naturalism is thus the final 
characteristic and the capital vice of Brahmanism,^ as 
indeed of philosophic mysticism in general, whether 
it be the product of Brahmanism, Buddhism, neo- 
Platonism, or Islam. 
(ii) From the sixth century onwards new schools 
^ We do not mean that Brahmanism descends to the adoration of 
sensible nature, above which, on the contrary it claims to rise completely. 
By the term " naturalism " we here mean the claim to arrive at union 
with God and perfection without the supernatural assistance of grace. 
33
AN INTRODUCTION TO PHILOSOPHY 
— 
. 
arose in India, some orthodox, others heterodox. Of 
these the principal was that founded by Cakya-Muni, 
surnamed the Buddha ^ (the enlightened, the sage) 
Buddhism, a doctrine essentially negative and solvent, 
directed, moreover, to practice rather than to specu-lation, 
may be regarded as the corruption and dis-solution 
of the Brahman philosophy. 
Substituting for that which is that which passes away, 
refusing to say that anything does or does not exist, 
and admitting only a succession of impermanent 
forms without fixed foundation or absolute principle 
in other words subordinating being to what is known 
as becoming or feri—^it showed, at the very time at 
which in Greece Heraclitus formulated the philosophy 
of flux, all the characteristics of a perfect evolutionary 
system, and, if it declared the existence of God, as 
of a substantial self and an immortal soul, unknow-able 
{agnosticism), its real tendency was to deny the 
existence of God (atheism), and to substitute for 
substance of any kind a stream or flux, regarded 
indeed ^ as itself real, of forms or phenomena 
(phenomenalism).^ Hence for Buddhism metempsy-chosis 
consists in a continuous chain of thoughts and 
feeHngs (a stream of consciousness, as we should term 
it to-day) passing from one mode of existence to 
another in virtue of a sort of urge towards Ufe, due 
itself to the desire to live : it is desire which is the 
" His actual name was Gautama. The name Cakya-Muni means 
the ascetic or hermit {muni) of the race or clan of the Cakya. Buddha 
lived during the second half of the sixth century B.C. He would seem 
to have died about the year 477. 
2 At least by Buddha's original disciples. 
' " Ever>'thing is empty, everything vmsubstantial " was a saying 
of Buddha's. 
34
PRE-PHILOSOPHIC THOUGHT 
cause of existence and " we are what we have 
thought." 
At the same time, the teaching of deliverance from 
suffering, which in Buddhism, even more than in 
Brahmanism, dominates the entire system, assumes a 
different and even more radical form. Evil is no 
longer merely the possession of individual or personal 
existence ; it is existence itself : it is evil to be, and the 
desire of existence is the root of all suffering. The 
wise man must therefore destroy in himself man's 
natural longing for existence and for beatitude, the 
fullness of being ; he must abandon all hope and 
extinguish every desire. He will thus attain the state 
of emptiness or toial indetermination called nirvana 
(literally nakedness, metaphorically immortality^ refresh-ment, 
thefarther bank—the term, in itself indefinite, was 
never defined by Buddha), which will deliver him 
from tlie evil of existence and the yoke of trans-migration, 
and which, in the logical consequence of 
Buddhist principles, must be regarded as the annihila-tion 
of the soul itself. For since the soul is only the 
chain or current of thoughts and feelings which derive 
their existence from the desire to be, to extinguish that 
desire is to extinguish the soul. 
This nirvana is the goal for whose attainment 
Buddhism made use of the ascetic practices which it 
took over with considerable mitigation from Brahman-ism, 
also of its moral code ^ which is thus directed, 
not to God, but to a species of mystical nothingness 
* We here tmderstand moral code in a very wide sense as meaning 
a code of behaviour. If the expression be taken as implying moral 
obligation, whose ultimate basis is the Christian doctrine of God 
the transcendent Creator, we must conclude that Buddhism, as indeed 
all the Oriental religions, Indian or Chinese, has no moral code. 
35
AN INTRODUCTION TO PHILOSOPHY 
as its last end. Moreover, the source and ultimate 
measure of Buddhist ethics is man, not God. If it 
rejected the system of castes which exaggerated 
the demands of social order and divided man almost 
into distinct species, it was only to dissolve social 
order of any kind in an absolute equality and 
individualism. And though it prescribed a universal 
benevolence (which extended even to prohibiting 
the slaughter of animals and to a compulsory 
vegetarianism), almsgiving, pardon of injuries, and 
non-resistance to the wicked, its motive was not love 
of one's neighbour as such, whose positive good and 
(by imphcation) existence we are bound to will, but 
to escape suffering to oneself by extinguishing all 
action and energy in a kind of humanitarian ecstasy. 
Buddhism is, therefore, a proof that gentleness and pity, 
when they are not regulated by reason and dictated 
by love, can deform human nature as much as 
violence, since they are then manifestations of 
cowardice, not of charity. 
This doctrine of despair is not only a heresy from 
the point of view of Brahmanism ; it is an intellectual 
plague to humanity, because it proceeds from the 
negation of reason. It is not, therefore, surprising 
that we find in it the majority of the fundamental 
errors by which contemporary attacks on reason are 
inspired. If at the present day it has found a warm 
welcome among certain circles in Europe, it is because 
all those who hope to derive from humanitarianism a 
moral code of human kindness for the acceptance of an 
atheistic society are already implicitly Buddhists. 
(iii) Buddhism is a philosophy, agnostic and 
atheistic, which nevertheless usurps the social and 
36
PRE-PHILOSOPHIG THOUGHT 
ritual functions of a religion. It is as a religion that 
it has won the allegiance of so many millions.^ In 
certain other schools to which Brahmanism gave 
birth—schools recognised as " orthodox "—we find, 
on the other hand, a tendency towards the normal 
distinction between philosophy and religion. 
These darshanas, it is true, would seem to be not so 
much distinct systems as complementary aspects of 
one and the same doctrine, the Brahmanist meta-physics. 
Here we may pass over the Vedanta, the 
most complete statement of that metaphysics and its 
doctrine of deliverance ; the Mimamsa, a species of 
commentary on the ritual and an explanation of the 
unseen forces set in motion by every act ; the Sankhya, 
founded, it is said, by Kapila (fifth or sixth century 
B,c. ?), which treats of the emanation of all things from 
their source, and seems to have taught, like Plato, a 
psychological dualism which explziins suffering by 
the union souls contract with matter ; and also Toga^ 
which teaches the practical methods which lead to 
contemplation, that is to say, the total loss of conscious-ness 
and identification with the universal Being 
(Iskvara) by a supra-rational knowledge. But the 
darshana Vaisesika, ascribed to Kanada (about the fourth 
century B.C. ?), which includes a rough outline of 
cosmology, and divides everything which exists into a 
number of fundamental classes or categories, sub-stance, 
quality, movement, association, difference, and 
^ However, in proportion as it has secured wide acceptance, Buddhism 
has ceased to be atheistic, only to fall into the most degraded concep-tions 
of deity. Popular Buddhism as practised to-day in many parts 
of Asia, where, to adapt itself to existing beliefs, it has assumed the 
most varied shapes, is nothing more than a form of idolatry, totally 
different from philosophic Buddhism. 
37
AN INTRODUCTION TO PHILOSOPHY 
— 
inherence, and explains the four elements of ponder-able 
matter, earth, water, air, and fire, by the union 
of indivisible and indestructible particles, " atoms " in 
the language of philosophy,^ and the darshana Nyaya^ 
founded by Gotama, which attempts to construct a 
theory of reasoning and proof—that is to say a logic, 
though a logic extremely confused and incomplete 
are clearly the rough sketches of a work strictly and 
solely philosophical. But these crude attempts did 
not lead to a completed system, and Indian thought 
never achieved a rational and autonomous philosophy. 
{c) When we turn to the Far East and consider the 
very ancient civilisation of China^^ we find that when 
the primitive religion of the Chinese, which seems to 
have been fairly pure,' had from the twelfth century b.c. 
undergone gross corruption and materialisation, 
substituting the sky for God,* worshipping the sun and 
moon, paying divine worship to the souls of ancestors 
and to spirits, and allowing itself to become tainted 
by magic and sorcery, wise men were compelled here 
1 Kanada, however, to explain this union, attributed real qualities 
to his atoms. Observe that Brahmani^m, which rejects atomism, 
admits five elements (ether being the fifth) ; Buddhism on the contrary, 
which has welcomed atomism, only four. 
2 Whatever be the racial appurtenance of the Chinese, their history 
undoubtedly shows closer connections with the Aryans than with the 
Semites. It is for this reason that Chinese philosophy is discussed 
in the present section. 
* It taught the existence of one sole God 
— 
Shang-ti—personal, 
intelligent, distinct from the world, Sovereign Ruler of the races of 
mankind ; also the immateriality and immortality of the human soul, 
and even offered to the spirits of ancestors the same sacrifices and 
marks of reverence as to the good spirits who are guardians of men. 
* In all probability Heaven {Tien) was in origin simply a meta-phorical 
synonym of the Sovereign Ruler {Skang-H). 
38
PRE-PHILOSOPHIC THOUGHT 
also to seek a remedy for a decadence which about 
the sixth century B.C. threatened their civiHsation with 
utter ruin. 
It has long been beUeved that the Chinese sages 
were simply moralists wholly occupied with laying 
down rules of conduct and completely indifferent to 
metaphysical speculation. This is a true account 
only of Confucius and his followers ; it does not seem 
to be appHcable to Lao-Tse, though we can only 
accept with considerable reserve the interpretations 
of his teaching offered by certain modern Taoists. 
According to their account, Lao-Tse (born 604 b.c) 
was himself the disciple of a tradition whose oldest 
monument is the Ti-King, a book which consists 
essentially of sixty-four graphic symbols (hexagrams or 
double trigrams) arranged in a series of mechanical 
groups/ formed by combining simpler signs and 
susceptible of very many interpretations (metaphysical, 
logical, mathematical, moral, political, astronomic) , each 
number corresponding analogically with the others. 
The metaphysical speculation of the Ti-King appears 
to have been primarily concerned with the question, 
How can the Absolute, being wholly self-sufficient, act 
and manifest itself? It distinguishes in the supreme 
and sole First Principle or Perfection two different 
aspects, Chien^ the unmovirig and unknowable source 
of all activity, and Chuen, knowable activity, which 
eternally manifests perfection in a process of spiral 
evolution and an endless flux of forms. But these 
two aspects merge in one single and self-identical 
being, and all things, after passing through all the forms 
1 Raymond Lull, in his attempts to create an ideographic algebra, 
employed an analogous procedure. 
39
AN INTRODUCTION TO PHILOSOPHY 
of evolution (of which the human cycle is but one 
curve), must return to Chien. This metaphysic may 
therefore be described as a species of evolutionary 
pantheism. It constitutes the foundation of Lao-Tse's 
system (Taoism), his chief contribution being an 
element of occultism and asceticism,^ Tao (the Way), 
the eternal goal and process of evolution, is the road 
by which all things must pass to arrive finally at the 
complete cessation of activity [Nibban, the Chinese 
nirvana), in which they are reabsorbed in nothingness 
and become one with the first principle of ail activity. 
The wise man will imitate the Tao by cutting himself 
off from all things, for the Way, though it has pro-duced 
beings, does not partake of their movements. 
" Having built this house, it dwelleth not therein." 
Detached from wealth, passions, and sensible experi-ence, 
and knowing that evil is mere appearance, he 
trains himself in solitude, secrecy, and humility (a 
humility which has nothing in common with the 
Christian virtue of that name, being nothing more 
than prudence and contempt for one's fellow men), 
until he reaches a state of perfect knowledge in which 
he no longer acts except by the pure intelligence. 
The wisdom, the illusory wisdom, to which Taoist 
asceticism leads its disciples, an asceticism which 
makes use of opium, as Buddhist asceticism of hypnosis, 
is for man a principle of revolt, therefore the adept 
1 It may be added that in the twelfth century a.d. Chu-Hi, who 
has been regarded, mistakenly it would seem, as a materialist, 
formulated, in the tradition of Lao-Tse, a system which in the Chinese 
system of education has become, practically speaking, the official 
philosophy. It explains the constitution of things by a dual principle 
{li and ki) which is not without resemblance to the duality of form 
and matter in Aristotle and the Alexandrians. 
40
PRE -PHILOSOPHIC THOUGHT 
must keep it a secret for himself and a narrow circle 
of initiates.* 
No doubt Confucius (Kung-fu-tse, 551-479 B.C.), 
who, unlike Lao-Tse, represents for the Chinese a 
moderate and practical wisdom (which, from its place 
in their system of education and its own active 
character, is generally accessible), preserved many 
truths of the primitive wisdom. He avoided, however, 
every ultimate question, and confined himself to an 
ethic purely human, social, earthly, and even 
commonplace. Opportunism, he observes, is the 
distinctive mark of the wise man. Every predeter-mined 
line of action, every preconception is a mistake. 
In all matters one should pursue a middle course, 
live unfettered by one fixed purpose, embrace no 
opinion with enthusiasm, reject nothing because it is 
antipathetic, do whatever seems best in the circum-stances 
of the moment and as the situation demands. 
Confucianism, a system intended for the multitude, 
ended in pure materialism. Taoism, which claimed 
to address a small circle, and which, if the interpreta-tion 
given above be correct, constitutes, together with 
Brahmanism, one of the most singular attempts ever 
made by man to attain, in that ignorance of love 
which seems an aboriginal characteristic of Oriental 
thought, a wisdom exclusively of the intellect, by 
which he could deify himself in metaphysics, has 
experienced in China alternate periods of triumph 
and persecution, and has organised, apparently ever 
since the opening centuries of our era, secret societies 
* " Empty their heads, and fill their bellies," was Lao-Tse's advice 
to a statesman ; 
" weaken their minds and strengthen their sinews. 
To teach the people is to ruin the State." 
41
AN INTRODUCTION TO PHILOSOPHY 
in which it has definitively taken refiige since the 
seventeenth century and in which it has degenerated 
into a philosophic and political occultism of the most 
pernicious type. 
This brief historical sketch has shown the important 
part in the hfe of humanity played by sages and their 
wisdom. All these nations, situated on the frontiers 
of darkness, and lacking a divine revelation of truth, 
were obliged, when their religions proved incapable 
of satisfying the needs of the individual soul or of 
society, to have recourse to the wisdom supplied by 
human reason. This wisdom, in the civilisations of 
which we have spoken hitherto, was never differentiated 
from religion, but, on the contrary, encroached on 
the domain of the latter and claimed to conduct men 
to their last end, until in India we actually find that 
Brahmanism successfully achieved that canonisation 
of metaphysics which threatened the Greco-Roman 
world in the reign of the neo-Platonic emperor, 
Julian the Apostate. It achieved also that transfusion 
into religion of a human philosophy attempted by 
Kantian metaphysics in the nineteenth century 
{Modernism) . 
It has also shown how this human wisdom has 
everywhere proved bankrupt, and how, even before 
philosophy took shape as an independent discipline, 
most of the great philosophic errors had been already 
formulated. From the very first, the most arduous 
problems tower hke mountains before the intellect 
of man ; the problem of evil, the problem of being, 
the problem of the becoming and flux of things. 
It is not, therefore, surprising that a reason liable to 
42
PRE-PHILOSOPHIG THOUGHT 
error the moment it transcended the elementary 
truths within the range of common sense, a reason 
still unstable and undisciplined, and therefore all the 
more ambitious, went astray from the outset and 
opened the history of metaphysics with the dualism 
of Zoroaster and the pessimism of the Hindu, the 
pantheism and idealism of the Brahmans, the atheistic 
evolutionism of Buddha, and the illusory wisdom of 
Lao-Tse. When it became more modest, it was only 
to fall into the ethical positivism of Confucius, re-nouncing 
all sublimity and even denying its own 
raison d'etre. Nor should it surprise us to find the 
same errors reappearing at a later stage, v/hen philo-sophy 
had been fully elaborated. Error, at whatever 
period ofhuman history it may arise, is due to a failure 
of man's reasoning power—is, so to speak, a return of 
its primitive weakness, and therefore of its very nature 
retrograde. 
A further fact, however, calls for remark here, a 
fact only too well established by this prehistory, so 
to term it, of philosophy : namely, that these funda-mental 
errors are not unsubstantial and insignificant 
dangers ; they may succeed, to the bane of those 
diseased cultures which they condem.n to sterility. 
Truth (in all matters which transcend the data of 
common sense) is not, as those are apt to beheve who 
have had the good fortune to be bom into a culture 
formed by it, given to man ready made, like a natural 
endowment. It is difficult to attain, and hard to 
keep, and only by a fortunate exception is it possessed 
uncontaminated by error and in the totahty of its 
various complementary aspects. We have therefore 
the most urgent cause to be grateful for the possession 
43
AN INTRODUCTION TO PHILOSOPHY 
of a revelation, by which God has given us from on 
high, besides a knowledge of supernatural truth 
inaccessible to reason, a sure and easy access to the 
essential elements of the same truth which, so far as it 
falls within the natural order, is indeed accessible to 
our speculation, but can be so easily missed by it. 
Those also have the strongest claim on our gratitude 
who from below, by the strenuous exercise of their 
reason and unaided by revelation, succeeded in 
bringing to light the principles and laying the perma-nent 
foundation of this natural truth, and in con-structing 
a true and progressive human wisdom, in 
other words a philosophy, which, when met later 
and raised by the truth revealed from heaven, would 
be incorporated into the fabric of a higher wisdom, 
theology, the wisdom of man deified by grace, wisdom 
in the highest sense of the term. How highly there-fore 
ought we to prize the sacred heritage of Greek 
thought 
! 
In Greece^ alone in the ancient world, the wisdom of 
man found the right path, and as the result of a 
fortunate harmony of the soul's powers and of a long 
effort to achieve mental order and discipline human 
reason attained its full vigour and maturity. In con-sequence, 
the small Hellenic race appears among the 
great empires of the East like a man amidst gigantic 
children, and may be truly termed the organ of the 
reason and word of man as the Jewish people was the 
organ of the revelation and word of God. 
It was in Greece alone that philosophy achieved 
her autonomy and was explicitly distinguished from 
religion. At least during the purest and most glorious 
44
PR E-PHILOSOPHIG THOUGHT 
age of the Heilenic mind, it recognised its own 
boundaries and was content to claim a strictly limited 
territory—the scientific study of purely rational truths 
—whereas Greek religion, already very much degraded 
in the time of Homer, became increasingly incapable 
of satisfying the needs of the intelligence, and grew 
more corrupt every day. True, the time would come 
when the Greeks, arrogantly abusing philosophy and 
reason, would attempt to embrace the things of God 
within the limits of their wisdom, " would become 
vain in their thoughts " and deserve the condemnation 
pronounced by St. Paul on the wisdom of this world, 
" which is foolishness in the sight of God." But their 
philosophy, though born of their mind, is undefiled 
by their corruptions, and its sole object is the truth. 
45
II 
THE PRE-SOGRATIG PHILOSOPHERS 
The earliest thinkers of Helias were the poets, the 
interpreters of traditional religion. Myth-makers, like 
Hesiod or Homer, sometimes prophets, such as that 
Epimenides of Cnossos who purified Athens from 
pestilence by erecting altars to unnamed divinities, 
they have no place in the history of philosophy in the 
strict sense, Greek philosophy, as Aristotle shows, 
only began with Thales of Miletus, one of the Sages 
or Gnomics, who lived in the seventh or sixth cen-turies 
B.G.^ 
The primary aim of these Sages, traditionally seven 
in number (their names are variously handed down 
by ancient writers), was to improve the conduct of 
their fellow citizens. Their aphorisms, some of which 
Plato quotes in the Protagoras, do no more than embody 
the practical lessons they had learned from their 
experience of life. They were men of acdon, legisla-tors, 
or moralists, men of prudence, but not yet 
philosophers. Alone among them Thales embarked 
on scientific speculadon. Geometrician and astrono-mer, 
he demonstrated that ail the angles inscribed in 
a semicircle are right angles, and appears to have 
predicted—no doubt owing to his acquaintance with 
^ For the iragitnents from the early philosophers quoted in this 
chapter Professor Burnet's translation (Early Greek Philosophy) has been 
used I for Aristotle's Metaphysics Professor W. D. Ross's translation. 
46
. 
THE PRE-SOCRATIC PHILOSOPHERS 
Babylonian science—the total eclipse of the sun which 
occurred on May 28th, 585. 
The philosophers who succeeded him were still for 
the most part men who played an active part in public 
affairs, ardent politicians of the city state ; but, in 
spite of this practical activity, they were more or less 
clearly conscious from the beginning of the true nature 
of their wisdom. Moreover, save in the case of a few 
exceptional individuals (for instance, Empedocles, 
the miracle worker, and Pythagoras, who founded a 
religious sect) , Greek philosophy was from the very first 
distinct from religion—indeed it took shape as a critic 
and foe of the popular mythology and was manifestly 
the product of pure reasoning. 
In this work we are concerned only with the 
progressive development of Greek philosophy from 
Thales to Aristotle, for it was during this period that 
philosophy, with its absolute validity for mankind as 
a whole, took definite shape. The process occupied 
some three centuries and is divisible into three great 
epochs—the period of formation (the pre-Socratic 
philosophers), the period of crisis (the Sophists and 
Socrates), the period of fruitful maturity (Plato and 
Aristotle) 
THE lONIANS 
(a) Human reason now set out with its unaided 
powers in search of the principles and causes of things. 
What first strikes man's intelHgence is what he sees 
and touches, what he knows by his senses, and when he 
attempts to understand anything, he begins by asking 
what it is made of. Therefore the first thinkers of 
Hellas only considered in things the material of which 
47
AN INTRODUCTION TO PHILOSOPHY 
" 
they are made, their matter (what we shall learn to 
call the material cause), which they naively took to 
be a complete explanation of the object. Moreover, 
since the most universal and most important 
phenomenon of nature is change, especially the change 
by which one body becomes another {e.g. bread 
becomes flesh, wood fire), they concluded that the 
original matter of which all things are fashioned must 
be identical in all, the common subject of all corporeal 
changes. But since they were still unable to conceive 
any impalpable or invisible principle, they thought 
they had discovered this matter in some one of the , 
elements perceived by the senses. 
Thales, for example (624-546), influenced by 
traditional myths which derived all things from the 
primordial waters, and arguing from the fact that 
plants and animals " are nourished by moisture 
and that the germ of animal life is moist, concluded 
that water is the sole substance, preserving its identity 
through all the transformations of bodies. For 
Anaximenes (588-524) this substance was air, for 
Herachtus (540-475 ?) fire, for Anaximander (610- 
547) the boundless (by which he understood the 
indeterminate, arceipov), a fusion of all the contraries. 
Moreover, water, air, fire, and the boundless were 
regarded as something active, Uving, and animate, 
endowed by an internal force with a manifold and 
unlimited fecundity. This was the meaning of Thales's 
dictum, all things " are full of gods," Travxa nkripy] 
6eoiv.^ From the history of this extremely primitive 
Ionian school, whose philosophy is termed hylozoist 
because it ascribed life (Ccoy)) to matter (uXv)), we learn 
1 Aristotle, De Anima, i, 5, 41 1 a 7. 
48
THE PRE-SOCRATIC PHILOSOPHERS 
to regard as the most elementary and crude of 
philosophic doctrines the materialistic monism which 
teaches the existence of a one single substance of a 
material nature, and evolutionism^ which attempts to 
explain everything by an historic process of unfolding, 
development, or evolution of something pre-existent. 
Evolutionism, which, owing, on the one hand, to 
German metaphysics, on the other to Darwin and 
Spencer, became so popular in the nineteenth century, 
was already taught in Greece by the physicists of 
the sixth and fifth centuries b.c.^ Anaximander in 
particular taught the eternal evolution of worlds 
" which rise and set at long intervals," and held that 
animals sprang from the mud of the sea floor, clothed 
at first, as with a species of shell, in a prickly bark 
which they shed on dry land, ^ and that man arose from 
animals of another species,' having been originally 
formed within the bodies of fishes, where he developed, 
being ejected as soon as he had become sufficiently 
large to provide for himself.* 
Later Empedocles of Agrigentum (493-433 ?), whose 
speculation in other respects marks an advance on that 
of the lonians,^ explained the origin of Uving beings 
1 In India about the same date Buddhism was formulating, as we 
have seen, the religion of evolutionism. 
* Plac. Philos., V, 19, i. Dox. 430, 15. 
' Pseudo-Plut., Strom., frag. 2, Dox, 579, 17. 
* Plut., Sjmp., q, viii. 579, 17. 
" For a single corporeal substance Empedocles substituted four 
elements specifically different, the four which became later the four 
classical elements of ancient chemistry—earth, water, air, fire. His 
dominant interest was to discover the efficient cause of the evolution 
of things, which he believed to consist in the two great motive forces 
love and hate. Empedocles was not only a philosopher : he was also 
a magician, doctor, poet, orator, and statesman. Aristotle ascribes to 
him the invention of rhetoric. 
49
. 
AN INTRODUCTION TO PHILOSOPHY 
by the separate production of the individual organs 
and members, e.g. the head, eyes, arms, which were 
subsequently joined by chance in every possible 
combination, of which only those have survived which 
were fitted to Uve {cf. the Darwinian principle of 
the survival of the fittest) 
It is also worth remark that, before Democritus, 
Anaximander and Empedocles also sought, Uke the 
pseudo-scientific evolutionism of modem times, to 
explain all things mechanically, that is to say—as the 
result of a simple aggregation of material elements 
effected by local motion. 
(b) Among these physicists, as Aristotle termed them, 
or philosophers of sensible nature, must be reckoned 
three great thinkers, Heraclitus, Democritus, and 
Anaxagoras. 
(i) Heraclitus of Ephesus,^ a lonely and proud genius 
who despised the multitude and popular reHgion, 
drew heroically fi*om the thought of the Ionian 
philosophers its ultimate metaphysical presuppositions, 
and thereby fixed for aU succeeding ages one of 
the possible extremes of speculation and error. A 
particular reality perceived in things had taken hold 
of his intellect with such force that he became its 
hopeless slave. That reality was change or becoming. 
His vision was so fixed on the change which all things 
undergo that he declared that change alone is real. 
IlavTa pel, aU things are in flux ; and men are fools to 
trust in the stability of their false happiness, " when 
they are born, they wish to live and to meet their 
^ The dates of Heraclitiis's birth and death are uncertain. He was 
in his a-Kix-fif the prime of his age, about 500 B.C. 
50
THE PRE-SOGRATIC PHILOSOPHERS 
doom—or rather to rest—and they leave children 
behind them to meet their doom in turn." We do 
not touch the same thing twice nor bathe twice 
in the same river. The very moment we touch an 
object, it has already ceased to be what it wa5 before. 
Whatever exists changes from the very fact of its 
existence. 
That is to say change has no abiding and permanent 
subject identical with itself, Uke an ivory billiard ball 
which remains an ivory billiard ball while it is in 
motion. We are therefore compelled to pronounce 
boldly that that which is (the thing which changes) 
at the same time is not (because there is nothing which 
persists throughout the change). " We step and do 
not step into the same river ; wc are and are not." 
Moreover, contraries must be pronounced identical. 
The sea is the purest and the impurest water. Good 
and ill are one. *' No one," writes Aristotle in a 
famous passage, " can possibly conceive that the same 
thing does and does not exist. According to some, 
Heraclitus was of a different opinion, but we are not 
obliged to believe that a man really thinks whatever 
he says. The reason of the opinion held by these 
philosophers was that the only realities they admitted 
were sensible objects, and, since they perceived that 
sensible nature is in perpetual motion, some have held 
with Cratylus ^ that no statement can be made about 
it ; he was content to wag his finger." * This 
scepticism was the inevitable consequence of Heraclitus's 
philosophy of pure flux, despite his personal conviction, 
* One of Heraclitus's most famous disciples. He was Plato's first 
teacher. (Aristotle, Metaph., i, 6.) 
* Metaph., iv, 5, loio a 13. 
51
AN INTRODUCTION TO PHILOSOPHY 
passionately held, of the reality and value of truih. 
" If you do not expect the unexpected," he said, " you 
will not find truth, for it is hard to be sought out and 
difficult." 
HeracUtus is thus the philosopher of evolution and 
becoming. In his view, all things are diflbrentiations 
produced by discord or strife {7z6'kz[io<^ Tcar^p Tcavrcov) 
of a single mobile principle which he conceives in 
the form of fire, a fire ethereal, living, and divine. 
So from the outset stands out in the clearest light 
that fatal necessity which chains every philosophy 
of pure becoming to monism ' or to pantheism^ " If," 
wrote Aristotle,' '' you maintain that all beings are 
one, you simply return to Heraclitus's opinion. All 
things are then confused, good and evil become 
identical, man and the horse are one and the same 
thing. But this is really to maintain not that beings 
are one, but that they are nothing." 
(ii) Born within a few years of Heraclitus's death, 
Democritus of Abdera (470-361 ?), who had a more 
superficial intellect and a predilection for ideas easily 
comprehensible, attempted to discover in the flux of 
sensible phenomena a permanent and unchanging 
element ; but in his search for this unchanging 
element he made use of his imagination rather than 
his understanding. Therefore the sole reality he 
would recognise was something which, though it is 
inaccessible to the senses, can nevertheless be appre-hended 
by the imagination—namely, pure geometrical 
quantity as such, stripped of all qualities (colourless, 
^ The doctrine that ail things are one single being. 
' The doctrine which identifies the world with God. 
* Phys., i, 2, 185 b 19. 
52
THE PRE-SOCRATIC PHILOSOPHERS 
scentless, tasteless, etc.), and possessed solely ofextension 
in the three dimensions of space. Democritus found 
the explanation of everything in the plenum, which he 
identified with being, and the void, identified with 
nonentity. The plenum was divided into indivisible 
parts of extension (" atoms "), which were separated 
one from another by the void and in a state of ever-lasting 
motion, and differed only in shape,^ order,' 
and position.' The order of the universe and the 
structure of individual beings he attributed to the 
blind necessity of chance. Thus Democritus * intro-duced 
into Greek philosophy during the lifetime of 
Socrates the doctrine of atomism and more generally 
the philosophy termed mechanical, which raises 
geometry to the position of metaphysics, reduces 
everything to extension and motion, and professes to 
explain the organisation of the universe by a host of 
fortuitous coincidences. In this fashion the Parthenon 
could be " explained " as the result of throwing 
stones one on another during an indefinite term of 
^ As, for example, A differs from N. 
» As AN differs from NA. 
' As N differs from the same letter placed differently : Z. 
* As also his master Leucippus. Had Leucippus and Democritus 
come in anyway under the influence of the Indian philosopher Kanada? 
The more likely hypothesis is a coincidence due to similarity of intellec-tual 
oudook, particularly if Kanada, whose date is very uncertain, was 
contemporary with or even posterior to Democritus. Speaking 
generally, there seems no reason to believe that Oriental speculation 
so influenced Greek thought as to teach it in the strict sense or transmit 
any particular system. That, on the other hand, it influenced the 
Greeks by arousing a spirit of speculative inquiry and providing in-tellectual 
material (which they alone were able to treat scientifically) 
is the obvious conclusion from the simple fact that Greek philosophy 
originated in those provinces of the Hellenic world which were in 
contact with the Ea^t. 
53
AN INTRODUCTION TO PHILOSOPHY 
years, the tragedies of Racine as due to the in-discriminate 
shuffling of type for a sufficient length 
of time. 
(iii) Finally Anaxagoras of Glazomenae (500-428), 
who was a friend of Pericles, and in the maturity 
of his powers when Democritus was born and 
HeracUtus had just died, turned Greek philosophy 
towards a higher source of illumination, and cor-rected 
rather than continued the speculation of the 
lonians by the aid of ideas which, it must be con-fessed, 
he either worked out badly or was unable 
to use. 
On the one hand, he perceived that the material 
principle ofwhich all bodies are formed, and which the 
lonians had identified with one particular element, 
must already somehow contain in itself the entire 
diversity to which it will give birth : unless everything 
were in everything, nothing could come from nothing.^ 
He therefore concluded that the principle in question 
consisted of an endless mixture of all natures and 
quahties in such fashion that each corporeal particle 
contained within itself the elements {homoeomeries) 
of all the rest ; for example, each particle of the bread 
we eat contains invisible elements of the bone, blood, 
flesh, etc., which will be discovered later, changed 
only in their relative proportions, in each particle of 
bone, blood, flesh, etc. It was a bizarre conception, 
and, as taught by Anaxagorsis, not worth serious 
discussion, but nevertheless a crude adumbration of 
Aristotle's great conception of first matter {materia 
prima) which is nothing in act, but aU bodies in 
potentiality. 
1 Cf. Aristotle, Phys., i, 4, 187 a 26. Simpliciiu, Phys,, 155, 23. 
54
— 
THE PRE-SOCRATIG PHILOSOPHERS 
On the other hand—and it is his chief claim to 
distinction—he realised that the material principle, 
that of which all things are made, is insufficient to 
explain them. We must also discover the agent 
that produces them (the efficient or motor cause) and 
the end for which the agent acts (the final cause). 
Is it, as Plato was to ask later, a sufficient explana-tion 
of the fact that Socrates is sitting in prison 
to say that he has bone, joints, and muscles arranged 
in a particular fashion? We must also know who 
brought about that disposition of these bones and 
muscles—namely, Socrates himself by his will—and 
why he willed it. 
Because Anaxagoras arrived at the recognition that 
there must necessarily exist, besides the material 
elements of the world, a separate Intelligence (voui;') 
to which the ordering of the universe is due, he alone, 
as Aristotle remarks " kept sober " when all the other 
philosophers of his period, drunk with the wine of 
sensible appearances, " spoke at random." ^ 
THE ITALIANS 
Besides the school of Ionian philosophy, there 
existed in the Hellenic world of the sixth and fifth 
centuries B.C. two other great philosophic schools 
the Pythagorean and the Eleatic. 
Pythagoras of Samos (572-500 ; according to other 
authorities, 582-497), the founder of a philosophic 
society of a religious and political character, which 
held the reins of government in several cities of Magna 
Graecia (Southern Italy), and was later dissolved by 
1 Metaph., i, 3, 984 b 18. 
55
AN INTRODUCTION TO PHILOSOPHY 
violence,^ understood that there existed realities of a 
higher order than those perceptible by the senses. 
But it was by the study of numbers that he had arrived 
at the knowledge of these invisible reaUties, whose 
immutable order dominated and determined the 
process of becoming ; and henceforward he had 
understanding only for numbers. Not content with 
teaching that there is present in all objects and in the 
universe as a whole a hidden principle of measure and 
harmony, he taught that numbers—by which this 
harmony is revealed to our senses—are the sole true 
reality, and regarded them as the very essence of 
things. Pythagoras was not only conversant with the 
important observations of Oriental astronomy, but, 
by his fundamental discovery of the relationship 
between the pitch of sounds and the length of vibrating 
strings, had reduced to the rigidity of numerical law 
so fugitive a phenomenon as sound. Imagine the 
awed astonishment with which he must have dis-covered 
behind the flux of sensible phenomena the 
intelligible constant and immaterial proportions which 
explain to the mathematician the regularities we 
observe. Consider, moreover, the mysterious symbolic 
value of numbers attested aUke by the sacred traditions 
of mankind and the most positive of philosophers (from 
Aristotle, who paid homage to the holiness of the 
number 3, to Auguste Comte, who will construct an 
entire mythology of the prime numbers) , and it is easy 
to understand how naturally the thought of Pythagoras 
* In this society absolute obedience prevailed even in the speculative 
sphere. It was in the Pythagorean brotherhood, not in the schools of 
the Christian Middle Ages, that everything yielded to the Magister 
dixit, airrbs ^<pa 
56
THE PRE-SOCRATIC PHILOSOPHERS 
and his disciples passed from the sign to the cause and 
made the symbol a principle of reality. 
Consequently, numerical principles were regarded 
as the principles of everything that exists ; from the 
opposition between the determinate and the indeter-minate 
(infinite) are derived all the fundamental 
pairs of opposites—odd and even, the elements of 
number, the one and the many, right and left, male 
and female, rest and motion, straight and crooked, 
light and darkness, good and evil, the square and 
the quadrilateral with unequal sides—which determine 
the nature and activity of things. Every essence has 
its number and every essence is a number. The 
number 4, for example, is not simply a figure ofjustice, 
it constitutes the essence of justice ; similarly the 
number 3 constitutes holiness, the number 7 time, 
the number 8 harmony, the number 5 the union of the 
sexes, the number 10 perfection. When numbers 
which in themselves are not localised receive a position 
in space, bodies come into existence. Thus all 
speculation on the origin or nature of things resolves 
itself into speculation on the genesis and properties 
of numbers. 
Pythagoras, therefore, and his school, to whom 
mathematics, music and astronomy owe so much, 
never arrived at the true conception of the first 
philosophy or metaphysics. They achieved, it is true, 
a degree of abstraction superior to that at which the 
lonians had halted, and did not, like them, confuse 
metaphysics with physics. But they confused it 
with the science of number, into which, moreover, 
they imported qualitative interpretations ; and, con-sequently, 
in spite of their effort to reach the object 
57
AN INTRODUCTION TO PHILOSOPHY 
of pure intelligence, they were held fast in the bonds 
of imagination. And if, on the other hand, they 
perceived that the nature of things is intrinsically 
determined by immaterial principles more real and 
truer than that which is tangible and visible, they 
were not yet able to attciin the notion of the formal 
cause, whose full elucidation was reserved to Aristotle 
alone. 
It is to Pythagoras, as we have already remarked, 
that we owe the term, philosophy. A passage ofDiogenes 
Laertius (viii, 8) shows that for him the dignity of 
science consisted in its purely speculative and dis-interested 
character, a point on which Aristotle, 
at the beginning of his Metaphysics, was to insist strongly. 
" Human life," he said, " may be compared to the 
pubhc games, which attract diverse sorts of men. 
Some come to compete for honours and the crowns of 
victory, others to trade, others, the more noble sort, 
solely for the enjoyment of the spectacle. Similarly 
in Ufe some work for honour, others for profit, a few 
for truth alone; they are the philosophers. ." 
. . 
Pythagoras appears to have taught the unity of God, 
whom he regarded as one omnipresent Spirit from 
whom our spirits emanated. He was the first to give 
the universe the name y.6a[ioc^, which, like the Latin 
mundus, conveys the idea of beauty and harmony. 
The most famous and the most derided of his tenets 
was the doctrine of the transmigration of souls, or 
metempsychosis, which he probably derived not from 
Egypt, as Herodotus suggests, but from Hinduism 
(by way of Persia) ^ a doctrine which very early 
; 
^ As Gomperz observes, " the Asiatic Greeks and a portion of the 
population of India were already subject, when Pythagoras left his 
58
THE PRE-SOCRATIC PHILOSOPHERS 
obtained in Greece the adhesion of the Orphics and 
the Pythagoreans. " Coming one day upon a puppy 
which was being cruelly beaten," the aged Xenophanes 
wrote of Pythagoras in mordant verse, " he lamented 
its fate and cried out in pity ' Stop Don't beat him. 
! That is the soul of one of my friends him 
; I recognise by " 
his voice.* 
The Pythagoreans also believed that the revolution 
of cosmic cycles must produce the everlasting recur-rence 
at enormous intervals of all things, reproduced 
identically even in the most insignificant details. 
" According to the Pythagoreans," Eudemus told his 
disciples, " a day will come when you will be all 
gathered again, sitting in the very same places to 
listen, and I shall be telUng you the same story once 
more." ^ 
Astronomy was among the sciences which the 
Pythagorean school cultivated most successfully. Philo-laus, 
who taught that the earth, the sun, and all the 
stars revolved around a mysterious centre of the 
universe filled with fire, may be regarded as a distant 
precursor of Copernicus. But even in this sphere the 
Pythagoreans betrayed in the most flagrant fashion 
the vices of the exclusively mathematical mind. 
native Ionia, to the same ruler, Cyrus, the founder of the Persian 
Empire " {Thinkers of Greece, i, 3). 
Speaking more generally, it was, it would apf>ear, by way ol the 
Pythagorean school that certain distinctively Oriental conceptions and 
modes of thought first entered Greece, to pass from Pythagoreanism to 
Platonism and neo-Platonism, and thence, swollen by further additions, 
into Gnosticism and the more or less undergroimd stream of heterodox 
speculation. 
1 Simplicius, Phys., 732, 30 D. Nietzsche, who was obsessed and 
driven to despair by the thought " of the everlasting recurrence of 
things," derived this singular conception from Greek philosophy. 
59
AN INTRODUCTION TO PHILOSOPHY 
" The Pythagoreans," wrote Aristotle/ " having been 
brought up in the study of mathematics . . . suppose 
the whole heaven to be a musical scale and a number. 
All the properties of numbers and scales which they 
could show to agree with the attributes and parts, 
and the whole arrangement of the heavens, they col-lected 
and fitted into their scheme ; if there was a 
gap anywhere, they readily made additions so as to 
make their whole theory coherent. For example, as 
the number i o is thought to be perfect and to comprise 
the whole nature of numbers, they say that the bodies 
which move through the heavens are ten ; but as 
the visible bodies are only nine, to meet this they 
invent a tenth—the counter-earth—not studying the 
phenomena to discover their causes and test their 
hypotheses, but imposing upon the phenomena their 
hypotheses and preconceived beliefs, thereby claiming 
to assist God to fashion the universe." 
THE ELEATICS 
Though it cannot, strictly speaking, be said that the 
school of Elea founded metaphysics, since it failed to 
keep a firm grasp of the truth, it must receive the 
credit of having raised Greek thought to the meta-physical 
level and attained the necessary degree of 
abstraction. The oldest of the Eleatics was Xeno-phanes, 
a wandering rhapsodist, born about the year 
570 at Colophon, whence he migrated to Elea in 
southern Italy—banished, no doubt, by the Persian 
invasions. Xenophanes poured scorn upon the mytho-logy 
of the poets and the opinions of the common 
1 Melaph., , 5, 986 a. De Caelo, ii, 13, 293 a. 
60
THE PRE-SOGRATIG PHILOSOPHERS 
people. " Far better," he said, in slighting reference 
to the honours paid to athletes, " is our art than the 
strength of men and horses." He taught the absolute 
unity of God, but confounded him with the universe, 
declaring in a pantheistic sense that God is one and 
all, ev xal Tcav. 
But the most profound thinker, indeed the true 
founder of this school, was his disciple Parmenides of 
Elea (born 540), the Great Parmenides, as Plato 
called him. Transcending the w^orld of sensible 
phenomena and even that of mathematical forms or 
essences and numbers, he attained to that in things 
which is purely and strictly the object of the intellect. 
For it can scarcely be denied that the first truth about 
things which the intellect perceives is that they exist, 
their being. The notion of being, thus abstracted, 
impressed Parmenides so powerfully that it fascinated 
him. As his contemporary Heraciitus was the slave 
of change, Parmenides was the slave of being. He 
had eyes for one thing alone : what is is, and cannot 
not be ; being is, non-being is not. Parmenides was 
thus the first philosopher who abstracted and formu-lated 
the principle of identity or non-contradiction, the 
first principle of all thought. 
And as he contemplated pure being, he perceived 
that this being is completely one, absolute, immutable, 
eternal, without becoming, incorruptible, indivisible, 
whole and entire in its unity, in everything equal to 
itself, infinite ^ and containing in itself every perfec-tion.- 
But while he thus discovered the attributes of 
him who is, he refused to admit that any other being 
^ Simpiicius, Phys., 144, 25-145, 23. (Diels, frag. 8, 22.) 
' Aristotle, Phys., i, 3. 
61
AN INTRODUCTION TO PHILOSOPHY 
could exist, and rejected as a scandal to the reason 
the being mingled with non-entity, because produced 
from nothingness, of every creature. 
He was thus led so far astray that he ascribed to 
the being of the world that which belongs only to 
uncreated being. And rather than be false to what 
he believed were the exigencies of being and reason, 
he preferred to refuse heroically the witness of the 
senses and deny the existence in the universe of change 
or multiplicity. Change, motion, becoming, as also 
the diversity of things, are but an illusory appearance. 
There exists only being, the one. 
Does not change imply that an object both was 
and was not (what it becomes), and at the same time 
continues and ceases to be (what it was) ? Does not 
multiplicity imply that what is (this) is not (that) ? 
Do not, therefore, multiphcity and change contradict 
the fundamental principle that what is possesses in 
itself being and not non-being ? 
It was in defence of Parmenides's doctrine of the 
impossibility of change that his disciple Zeno of Elea ^ 
(born 487) composed his famous arguments, by which 
he claimed to prove that the very concept ofmovement 
is self-contradictory : arguments fallacious, no doubt, 
but of singular force and refutable only by the doctrine 
of Aristotle. 
Thus Parmenides, reaching the opposite pole to 
Heraclitus, fixed, as he did once for all one of the 
extreme hmits of speculation and error, and proved 
that every philosophy of pure being, for the very 
reason that it denies that kind of non-being which 
^ Not to be confused with Zeno the Stoic, who lived much later 
(350-264) and was born at Cittium in Cyprus. 
62
THE PRE-SOGRATIC PHILOSOPHERS 
Aristotle termed potentiality and which necessarily 
belongs to everything created, is obliged to absorb all 
being in absolute being, and leads therefore to monism 
or pantheism no less inevitably than the philosophy 
of pure becoming. 
63
Ill 
THE SOPHISTS AND SOCRATES 
The long effort of these speculative pioneers which 
we have briefly recapitulated had equipped human 
thought with a number of fundamental truths. But 
if, looking backwards with a knowledge of the mighty 
synthesis in which all those truths, then partially per-ceived, 
have been harmonised and balanced, we can 
contemplate with admiration the gradual formation 
of the vital centres and arteries of philosophy, at 
the time, in fifth-century Hellas, these good results 
were concealed not only by the medley of con-tradictory 
theories, but by the number and gravity 
of prevalent errors, and it seemed as though the 
entire movement had achieved nothing but disorder 
and chaos. 
The Greek thinkers had set out with high hopes of 
knowing everything, and climbing the sky of wisdom 
in a single step. As a result of this immoderate 
ambition, and because they lacked discipline and 
restraint in handling ideas, their concepts were 
embroiled in a confused strife, an interminable battle 
of opposing probabilities. The immediate and obvious 
result of these attempts at philosophising seemed the 
bankruptcy of speculative thought. It is not, therefore, 
surprising that this period of elaboration produced a 
crisis in the history of thought, at which an intellectual 
disease imperilled the very existence of philosophic 
64
THE SOPHISTS AND SOCRATES 
speculation. This intellectual disease was sophistry, 
that is to say, the corruption of philosophy. 
THE SOPHISTS 
Sophistry is not a system of ideas, but a vicious 
attitude of the mind. Superficially the sophists were 
the successors and disciples of the thinkers of an earlier 
generation—even the word sophist originally bore no 
derogatory significance—in reaHty they differed fi-om 
them completely. For the aim and rule of their 
knowledge was no longer that which is, that is to say, 
the object of knowledge, but the interest of the knowing 
subject. 
At once wandering professors accumulating honours 
and wealth, lecturers, teachers of every branch of 
learning, journalists, if one may so call them, supermen, 
or dilettanti, the sophists were anything in the world 
but wise men. Hippias, who achieved equal eminence 
in astronomy, geometry, arithmetic, phonetics, pro-sody, 
music, painting, ethnology, mnemotechnics, epic 
poetry, tragedy, epigram, dithyramb, and moral 
exhortation, ambassador of EUs, and jack-of-all-trades 
(he attended the Olympic Games in clothes made 
entirely by himself), reminds us of some hero of the 
ItaUan Renaissance. Others resemble the philosophes 
of the eighteenth or the " scientists " of the nineteenth 
century. But the most characteristic feature of all 
aUke was that they sought the advantages conferred 
by knowledge without seeking truth. 
They sought the advantages conferred by knowledge 
so far as knowledge brings its possessor power, pre-eminence, 
or intellectual pleasure. With this in view, 
65
AN INTRODUCTION TO PHILOSOPHY 
they put themselves forward as rationalists and 
walking encyclopaedias ; to every question they had 
an answer ready, deceptively convincing ; 
^ and they 
claimed to reform everything, even the rules of 
grammar and the gender of nouns. ^ But their 
favourite study was man, of all the most complex and 
uncertain, but one in which knowledge is easiest 
coined into power and reputation ; and they culti-vated 
most assiduously law, history, casuistry,' politics, 
and rhetoric. They professed to be teachers of virtue. 
They did not seek truth. Since the sole aim of 
their intellectual activity was to convince themselves 
and others of their own superiority, they inevitably 
came to consider as the most desirable form of know-ledge 
the art of refuting and disproving by skiliul 
arguments, for with men and children alike destruction 
is the easiest method of displaying their strength, and 
the art of arguing with equal probability the pros 
and cons of every question—another proof of acumen 
and skill. That is to say, in their hands knowledge 
altogether lost sight of its true purpose, and what 
with their predecessors was simply a lack of intellectual 
discipline became with them the deliberate intention 
to employ concepts without the least regard for that 
delicate precision which they demand, but for the pure 
^ Critias, for example, considered belief in the gods as the invention 
of an astute statesman who sought to keep the citizens obedient by 
clothing the truth in a garment of fiction. 
2 It was Protagoras who attempted to rationalise the genders of 
nouns : desiring, for example, that nijvis (anger) should be given the 
masculine gender, also ttt^Xt?^ (helmet), etc. 
* Recall, for example, the celebrated discussion between Protagoras 
and Pericles after an accidental homicide in the course of an athletic 
contest, on the question who ought to be punished : the man who 
arranged the contest, the unskilful player, or the javelin itself. 
66
THE SOPHISTS AND SOCRATES 
pleasure of playing them off one against the other 
— 
an intellectual game of conceptual counters devoid 
of solid significance. Hence their sophisms or quibbles. 
Their ethics were of a piece. Every law imposed 
upon man they declared to be an arbitrary convention, 
and the virtue they taught was in the last resort either 
the art of success, or what our modern Nietzscheans 
call the will to power. 
Thus, of the spirit which had inspired the lofty 
intellectual ambitions of the preceding age, the sophists 
retained only the pride of knowledge ; the love of 
truth they had lost. More ardently than their pre-decessors 
they desired to achieve greatness through 
knowledge, but they no longer sought reality. If we 
may use the expression, they believed in knowledge 
without believing in truth. A similar phenomenon 
has recurred since in the history of thought and on a 
far larger scale. 
Under these conditions the sole conclusion which 
sophism could reach was what is termed relativism or 
scepticism. Protagoras of Abdera (480-410), for ex-ample, 
maintained that " man is the measure of all 
things—of what is, that it is, and of what is not, that 
it is not," by which he meant that everything is 
relative to the dispositions of the subject and the 
truth is what appears true to the individual. And his 
contemporary, Gorgias of Leontini (died 375), a famous 
orator, in his book entitled Of Mature or the Non- 
Existent, taught (i) that being is not, in other words, 
that nothing exists : non-existence is non-existence, 
and therefore it is—a quibble on the word is which 
one day would be Hegel's grand metaphysical game 
hence being, its contrary, is not ; 
(ii) that if anything 
67
AN INTRODUCTION TO PHILOSOPHY 
existed we could not know it ; (iii) that if anyone 
could know anything, he could not communicate his 
knowledge to another. 
SOCRATES 
It was Socrates (469-399) who saved Greek thought 
from the mortal danger into which the sophists had 
brought it. Except for the fact that he took no fees 
for his instruction, his manner of Ufe was externally 
the same as theirs. Like them, he spent his time in 
discussions with young men, and it is not surprising 
that a superficial observer, such as Aristophanes, 
confused him with the sophists. In reality, he waged 
against them an unrelenting war and opposed them 
at every point. The sophists claimed to know 
everything and did not believe in truth ; Socrates pro-fessed 
ignorance and taught his hearers to seek nothing 
but the truth. Thus his entire work was a work of 
conversion. He reformed philosophic reasoning and 
directed it to the truth, which is its proper goal. 
This work was of such importance for the fiiture of 
the human intellect that it is not strange that Socrates 
accomplished it as a mission divinely imposed. He 
possessed not only an extraordinary power of philo-sophic 
contemplation (we are informed by Aulus 
Gellius and Plato that he sometimes passed entire 
days and nights motionless, absorbed in meditation) 
but also something which he himself regarded as 
daimonic or inspired, a winged fervour, a free but 
measured force, even perhaps at times an interior 
instinct of a higher order, gifts suggestive of that 
extraordinary assistance in regard to which Aristotle 
68
THE SOPHISTS AND SOCRATES 
said that those who are moved by a divine impulse 
need no guidance of human reason, since they possess 
in themselves a better principle.^ He compared him-self 
to a gadfly sent to sting the Athenians awake 
and force upon their reason a constant examination 
of conscience, a service which they repaid with hem-lock, 
thus affording the aged Socrates, already on the 
verge of the grave, opportunity for the most sublime 
death to which merely human wisdom can lead. 
(a) Socrates was not a metaphysician, but a prac-titioner, 
a physician of souls. His business was not to 
construct a system, but to make men think. This 
was the method by which he could best conquer a 
sophistry whose radical vice was not so much an error 
of doctrine as a deformity of the soul. 
For the chief topic of Socrates's discussions was the 
problem of the conduct of human Hfe, the moral 
problem. His ethics, as far as we can judge it by the 
reports of Plato and Xenophon, seems at first sight to 
have been dictated by narrowly utiUtarian motives. 
What I ought to do is what is good for me, and what 
is good for me is what is useful to me—really useful. 
But at once the need becomes evident of discovering 
what is really useful to man ; and at this point 
Socrates compelled his hearers to acknowledge that 
man's true utility can only be determined by reference 
to a good, absolute, and incorruptible. By thus 
constantly raising the question of man's last end,^ 
and directing men towards their sovereign good, he 
went beyond utilitarianism of every description, and, 
^ Magn. Moral., vii, 8. Cf. Eth., vii, i. 
' A question which he himself seems to have answered some-what 
ambiguously. 
69
AN INTRODUCTION TO PHILOSOPHY 
with the full force of a sane common sense, vindicated 
the supremacy of moral good and our great eternal 
interests ; his ethics thus passed over into the meta-physical 
sphere. In the second place Socrates proved 
by every method of argument that in order to behave 
rightly man's first requisite is knowledge ; he even went 
so far as to maintain that virtue is identical with 
knowledge and therefore that the sinner is simply 
an ignorant man. Whatever we think of this mistake, 
the fact remains that for Socrates ethics was nothing 
if not a collection of truths established by demonstra-tion, 
a real and a genuine science. In this two-fold 
character, metaphysical and scientific, of his 
moral teaching, he stood in radical opposition to the 
sophists, and may be regarded as the founder of ethics, 
{b) But it was impossible to found scientific ethics 
without defining at the same time the laws which 
determine scientific knowledge of every description. 
Here we reach the essence of the Socratic reform. 
By returning to reason itself to study the conditions 
and value of its progress towards truth, that is to say, 
by the use of logical and critical reflection, Socrates 
disciplined the philosophic intelligence, showed it the 
attitude to adopt and the methods to employ in order 
to attain truth. 
The first requisite was to cleanse the mind of the 
false knowledge which pretends to get to the bottom 
of things with a few facile ideas. That is why he 
always began by leading those whom he entangled in 
the net of his questions to confess their ignorance of 
that which they had been certain they knew best 
(Socratic irony). But this was merely the preliminary 
stage of his method. Soon the questioning began 
70
THE SOPHISTS AND SOCRATES 
afresh, but its object now was to lead Socrates's inter-locutor, 
whose thought it guided in the direction 
desired, to discover for himself the truth of which he 
had admitted his ignorance. This was the essence 
of the Socratic method, his maieutic, the art of intel-lectual 
midwifery. Moreover, Socrates reaUsed so 
thoroughly that the attainment of truth is a vital 
and personal activity, in which the teacher can only 
assist his pupil's intelUgence, as a doctor assists nature, 
but the latter is the principal agent,^ that he compared 
the acquisition of knowledge to the awakening of a 
memory dormant in the soul, a comparison from which 
Plato was to derive his famous theory of reminiscence 
(dva[xvif)(Ti.(;'). 
How, then, did this maieutic form the philosophic 
intellect ? By determining its proper object, teaching 
it to seek the essences and definitions of things. ^ Socrates 
was never weary of recalling reason to this one object : 
that which the subject of discussion is, what is courage, 
piety, virtue, the art of ship-building or cobbling, etc. 
All these have a being peculiar to themselves, an 
essence or nature which the human understanding 
can discover and express by a definition which dis-tinguishes 
it from everything else. Because Socrates 
thus required that the essential should in all cases 
be distinguished from the accidental, and because he 
persistently employed his intellect in the search for 
essences, his philosophy may be termed the philosophy 
of essences. It was no longer a question of reducing 
everything to water, fire, numbers, or even absolute 
being, nor yet of finding some indeterminate concept 
1 Cf. St. Thomas, Sum. Theol., i, q. 117, a. i. 
2 Cf. Aristotle, Metaph., xi, 4, 1078 b 17-3^'. 
71
Am INTRODUCTION TO PHILOSOPHY 
sufficiently elastic to enfold everything like a shapeless 
mantle. On the contrary, Socrates's aim was to attain 
the proper intellectual expression of everything—to 
define and determine its essence by a concept applicable 
only to itself. 
At the same time Socrates taught the reason, if not 
by a finished theory of ratiocination, or by construct-ing, 
as Aristotle would construct later, a logic of 
syllogism and demonstration, at least practically and 
of set purpose, to employ concepts, not, as in the 
barbarous word-play of the sophists, as weapons to 
deliver strokes at haphazard, but in such a fashion 
that they fitted exactly the outhne and structure of 
reahty. He thus created dialectic, an instrument of 
knowledge, as yet no doubt defective, but which 
paved the way for the correct notion of scientific 
knowledge, and was compared by Plato to the art of 
the expert chef who cuts up a fowl by distinguishing 
and following carefully the smallest joints of its 
anatomy. 
(c) Thus this unwearied disputer, for all his 
superficial scepticism, possessed an invincible con-fidence 
in the intellect and in science—but of an 
intellect disciplined, humble in its attitude towards 
reality, and a science aware of its limitations, advancing 
successfiilly and securely in the apprehension of truth 
only so far as it respected the sovereignty of the real 
and was conscious of its ignorance in every direction. 
In this we recognise Socrates as the teacher of the 
scientific spirit, as also of the philosophy which we 
shall learn to know as moderate intellectualism. By his 
logical and critical work he forged the instrument 
indispensable for the progress of the mind and turned 
72
THE SOPHISTS AND SOCRATES 
the crisis created by sophistry to the profit and salva-tion 
of reason. By his work as a teacher of morality, 
he not only founded the science of ethics, but Uberated 
thought from the fascination of the sensible, and 
unintentionally, perhaps, set philosophic speculation 
on the road to metaphysics, wisdom in the strict 
sense. This he did simply by raising philosophy (this 
was the true significance of the Socratic demand for 
self-knowledge) from exclusive occupation with the 
physical universe ^ to the study of human nature and 
human activities, which contain a spiritual element 
of a higher order altogether than Uie stars or the 
entire universe of matter. 
But Socrates was no more than a pioneer of genius. 
He gave the impulse, but never reached the goal. 
When he died everything was still in the air. For 
method is not enough, a systematic body of doctrine 
is necessary ; and Socrates, though his teaching was 
fertile in fruitful hints, possessed, apart from the 
elements of ethics, no doctrine in the strict sense. The 
doctrinal completion of his work and the construction 
of the true philosophy were reserved for Plato and 
Aristotle. 
1 Parmenides himself arrived at the metaphysical conception of 
being by an exclusive consideration of the corporeal imiverse. 
73
IV 
PLATO AND ARISTOTLE 
So undoctrinal was Socrates's teaching that his 
disciples developed it along very divergent lines. The 
philosophers known as the minor Socratics, who seized 
upon some partial aspect of the Master's thought, 
which they distorted more or less, were either moralists 
pure and simple (like the Cyrenaics,^ who placed man's 
last end in the pleasure of the moment, and the 
Cynics,^ who, going to the opposite extreme, deified 
force of character or virtue), or logicians infatuated 
by the love of argument (eristics) , like the neo-sophists 
of Elis and especially of the school of Megara,' who 
tended to deny the possibility of knowledge and, by 
compelling philosophers to find answers to their argu-ments, 
indirectly contributed to the progress of logic. 
The Megarians denied that in any judgment one thing 
could be predicated of another. According to them, 
this amounted to affirming that the predicate was the 
subject and thus everything became identical with 
1 The leading philosophers of this school were Aristippus of Gyrene, 
Theodore the Atheist, Hegesias, and Anniceris. 
2 The name was derived from the gymnasium in Athens (Kwdaapye^) 
where Antisthenes taught. The chief Cynics were Antisthenes (born 
445 B.C.), Diogenes of Sinope (400-323), and Crates of Thebes. 
* The principal representatives of the school of Elis were Phaedo 
and Menedemus, of the Megarian, Euclid of Megara (not to be confused 
with Euclid the geometrician), Eubulides of Miletus, Diodorus Cronos, 
and Stilpo. 
74
PLATO AND ARISTOTLE 
everything else. Hence being exists is the sole legitimate 
proposition, and the Eleatic metaphysics the only true 
philosophy. 
PLATO 
The appellation major Socratics belongs only to 
Plato, Socrates's intellectual heir, and his disciples. 
Plato (427-347),^ whose father was of royal descent 
and whose mother traced her pedigree to Solon, 
ambitious to reign as king in the intellectual domain, 
endeavoured to combine in the powerful unity of an 
original system the entire host of speculations which 
he found scattered and fragmentary in the conflicting 
systems of his philosophic predecessors. With him 
philosophy attained her majority. But the work 
which he attempted and which the Socratic reform 
had made possible remained incomplete and defective. 
Under the impulse of his lofty and daring genius, the 
intellect soared too fast and too high, and failed to 
achieve by a final victory the conquest of reaHty. 
(fl) Like Parmenides, Plato understood that the 
subject of metaphysics is being itself But he refused 
to absorb everything which exists in the unity of 
immutable and absolute being. And he was thus led 
to the discovery of important metaphysical truths. 
He perceived that, since things are more or less 
perfect, more or less beautiful and good, more or less 
deserving of love, and since there are things whose 
goodness is mixed with evil—which in philosophic 
terminology participate in goodness—there must neces- 
1 After extensive travels, Plato settled in Athens, v^'here he purchased 
the estate of a certain Academus to be the home of his school, known 
therefore as the Academy. 
75
AN INTRODUCTION TO PHILOSOPHY 
sarily exist a being in which goodness, beauty, and 
perfection are full and entire, unmingled with their 
contraries, a being which is the ground of the beauty 
and goodness of everything else. His thought thus 
ascended to the true God transcending the world and 
distinct from it, whom he saw as goodness itself, the 
absolute good, the good, so to speak, in person. But 
this was not the most prominent aspect of Platon-ism. 
As we pointed out above, Socrates's philosophy 
—a philosophy rather suggested by his practice 
than formulated theoretically—was the philosophy of 
essences ; Plato's philosophy was, before everything 
else, the philosophy of ideas. 
Socrates had shown that what we must seek and 
attain at all cost are the essences of things which the 
mind apprehends and expresses in a definition. What, 
then, is it wliich the intellect perceives when it appre-hends 
the essence of a man, a triangle, white, or 
virtue ? Clearly man, abstracting from Peter, Paul, 
John, etc., triangle, abstracting from any particular 
triangle, isosceles or equilateral, and similarly white 
and virtue. Moreover, the concept or idea of man or 
triangle persists the same when applied to a host of 
men or triangles individually different. In other 
words, these ideas are universals. Further, they are 
immutable and eternal in this sense, that even if, for 
example, no actual triangle existed, the idea of triangle, 
with all the geometrical truths it involves, would 
remain eternally the same. Moreover, these ideas 
enable us to contemplate, pure and unalloyed, the 
humanity, triangularity, etc., in which the different 
beings we know as men, triangles, etc., participate. 
Failing to analyse with sufficient accuracy the nature 
76
— 
PLATO AND ARISTOTLE 
of our ideas and the process of abstraction, and 
applying too hastily his guiding principle, that what-ever 
exists in things by participation must somewhere 
exist in the pure state, Plato arrived at the conclusion 
that there exists in a supra-sensible world a host of 
models or archetypes, immaterial, immutable, eternal, 
man in general or man in himself, triangle in itself, 
virtue in itself, etc. These he termed ideas, which are 
the object apprehended by the intellect, the faculty 
which attains truth—that is to say, they are reality. 
But what, then, is the status of the sensible world ? 
What are we to think of the individual changing and 
perishable things we see and handle ? Since they are 
not the ideas, they are not reality. They are, as 
Heraclitus taught, mere becoming. Plato did not 
deny their existence, but regarded them as feeble 
and deceptive images of reality, the object of opinion 
(So^a), not of certain or true knowledge, fleeting as 
shadows cast upon a wall. Man, therefore, captive 
of the body and the senses, is like a prisoner chained 
in a cave, on the wall of which he sees pass before him 
the shadows of the living beings who move behind 
his back in the sunlight—fugitive shadows, evading 
his grasp, of the idea-substances lit by the Sun of the 
intelligible world, God or the idea of the good. 
But, after all, is a metaphor an explanation ? The 
Platonic ideas are that in virtue of which things 
possess their specific natures 
man in himself or 
humanity is that which makes Socrates a man, the 
beautiful in itself or beauty is that which makes Alci-biades 
or Gallias handsome. In other words, the 
Platonic ideas are the essences and the perfections of 
things. But, on the other hand, they are distinct 
77
AN INTRODUCTION TO PHILOSOPHY 
from things—belong, indeed, to another world. How, 
then, are we to explain the relationship between things 
and their ideas ? Plato replied by calling them 
likenesses or participations of the ideas. But these 
terms, which later will receive in Scholasticism a 
profound significance, are in Plato's system nothing 
more than metaphors devoid of strictly intelligible 
content. And the further question immediately arises, 
why and how anything exists except ideas—that is to 
say, anything not pure reality. In other words, 
Plato has to explain what the thing is which partici-pates 
in the ideas, receives their likeness or reflection. 
Plato replied that it is matter (or the boundless, 
(ScTCEtpov) . And since the ideas are that which is, he 
was compelled to regard matter as that which is not, 
a sort of non-existent being : a pregnant conception 
which, in Aristotle's hands, was to be purged of all 
internal contradiction, but which, as presented by 
Plato, seems self-contradictory, the more so since he 
confuses it elsewhere with the pure space of the 
mathematician. 
{b) Undisturbed by these metaphysical difficulties, 
Plato proceeded to build up, in accordance with its 
inner logic, the edifice of his system. For the theory 
of ideas involved an entire system of philosophy, 
embracing the nature of knowledge, man, and the 
physical universe. 
Human knowledge was divided into two totally 
different categories : imagination (elxacria) and opinion 
(So^a), which are concerned with that which of its 
nature cannot be the object of knowledge, the visible 
and corruptible world and its deceptive shadows ; and 
78
PLATO AND ARISTOTLE 
intellectual knowledge (voyjcji,!;'), which is concerned 
with intelligible things and is itself subdivided into 
reason (Siavoia), whose object is mathematical number, 
and intellect (vou(^), which rises by means of dialectic 
to the intuitive contemplation of the idea-essences 
and finally of God, the super-essential good. 
Plato had now to explain this intellectual knowledge, 
the origin of the ideas in our minds, images of the 
eternal ideas. Since these ideas cannot be derived 
from the senses, which are fettered to illusion, we 
must receive them immediately from on high, and 
they must be innate in our soul. In a former existence, 
before its union with the body, the soul beheld the 
ideas and possessed intuitive knowledge. That know-ledge 
still remains with us, but, clouded and darkened 
by the Ufe of the body, it abides in the soul as a 
dormant memory, and it is by gradually reviving it 
that the quest of wisdom enables us to reconquer our 
original intuition of truth. Thus man is a pure spirit 
forcibly united with a body, as it were an angel 
imprisoned in the flesh {psychological dualism). The 
human soul lived before it animated the body, to 
which it is tied as a punishment for some pre-natal 
sin, and after death it enters another body ; for, while 
Plato believed in the immortality of the soul, he also 
held the Pythagorean tenet of transmigration, or 
metempsychosis. 
Since the physical world is not an object of know-ledge, 
Plato can speak of it only in fables or myths, 
which he develops with all the resources of a con-summate 
art, although they serve only to disguise 
the impotence of his philosophy to account for material 
reality. 
79
AN INTRODUCTION TO PHILOSOPHY 
— 
It is in his myths that he ascribes the production, or 
rather the organisation, of the world to a demiurge— 
whom, in the opinion of many commentators, he 
regarded as distinct from God and inferior to him 
and expounds the queer notion that man is the origin 
of all living organisms : the first men created by the 
gods were of the male sex ; those who led evil lives 
were changed after death into women, who, in turn, 
if they continued to sin, were changed into irrational 
brutes, perhaps even into plants. 
In ethics Plato, like his master Socrates, but more 
clearly than he, established the fundamental truth of 
moral philosophy : neither pleasure nor virtue nor 
any partial good, but God himself, and God alone, 
is the good of man. But how does man attain his 
good ? By making himself, Plato replied, as like as 
possible to God by means of virtue and contemplation. 
Plato also examined, though inadequately, the concept 
of virtue, and outlined the theory of the four cardinal 
virtues, prudence, justice, fortitude, and temperance. 
He taught that it is better to suffer than commit 
injustice, and in the Republic ^ he paints so sublime and 
so flawless a portrait of the righteous sufferer that it is 
as though he had caught a glimpse of the Divdne Face. 
But, as a result of his exaggerated intellectualism, 
he failed to distinguish the acts of the practical from 
those of the speculative intellect and identified virtue, 
which requires rectitude of the will, with knowledge, 
which is a perfection of the reason alone. He there-fore 
misapplied the principle, in itself true, that the 
will always follows the guidance of the understanding, 
and maintedned that sin is simply due to lack of know-ii, 
362 A 
80
. 
PLATO AND ARISTOTLE 
ledge and that no one deliberately does evil : " the 
sinner is merely an ignorant person," The conse-quence 
of this theory, which Plato did not intend, is 
the denial of free will. 
Plato's sociology betrays the same idealist and 
rationaHst tendency which leads him to misapply 
another true principle, namely, that the part exists 
for the whole ; so that in his ideal republic, governed 
by philosophers, individuals are entirely subordinated 
to the good of the state, which alone is capable of 
rights, and disposes despotically of every possible 
species of property, not only the material possessions, 
but even the women and children, the life and hberty, 
of its citizens {absolute communism) 
{c) The radical source of Plato's errors seems to 
have been his exaggerated devotion to mathematics, 
which led him to despise empirical reahty. They 
were also due to an over-ambitious view of the scope 
of philosophy, in which Plato, like the sages of the 
East, though with greater moderation and discretion, 
placed the purification, salvation, and life of the 
entire man. 
Moreover, it is on account of these false principles 
latent in his system that all those philosophic dreams 
which tend in one way or another to treat man as a 
pure spirit can be traced directly or indirectly to 
Plato. 
But of Plato himself we may say that his false 
principles grew in an atmosphere too pure to allow 
them to yield their full fruit and poison the essence of 
his thought. St. Augustine was therefore able to 
extract from Plato's gold-mine the ore of truth. 
8i
AN INTRODUCTION TO PHILOSOPHY 
Plato's thought worked on a large scale and sought 
to embrace all things in a single grasp. But his 
superior wisdom and amazing gift of intuition pre-vented 
him from fixing in a final and definite state-men 
of doctrine many a speculation which floated 
vaguely before his mind. Weak points on which 
another philosopher would have insisted he touched 
lightly. Hence what in itself is a mark of imperfec-tion— 
vagueness, imprecision, hints, never worked out, 
with which he is often satisfied, a method of exposition 
more aesthetic than scientific, employing only m,eta-phors 
and symbols, a method which St. Thomas with 
good reason criticises severely ^—was actually his safe-guard, 
preserving him from a too disastrous distortion 
of the truths he had discovered. From this point of 
view it may be said that Platonism is false, if regarded 
in esse as a fully developed system, but, if regarded in 
Jieri as a progress towards a goal beyond itself, of the 
utmost value as a stage in the development of the 
true philosophy. 
ARISTOTLE 
To extract the truth latent in Platonism was the 
mighty reform effected by Aristotle. Aristotle success-fully 
took to pieces Plato's system, adapted to the 
exigencies of reality the formal principles he had dis-covered 
and misapplied, reduced his sweeping per-spectives 
within the limits imposed by a sublime 
common sense, and thus saved everything vital in his 
1 Plato habuit malum modum docendi ; omnia enim figurate dicit et per 
symbola, intendens aliud per verba, quam sonent ipsa verba. (St. Thomas, In 
I De Anima, viii.) 
82
PLATO AND ARISTOTLE 
master's thought. He did more : he founded for all 
time the true philosophy. If he saved whatever was 
true and valuable, not only in Plato, but in all the 
ancient thinkers of Greece, and brought to a successful 
conclusion the great work of synthesis which Plato 
had attempted, it was because he definitively secured 
the attainment of reality by the human intellect. 
His work was not only the natural fi^it of Greek 
wisdom purified from Plato's mistakes and the alien 
elements included in Platonism ; it contained, com-pletely 
formed and potentially capable of unlimited 
growth, the body of the universal human philosophy. 
Before Aristotle, philosophy may be regarded as in 
an embryonic stage and in process of coming to birth. 
Thenceforward, its formation complete, it was capable 
of indefinite development, and knew no bounds. 
Inventum philosophicum semper perfectibile. 
In fact, Greek speculation after Aristotle had spent 
its force, and was unable to keep firm hold of the 
truth. It would receive considerable material enrich-ment, 
but in essentials would deform instead ofperfecting 
philosophy.^ 
{a) For twenty years Aristotle was Plato's disciple ; 
but he was a disciple with the equipment of a formid-able 
critic. No one has refuted Plato's idealism more 
powerfully than he, or more effectively demolished a 
system which places the substance of things outside 
themselves. 
It is perfectly true that the primary object of the 
intellect is, as Socrates taught, the essences of things ; 
^ This is the reason why we have ended with Aristotle this introduc-tory 
sketch of the history of philosophy, or more exactly of the formation 
of philosophy. 
83
AN INTRODUCTION TO PHILOSOPHY 
perfectly true also, as Plato had perceived, that the 
essence of Peter, Paul, or John is humanity or human 
nature, abstracting from the individual characteristics 
peculiar to Peter, Paul, or John. But this essence as a 
universal exists only in the intellect—in our mind,* 
which extracts or abstracts it from the things in 
which it exists individualised ^—and, on the other hand, 
it is solely as an object of intelligence (inasmuch as it 
cannot be conceived apart from certain attributes), 
and not in its real existence, that it is eternal and 
necessary. Therefore the essences of perishable things 
possess no separate existence in the pure state,- and 
the entire Platonic world of archetypal ideas is 
sheer fiction. The truth of the matter is, as we shall 
prove later in detail, that there exists in everything an 
intelligible and immaterial element, which Aristotle 
calls form^ in virtue of which it possesses a specific 
nature or essence. But this principle is not separate 
from things ; it inheres in them as one of the factors 
which constitute their substance. Thus individual 
objects, though mutable and mortal, are no longer 
deceptive shadows ; they are reality. 
If real objects of a higher order exist, none are more 
immediately accessible to our knowledge. If the 
sensible world be, as it were, an imperfect likeness of 
the divine life of pure spirit, it is a being which 
resembles another being, not a mere image which has 
no existence in itself. If the world is subject to 
becoming, it is not pure becoming, but contains 
1 And primarily in the Divine Intellect, as the Schoolmen were to 
explain, thus taking accovmt of the truth contained in Plato's 
exemplarism. 
» Cf. St. Thomas, In I Metaph., 1. lo, 158 (ed. Gathala). 
84
PLATO AND ARISTOTLE 
enduring and substantial realities. If there is no 
science of the individual object of sense as such, never-theless 
a science of sensible reaUty is possible, because 
there exists, incarnate, so to speak, in that reality, 
something intelligible and immaterial. 
Thus the corporeal universe is the object not of 
mere opinion, which can be expressed only by myth 
and allegory, but of scientific knowledge, the science 
of physics. Aristotle was the true founder of physics.^ 
His incomparably powerful genius viewed mobihty 
in the immutable Hght of intellect, showed that all 
change obeys unchanging laws, laid bare the nature 
of motion itself, of generation and corruption, and 
distinguished the four species of causation operative 
in the sensible world. 
In language strangely trenchant and severe, he sums 
up his long polemic against the doctrine of ideas. 
Plato, he argues, completely misconceived the nature 
of the formal cause when he separated it from things. 
While he fancied " he was stating the substance of 
perceptible things," he asserted " the existence of a 
second class of substances," and his " account of the 
way in which they are the substances of perceptible 
things is empty talk ; for ' sharing ' (participating) 
means nothing." He thus made it impossible to give 
a satisfactory account of nature, and, by attributing 
all causation and all true reality to the ideas, he was 
unable to distinguish in the activity of things the 
lespective parts played by the efficient and the final 
1 Aristotle's experimental physics (the science of phenomena) is a 
magnificent intellectual construction totally ruined by mistakes of fact. 
But his philosophical physics (the science of moving being as such) 
lays down the foundations and principles of every true philosophy of 
nature. 
85
AN INTRODUCTION TO PHILOSOPHY 
cause. He thus neglected " the efficient cause which 
is the principle of change," He further failed to give 
any account of the cause of " that which we see to 
be the cause in the case of the arts, for the sake of 
which mind and nature produce all that they do 
produce." For " mathematics has come to be the 
whole of philosophy for modern thinkers, and they 
profess to explain all other things by mathematics." 
" And as to motion, if the ideas are motionless," there 
is no archetype of motion in the world of ideas, but 
in that case " whence," according to the Platonists, 
" did motion come ? If we cannot explain motion, 
the whole study of nature has been annihilated." ^ 
Refutation of the theory of ideas logically in-volved 
the criticism and correction of all the other 
parts of the Platonic system. In epistemology Aristotle 
showed that physics, mathematics, and metaphysics, 
or the first philosophy, are indeed three distinct 
sciences, but that they are distinguished by their 
subject-matter, not by the faculty employed, which in 
all alike is reason. But his most important achieve-ment 
in this sphere was to prove, by the marvellous 
analysis of abstraction which dominates his entire 
philosophy, that our ideas are not innate memories 
of pre-natal experience, but derived from the senses 
by an activity of the mind. 
In psychology, if in his reaction against Plato's 
metempsychosis, and from an excessive caution, he 
refrained from inquiring into the condition of the soul 
after death, at least he laid the firm foundations of the 
spiritualist doctrine by proving, on the one hand, in 
opposition to Plato's dualism, the substantial unity of 
1 Meiaph., i, 9, 992 a 25-992 b 10. 
86
PLATO AND ARISTOTLE 
the human being, composed of two substantial parts 
incomplete and complementary, and, on the other, 
against the materialists, the spiritual nature of the 
operations of the understanding and will. He thus 
created the only psychology capable of assimilating 
and explaining the vast material accumulated by 
modern experiments. 
In ethics, by distinguishing between the speculative 
judgment (which proceeds from the understanding 
alone) and the practical judgment (which proceeds con-jointly 
from the will), he showed how free will is 
possible, and how the sinner does what he knows to 
be evil, and drew, especially in his treatment of the 
cardinal virtues and in his analysis of human acts, 
the outlines of what was to be, so far as the natural 
order is concerned, the ethics of Christianity. • 
{b) But Aristotle must be studied, not only in his 
attitude to Plato, but absolutely in his attitude to that 
which is. For Plato did no more than furnish him 
with the occasion to wrestle with the problem of being. 
Aristotle won the match, leaving us his great concepts 
of potentiality and act, matter and form^ the categories^ 
the transcendentals, the causes, as weapons wherewith 
to wage the same intellectual contest, and teaching 
us, as a irue master of wisdom, to rise above the 
study of visible and perishable things to contemplation 
of the hving, imperishable reality which knows no 
change. " Immovable in its pure activity, this being 
is in no way subject to change. . . . On such a prin-ciple 
depend the heavens and the world of nature. 
Its Ufe is such as the best which we enjoy, and enjoy 
for but a short time. For it is ever in this state, since 
87
AN INTRODUCTION TO PHILOSOPHY 
its act is also pleasure—the act of the supreme intelli-gence, 
pure thought thinking itself. ... If God is 
always in that good state in which we sometimes are, 
this compels our wonder ; and if in a better, this is 
yet more wonderful. Life also belongs to God : for 
the act of thought is Ufe, and God is that act ; and 
God's essential act is Ufe most good and eternal. We 
say therefore that God is a living being, eternal and 
perfect, so that life which endures everlastingly 
belongs to God, for God is this life." ^ Moreover, this 
God is perfectly one, absolutely single. " Those who 
say mathematical number is first, and go on to generate 
one kind of substance after another and give different 
principles for each, make the substance of the universe 
a mere series of episodes and they give us many 
governing principles ; but the world must not be 
governed badly. As Homer observed, the rule of 
many is not good ; one is the ruler^ * 
Thus Aristotle, as Alexander of Aphrodisias remarks 
in a fine passage of his Commentary on the Metaphysics* 
" leads us fi"om the things which are themselves on the 
lowest plane, but most familiar to us, up to the Father, 
who has made all things, to God the most sublime, 
and proves that as the founder is the cause of the 
unity of the globe and the brass, so the Divine Power, 
author of unity and maker of all things, is for all 
beings the cause of their being what they are." 
Aristotle's mind was at once extremely practical 
and extremely metaphysical. A rigorous logician, but 
also a keen-sighted realist, he gladly respected the 
1 Metaph., xii, 7, 1072 b ; 9, 1074 b 35. 
* Metaph., xii, 10, 1076 a. 
' Alexander of Aphrodisias, In Metaph., ad 1045 a 36. 
88
PLATO AND ARISTOTLE 
demands of the actual, and found room in his specula-tion 
for every variety of being without violating or 
distorting the facts at any point, displaying an intellec-tual 
vigour and freedom to be surpassed only by the 
crystalline lucidity and angelic force of St. Thomas 
Aquinas. But this vast wealth is arranged in the light 
of principles, mastered, classified, measured, and 
dominated by the intellect. It is the masterpiece of 
wisdom, a wisdom which is still wholly human, but 
nevertheless, from its lofty throne, embraces with a 
single glance the totaUty of things, 
Aristotle, however, was a profound rather than a 
comprehensive thinker. He took little care to display 
the proportions and wide perspectives of his philo-sophy 
; his primary object w£is to apprehend by an 
absolutely reliable method and with a faultless preci-sion 
what in every nature accessible to human know-ledge 
is most characteristic, most intimate—in short, 
most truly itself. Therefore he not only organised 
human knowledge, and laid the solid foundations of 
logic, biology, psychology, natural history, meta-physics, 
ethics, and politics, but also cut and polished 
a host of precious definitions and conclusions sparkling 
with the fires of reality. 
It can therefore be affirmed without hesitation that 
among philosophers Aristotle holds a position alto-gether 
apart : genius, gifts, and achievement—all are 
unique. It is the law of nature that the sublime is 
difficult to achieve and that what is difficult is rare. 
But when a task is of extraordinary difficulty both in 
itself and in the conditions it requires, we may expect 
that there will be but one workman capable of its 
accomplishment. Moreover, a well-built edifice is 
89
AN INTRODUCTION TO PHILOSOPHY 
usually built not on the plans of several architects, 
but on the plan of a single one. If, therefore, the 
edifice of human wisdom or philosophy is to be 
adequately constructed, the foundations must be laid 
once for all by a single thinker. On these foundations 
thousands of builders will be able to build in turn, 
for the growth of knowledge represents the labour of 
generations and will never be complete. But there 
can be but one master-builder.^ 
For that reason, in spite of the mistakes, de-fects, 
and gaps which betray in his work the limita-tions 
of human reason,^ Aristotle is as truly the 
^ Descartes remarks very truly in his Discours de la Mithode : " Works 
composed of many pieces and made by the hand of several workmen 
are not so perfect as those which are the work of a single individual." 
But he was wrong in believing (i) that he was the man destined to 
lay the foundation of philosophy, a work which the ancients had failed 
to accomplish ; (ii) that by himself he was competent—at least, given 
sufficient time and experience—not only to lay the foundation of 
science, but to complete the edifice ; and (iii) in rejecting contemp-tuously 
the entire achievement of preceding generations, together with 
the traditional wisdom of humanity. Aristotle, on the contrary, 
succeeded in his task by constant criticism and analysis of his 
predecessors' thought, and by making use of the accumulated results 
of human speculation in the past. 
2 Aristotle is often credited with certain errors made by his disciples 
or commentators, especially about the human soul and the divine 
knowledge and causality. But a careful study of the text proves that 
when the philosopher maintained that the intellect is separate, he 
meant that it is separate from matter, not from the soul itself {cf. St. 
Thomas, In III De Anima, 4 and 5) , and therefore he did not deny, as is 
often asserted, the personal immortality of the human soul {cf. also 
Metaph., xii, 3, 1070 a 26). Nor did he teach that God is not the 
efficient cause of the world and moves it only as the end, or good, 
which it desires. (The passage in the Metaph., xii, 7, means simply that 
God moves as final cause or object of love the intelligence which moves 
the first heaven ; he does not affirm that God can act only as final cause 
and has not made things. On the contrary, in Metaph., ii, i 
, 993 b 28, he 
says that the heavenly bodies are dependent on the first cause, not only 
90
PLATO AND ARISTOTLE 
philosopher par excellence, as St. Thomas is the theo-logian.^ 
for their motion, but for their very being. Cf. Metaph. vi, i, 1026 b 17.) 
Cf. also the passage from Alexander of Aphrodisias, quoted above in 
the text, in which God's efficient causality in Aristotle's system is 
admirably brought out. As for the passage {Metaph., xii, 9) in which 
Aristotle investigates the formal object of the Divine Intellect, 
remarking that it is better not to know than to know certain things of 
a lower order, it most certainly does not amount to the denial that God 
knows the things of the world ; the statement is put forward simply to 
prepare for the solution of the question discussed. That solution, as 
indicated by Aristotle, is formally true, and consists in the proposition, 
which St. Thomas later affirmed more explicitly, that the Divine 
Intellect, because of its absolute independence, has no other formal 
object than the Divine Essence itself, and therefore does not know 
the things of the world in themselves, but in that essence in which every-thing 
is life. 
It remains true, nevertheless, that Aristotle committed serious errors 
(for instance, his attempt to prove the existence of the world ab aeterno) , 
and was also guilty of many omissions. In particular, the doctrine of 
creation, which follows with absolute necessity from his principles, is 
nowhere explicitly formulated by him (indeed, no heathen philosopher 
reached a clear notion of creation ex nihilo); and on those questions 
which, though in themselves capable of rational proof, are most 
difficult to solve without the aid of revelation—the relation of the world 
to God, the lot of the soul after death—he maintained a reserve, which 
was perhaps very prudent in itself, but leaves his work manifestly 
incomplete. 
1 Goethe, repeating the theme of Raphael's wonderful School of Athens, 
in which Plato is depicted as an inspired old man, his face turned 
heavenward, Aristotle as a man in the full vigour of youth pointing 
triumphantly to the earth and its realities, has drawn in his Theory of 
Colours (Part 2, Ueberliefertes) a striking comparison between Plato and 
Aristotle. " Plato," he says, " seems to behave as a spirit descended 
from heaven, who has chosen to dwell a space on earth. He hardly 
attempts to know this world. He has already formed an idea of it, 
and his chief desire is to communicate to mankind, which stands in 
such need of them, the truths which he has brought with him and 
delights to impart. If he penetrates to the depth of things, it is to 
fill them with his own soul, not to analyse them. Without intermission 
and with the burning ardour of his spirit, he aspires to rise and regain 
the heavenly abode from which he came down. The aim of all his 
91
AN INTRODUCTION TO PHILOSOPHY 
(c) Aristotle was born in 384 at Stagira, in Thrace.' 
The son of a doctor, by name Nicomachus, he belonged 
to the family of the Asclepiadae, descended, it was 
said, from Aesculapius. At the age of eighteen he 
became a pupil of Plato, whose lectures he attended 
until his death (347). After Plato's death he went to 
live at Atarneus in Mysia where Hermias was king, 
and from there to Mytilene. He then spent eight 
years at the court of King Philip of Macedon, where 
he became Alexander's tutor. After his pupil's acces-sion 
to the throne, he settled in Athens, where he 
established his school at the Lyceum (a gymnasium 
consecrated to Apollo Lycaeus). He taught as he 
walked to and fro with his pupils under the trees of 
the Lyceum, whence the name Peripatetics (walkers) 
by which his disciples became known. He spent 
twelve years in Athens. When the party opposed 
discourse is to awaken in his hearers the notion of a single eternal being, 
of the good, of truth, of beauty. His method and words seem to melt, 
to dissolve into vapour, whatever scientific facts he has managed to 
borrow from the earth. 
" Aristotle's attitude towards the world is, on the other hand, entirely 
human. He behaves like an arcliitect in charge of a building. Since 
he is on earth, on earth he must work and build. He makes certain 
of the nature of the ground, but only to the depth of his foundations. 
Whatever lies beyond to the centre of the earth does not concern him. 
He gives his edifice an ample foundation, seeks his materials in every 
direction, sorts them, and builds gradually. He therefore rises like a 
regular pyramid, whereas Plato ascends rapidly heavenward like an 
obelisk or a sharp tongue of flame. 
" Thus have these two men, representing qualities equally precious 
and rarely found together, divided mankind, so to speak, between them." 
1 " We must remember, moreover, that Stagira, a city of Chalcidice, 
was a Greek colony where Greek was spoken. It is therefore incorrect 
to regard Aristotle, as he is sometimes regarded, as only half Greek. 
He was a pure Hellene, as pure as Parmenides, for example, or 
Anaxagoras." (Hamelin, Le Systeme d'Aristote, p. 4.) 
92
PLATO AND ARISTOTLE 
to the Macedonians brought against him an accusation 
of impiety, on the pretext of a hymn he had once 
composed on the death of his friend Hermias, he 
retired to Chalcis, where he died at the age of sixty-three 
(322). 
The story is told of him that his love of study was 
so great that he devised the plan of holding in his 
hand, while at work, a ball of copper which, if he fell 
asleep, would rouse him by falling into a metal basin. 
To assist his researches Philip and Alexander placed 
their vast resources at his disposal. He wrote books 
to be read by the general public {dialogues), which are 
all lost—Cicero praised their style in the highest 
terms : flumen aureum orationis fundens Aristoteles {Acad., 
II, 38, 119)—and acroamatic books, in which he 
summarised the lectures given to his disciples ; the 
majority of these have survived.^ 
The history of these books, as related by Strabo, is 
very strange, and illustrates, as aptly as Pascal's 
remark on Cleopatra's nose, the part played by trifling 
* The following is a list of Aristotle's works : 
— 
i. The collection of works dealing with Logic and known collectively 
by the name of Organon (instrument of scientific knowledge). They 
consist of the Kary^yoplcn, the Categories ; the 'AvaXvTiKa irpdrepa Kai 
Offrepa, the Prior and Posterior Analytics ; the TdmKa, the Topics ; the 
Hfpl <To(piaTiKQ)v iXiyx'^i', On Sophistic Arguments ; and the Uepl ipfiTivelai, 
On Interpretation, De Interpreiatione, a treatise on the meaning of proposi-tions, 
which in spite of Andronicus's rejection must be accepted as 
authentic. 
ii. The Physics, ^vciKi] dxpSaais (the authenticity of Book 7 is 
doubtful), with which we must group the followii^ treatises : On 
the Heavens {De Caelo), Tlepl dvpavoO ; On Generation and Corruption, Uepl 
yeveaiu^ Kai <p0opS.s ; On the Parts of Animals, Uepl ^i^wv fi6pitav ; On 
the Soul, Uepl ^vxv^ ; On Sensation, Tlepl al(r6r,ffews kcu aladriruv ; On 
Memory, Uepl tj.vf)fi.r)% Kai avafivva-eta^ ; the Meteorology, MereupoXoyiKd ; the 
History of Animals, riepi ri fi^a l<rroplai (the authenticity of Book 10 is 
doubtful) ; and many other treatises, several of which are of dubious 
93
AN INTRODUCTION TO PHILOSOPHY 
accidents in determining the destinies of mankind. 
At the philosopher's death they were inherited together 
with his Ubrary by his disciple and successor Theo-phrastus. 
Theophrastus bequeathed them to a disciple 
authenticity, especially the De Mundo. (The treatise on Physiognomy 
is spurious, but is apparently a compilation from authentic fragments.) 
iii. The Metaphysics, Td uer^ rk (f>v<nKd, of which the second book, 
a iXttTTov, is the work of a disciple, Pasicles of Rhodes. 
iv. The Nicomachean Ethics, 'HdiKo, ^iKOfiixfia, and the Eudemian Ethics, 
'KdiKct EvBrifida. The latter work was composed, not by Aristotle, but 
by Eudemus himself. To these we may add the Great Ethics (Magna 
Moralia), "ROiko, tieyaha, which is a risumi of the two preceding; and 
therefore not written by Aristotle ; the Politics, UoKitiko. ; the Poetics, 
Tlepi noti]TiKT]i ; and the Rhetoric, T^x^r) pr/TopiKifi. The treatise On 
Virtues and Vices, the Economics, and the Rhetoric to Alexander are spurious. 
In 1 891 was discovered and published the Constitution of Athens,'AO-nvalwv 
TToXiTela. It formed part of a collection in which Aristotle gave an 
outline of the constitutions of 1 58 Greek states. 
Of the Schoolmen who commented on the works of Aristotle, the 
most important are Albertus Magnus, St. Thomas, and Silvester 
Maurus, whose paraphrase and commentary following the text word 
for word may still be usefully consulted. Saint Thomas wrote 
commentaries : (a) on the De Interpretatione (unfinished and replaced 
by Cajetan's for lessons 3-14 of Book 2) ; [b) on the Posterior 
Analytics ; (c) on the Physics ; (d) on the De Caelo et Mundo (St. 
Thomas died before it was completed, and from Book 3, lesson 8, the 
commentary was continued by his pupil, Peter of Auvergne) ; [e) on 
the Generation and Corruption of Animals (St. Thomas's unfinished 
Commentary has been completed by passages borrowed from other 
writers, in particular from the Commentary of Albertus Magnus) ; 
(/) on the Meteorology (completed by another hand from Bk. 2, lesson 11); 
[g) on the De Anima (the commentary on Books 2 and 3 is by St. Thomas 
himself, the commentary on Bk. i compiled from his lectures by one of 
his pupils, Raynald of Piperno ; (A) on the Parva Naturalia [De Sensu et 
Sensato, de Memoria et Reminiscentia, de Somno et Vigilia) ; (i) on the 
Metaphysics (modern edition by Padre Cathala, Turin, Marietti, 1915) ; 
(j) on the Nicomachean Ethics ; {k) on the Politics (completed by Peter of 
Auvergne from Bk. 3, lesson 6, or, as others think firom Bk. 4) . Cf. De 
Rubeis, Dissert. 23 in vol. i. Op. Omn. S. Thomae Aq., Leonine edition. 
For the writings of St. Thomas and the authenticity of his various 
minor works, see Mandonnet, O.P., Des Ecrits autherUiques de Saint 
Thomas (reprinted fi:om the Revue Thomiste, 1909-19 10) Friburg. 
94
PLATO AND ARISTOTLE 
named Neleus, Neleus in turn to his heirs. The 
latter, fearing they might be seized for the royal 
library at Pergamum, hid them in an underground 
chamber. They died and the manuscripts were lost. 
They remained lost for a century and a half and were 
only recovered by the good fortune of a successful 
book-lover. About the year lOO b.g. the descendants 
of Neleus's heirs discovered the manuscripts (in a 
very bad condition, as we can well imagine) and sold 
them for a large sum to a wealthy collector, Apellicon 
of Teos, who published them, in a very faulty edition. 
At the capture of Athens by the Romans in 86 b.g. 
they came into the possession of Sulla. The gram-marian 
Tyrannic had access to them and made use 
of them, and finally Andronicus of Rhodes catalogued 
and republished them.^ Commentaries were composed 
by Alexander of Aphrodisias (second century a.d.), 
also by several neo-Platonic philosophers. Porphyry 
(third century), Themistius (fourth century), Sim-plicius, 
and Philo (sixth century). 
1 Strabo, Geog., xiii, i, 54 ; Plut. Sulla, 26. Strabo's testimony is of 
considerable weight. It has, however, been proved that some of the 
most important scientific treatises of Aristotle were known to the Peri-patetics 
and their opponents in the third and second centuries B.C. 
We must therefore conclude that Strabo's account is reliable in its 
positive assertions, so far as the history of Aristotle's " acroamatic " 
manuscripts are concerned, but inaccurate, or at least exaggerated, 
in its negative statements. More or less complete copies of the 
Philosopher's works must have been in circulation in the Peripatetic 
school before Apellicon's discovery. We may nevertheless agree with 
Hamelin's conclusion that " the scientific writings of Aristotle were 
little read even by the Peripatetic school in its degeneracy. Apellicon's 
discovery would have had the effect of making these works once more 
fashionable." The truth of the matter therefore would be that, 
before this discovery and the works of Andronicus, Aristotle's scientific 
treatises were not indeed unknown, as Strabo says, but at any rate 
litde and badly known. 
95
AN INTRODUCTION TO PHILOSOPHY 
The Scholastic tradition, which grew up from the 
eighth century onward in the Christian West, was for 
long ignorant of Aristotle's original works, with the 
exception of the Organon (the treatises on logic), 
which had been translated into Latin ^ by Boethius 
(480-526). But it was acquainted with his thought, 
which had been transmitted and popularised at 
second hand and formed an integral part of the great 
philosophic synthesis of late antiquity, Platonic though 
it was in the main, on which the Fathers, especially 
St. Augustine, had drawn so largely in the service of 
the faith. In the Christian schools Aristotle's logic 
was taught in Boethius's translation. But it was not 
until the latter part of the twelfth century that the 
other writings of the Philosopher {physics, metaphysics, 
ethics) began to reach the Schoolmen, mainly, it would 
appear, as a result of the ardent polemic conducted 
at that date by the leaders of Christian thought 
against the philosophy of the Arabs, who possessed 
these books together with the neo-Platonic commen-taries 
in a Syrian version translated later into Arabic, 
and appealed to their authority. At first the object 
of considerable suspicion * on account of the source 
^ Later Boethius's work was partially lost, and it was not until after 
H41 that certain books of the Organon, reintroduced from the Arabs, 
began to appear in the philosophic literature of the Middle Ages, 
where they formed what was then known as the Logica Nova. These 
were the Prior and Posterior Analytics, the Topics, and the Sophistic 
Arguments. Cf. de Wulf, Hist, de la Phil, midiivale, and ed. pp. 149 sqq. 
* Censures (issued in 12 10 by a council of the province of Sens which 
met at Paris, and renewed in 12 15, in the statute imposed on the 
University of Paris by the legate Robert de Courcon, a statute confirmed 
by Gregory IX in 1231 and by Urban IV in 1263) forbade the 
employment of Aristotle's writings in public lectures or private teaching. 
We must remember, however, that, as M, Forget points out {Rapp. cm 
96
PLATO AND ARISTOTLE 
from which they had been received and the mistakes 
which the Arab commentators had foisted into them, 
all the works of Aristotle were soon translated into 
Latin, at first from the Arabic text, later ^ from the 
original Greek. ^ 
Now took place the meeting of human wisdom and 
divine truth, of Aristotle and the Faith. All truth 
belongs of right to Christian thought, as the spoils of 
the Egyptians to the Hebrews. Quaecunque igitur apud 
omnes praeclara dicta sunt, nostra Christianorum sunt,* 
because according to that saying of St. Ambrose, 
which St. Thomas deHghted to quote, every truth, 
whoever said it, comesfrom the Holy Spirit.'^ But someone 
must actually take possession, someone must enrol in 
congr. scientif. intern, des cath., Brussels, 1894), individuals remained free 
to read, study, and comment on these books in private. Moreover, the 
censures bound only the University of Paris. In 1229 the University 
of Toulouse, founded and organised under the protection of the papal 
legate himself, attracted students by announcing lectxires on the books 
forbidden at Paris. Finally, even at Paris, when the faculty of arts 
included in its course from 1255 public lectures on the Physics and 
Metaphysics, ecclesiastical authority made no attempt to interfere. 
And, more significantly still. Pope Urban IV, a kw years later, took 
under his personal patronage William of Moerbeke's translation of 
Aristotle and St. Thomas Aquinas's Commentaries. See GhoUet, 
" Aristotelisme de la Scolastique " in Vacant and Mangenot's 
Dictionnaire de Thiologie, and de Wulf, op. cit., p. 242. 
1 Some of Aristotle's works were apparently read at first in a Latin 
translation from Arabic, others in a direct translation from the Greek. 
In any case it was not long before the latter entirely superseded the 
former. St. Thomas used only direct versions from the Greek. 
2 The best of these translations is that of the entire works of Aristotle 
made between 1260 and 1270 by William of Moerbeke at the suggestion 
and under the supervision of St. Thomas. It is an absolutely literal 
rendering of the Greek text. 
' St. Justin, ApoL, ii, 1 3. 
* Omne verum a quocumque dicatur, a Spiritu sancio est. 
97
AN INTRODUCTION TO PHILOSOPHY 
the royal service of Christ the marvellous intellect 
of Aristotle. This work, begun by Albert the Great 
(i 193-1280), was continued and brought to a success-ful 
conclusion by St. Thomas Aquinas (1225-1274). 
Its accomplishment demanded a rare conjunction of 
favourable conditions—the ripe culture of the age of 
St, Louis, the magnificent organisation of the Domini-can 
order, the genius of St. Thomas.^ St. Thomas, 
whom the Church has proclaimed Doctor par excellence, 
Doctor Communis Ecclesiae, and whom she has enthroned 
as the universal teacher of her schools, was not content 
with transferring the entire philosophy of Aristotle to 
the domain of Christian thought, and making it the 
instrument of a unique theological synthesis ; he 
raised it in the process to a far higher order, and, so 
to speak, transfigured it. 
He purged it from every trace of error—that is to 
say, in the philosophic order, for so far as the sciences 
of observation or phenomena are concerned, St. 
Thomas was no more able than Aristotle to escape 
the errors prevalent in his day, errors which do not 
in any way affect his philosophy itself. He welded 
it into a powerful and harmonious system ; he explored 
* For the successful performance of such a task it was also requisite 
that Christian thought should have attained the high degree of elabora-tion 
alike in the philosophical and theological order which it had 
received from the Fathers and the Scholastic predecessors of St. Thomas. 
Therefore the work of Albertus Magnus and St. Thomas was not to 
change, but on the contrary to complete Christian philosophy, giving 
it its mature expression. If contemporaries were primarily impressed 
with the novelty of their work—a novelty' of completion, not of alteration —the reason is that the final process which perfects a system must 
always come as a shock to those who witness it, and who most likely are 
attached by force of habit to certain aspects of its imperfect state as 
such. 
98
PLATO AND ARISTOTLE 
its principles, cleared its conclusions, enlarged its 
horizon ; and, if he rejected nothing, he added 
much, enriching it with the immense wealth of the 
Latin Christian tradition, restoring in their proper 
places many of Plato's doctrines, on certain funda-mental 
points (for example, on the question of essence 
and existence) opening up entirely new perspectives, 
and thus giving proof of a philosophic genius as 
mighty as that of Aristotle himself Finally, and this 
was his supreme achievement, when by his genius 
as a theologian he made use of Aristotle's philosophy 
as the instrument of the sacred science which is, so 
to speak, " an impress on our minds of God's own 
knowledge," ^ he raised that philosophy above itself 
by submitting it to the illumination of a higher light, 
which invested its truth with a radiance more divine 
than human. Between Aristotle as viewed in himself 
and Aristotle viewed in the writings of St. Thomas is 
the difference which exists between a city seen by the 
flare of a torchlight procession and the same city 
bathed in the light of the morning sun. 
For this reason, though St. Thomas is first and 
foremost a theologian, we may as appropriately, if 
not with greater propriety, call his philosophy Thomist 
rather than Aristotehan. 
This philosophy of Aristotle and St, Thomas is in 
fact what a modern philosopher has termed the natural 
philosophy of the human mind, for it develops and brings 
to perfection what is most deeply and genuinely 
natural in our intellect aUke in its elementary appre-hensions 
and in its native tendency towards truth. 
Sum, TTieoL, i, q. i, a 3, a</ 2. 
99
AN INTRODUCTION TO PHILOSOPHY 
— 
It is also the evidential philosophy, based on the 
double evidence of the data perceived by our senses 
and our intellectual apprehension of first principles 
the philosophy of being, entirely supported by and 
modelled upon what is, and scrupulously respecting 
every demand of reality—the philosophy of the intellect, 
which it trusts as the faculty which attains truth, and 
forms by a discipline which is an incomparable mental 
purification. And for this very reason it proves itself 
the universal philosophy in the sense that it does not 
reflect a nationality, class, group, temperament, or 
race, the ambition or melancholy of an individual or 
any practical need, but is the expression and product 
of reason, which is everywhere the same ; and in this 
sense also, that it is capable of leading the finest 
intellects to the most subHme knowledge and the 
most difficult of attainment, yet without once betraying 
those vital convictions, instinctively acquired by every 
sane mind, which compose the domain, wide as 
humanity, of common sense. It can therefore claim 
to be abiding and permanent [philosophia perennis) in 
the sense that before Aristotle and St. Thomas had 
given it scientific formulation as a systematic philo-sophy, 
it existed from the dawn of humanity in germ 
and in the pre-philosophic state, as an instinct of the 
understanding and a natural knowledge of the first 
principles of reason and ever since its foundation as 
a system has remained firm and progressive, a 
powerful and living tradition, while all other philo-sophies 
have been born and have died in turn. And, 
finally, it stands out as being, beyond comparison 
with any other, one ; one because it alone bestows 
harmony and unity on human knowledge—both meta-lOO
PLATO AND ARISTOTLE 
physical and scientific—and one because in itself it 
realises a maximum of consistency in a maximum of 
complexity, and neglect of the least of its principles 
involves the most unexpected consequences, distorting 
our understanding of reahty in innumerable directions. 
These are a few of the external signs which witness 
to its truth, even before we have studied it for ourselves 
and discovered by personal proof its intrinsic certitude 
and rational necessity. 
lOI
DEFINITION OF PHILOSOPHY 
We began by calling philosophy human wisdom. 
Now that the history of its origins has given us further 
information as to the nature and object of this wisdom, 
we are in a position to attempt a more precise definition 
of philosophy. 
For this purpose we shall take philosophy to mean 
philosophy par excellence, the first philosophy or meta-physics. 
What we shall say of it in the absolute sense 
[simpliciter) will be applicable relatively [secundum quid) 
to the other departments of philosophy. 
Philosophy is not a " wisdom " of conduct or 
practical fife that consists in acting well. It is a wisdom 
whose nature consists essentially in knowing. 
How ? Knowing in the fullest and strictest sense 
of the term, that is to say, with certainty, and in being 
able to state why a thing is what it is and cannot be 
otherwise, knowing by causes. The search for causes 
is indeed the chief business of philosophers, and the 
knowledge with which they are concerned is not a 
merely probable knowledge, such as orators impart 
by their speeches, but a knowledge which compels 
the assent of the intellect, Hke the knowledge which 
the geometrician conveys by liis demonstrations. But 
certain knowledge of causes is termed science. Philo-sophy 
therefore is a science. 
Knowing by what medium, by what light ? Know- 
102
DEFINITION OF PHILOSOPHY 
ing by reason, by what is called the natural light of the 
human intellect. This is a quality common to every 
purely human science (as contrasted vsdth theology). 
That is to say, the rule of philosophy, its criterion of 
truth, is the evidence of its object. 
The medium or light by which a science knows its 
objects is termed in technical language its lumen sub 
quo, the light in which it apprehends the object of its 
knowledge (itself termed the object, quod). Each of 
the different sciences has its own distinctive light 
{lumen sub quo, medium sen motivum formale) which 
corresponds with the formal principles by means of 
which they attain their object. But these different 
principles are aUke in this, that they are all known 
by the spontaneous activity of our intellect, as the 
natural faculty of knowledge, in other words by the 
natural light of reason—and not, hke the principles 
of theology, by a supernatural communication made 
to man {revelation), and by the light of faith. We have 
now to consider the object quod of philosophy. 
Knowing what ? To answer this question we may 
recall the subjects which engaged the attention of 
the different philosophers whose teachings we have 
summarised. They inquired into everything—know-ledge 
itself and its methods, being and non-being, 
good and evil, motion, the world, beings animate 
and inanimate, man and God. Philosophy there-fore 
is concerned with everything, is a universal 
science. 
This does not, however, mean that philosophy 
absorbs all the other sciences, or is the sole science, 
of which all the rest are merely departments ; nor on 
103
AN INTRODUCTION TO P H I L O S O P H 'y 
— 
the Other hand that it is itself absorbed by the other 
sciences, being no more than their systematic arrange-ment. 
On the contrary, philosophy possesses its 
distinctive nature and object, in virtue of which it 
differs from the other sciences. If this were not the 
case philosophy would be a chimera, and those 
philosophers whose tenets we have briefly sketched 
would have treated of unreal problems.^ But that 
philosophy is something real, and that its problems 
have the most urgent claim to be studied, is proved 
by the fact that the human mind is compelled by its 
very constitution to ask the questions which the 
philosophers discuss, questions which moreover in-volve 
the principles on which the certainty of the 
conclusions reached by every science in the last resort 
depends. 
" You say," wrote Aristotle in a celebrated dilemma, 
*' one must philosophise. Then you must philosophise. 
You say one should not philosophise. Then (to prove 
your contention) you must philosophise. In any case 
you must philosophise." * 
^ No doubt they also studied many questions relating to die special 
sciences, for the differentiation of the sciences had not been carried so 
far as in modem times. Nevertheless the primary object of their study 
lay elsewhere, and, at least after Socrates, those special sciences 
astronomy, geometry, arithmetic, music, medicine, and geography 
which the ancients cultivated with success developed separately, and 
in clear distinction from philosophy. The very .ki'ory of the special 
sciences, which in modern times have made such enormous progress 
independently of philosophy and as autonomous branches of study, 
plainly proves that they are no part of the former 
2 " el fikv (pCKoao(ptyreov, (ptoao(pr,Tiov, Kal, el fiii cp ' TocprjTiov, (pCKoaocpt)- 
riov, w6.vT(t}% &pa. (pioiTo(prjTiov." 
This dilemma is taken from the UporftirTiKd?, a lost work of which 
only a few fragments have come down to us, {Cf. fr. 50, 1483 b 29, 42 
; 
1484 a 2, 8, 18.) 
104
DEFINITION OF PHILOSOPHY 
But how can pliilosophy be a special science if it 
deals with everything ? We must now inquire under 
what aspect it is concerned with everything, or, to 
put it another way, what is that which in everything 
directly and for itself interests the philosopher. If, for 
example, philosophy studies man, its object is not to 
ascertain the number of his vertebras or the causes 
of his diseases ; that is the business of anatomy and 
medicine. Philosophy studies man to answer such 
questions as whether he possesses an intellect which 
sets him absolutely apart from the other animals, 
whether he possesses a soul, if he has been made to 
enjoy God or creatures, etc. When these questions 
are answered, thought can soar no higher. No 
problems Ue beyond or above these. We may say 
then that the philosopher does not seek the explanation 
nearest to the phenomena perceived by our senses, 
but the explanation most remote from them, the 
ultimate explanation. This is expressed in philo-sophical 
terminology by saying that philosophy is not 
concerned with secondary causes or proximate explana-tions 
; 
^ but on the contrary with Jirst causes, highest 
principles or ultimate explanations. 
Moreover, when we remember our conclusion that 
philosophy knows things by the natural light of reason, 
it is clear that it investigates the first causes or highest 
principles in the natural order. 
When we said that philosophy is concerned with 
everything, everything which exists, every possible 
object of knowledge, our statement was too indefinite. 
We determined only the matter with which philosophy 
^ That is to say, approximating to the particulars of sensible 
phenomena. 
105
AN INTRODUCTION TO PHILOSOPHY 
deals, its material object, but said nothing of the aspect 
under which it views that object, or the attributes of 
that object which it studies that is to say, we did 
; not define its,formal object, its formal standpoint. The 
formal object of a science is the aspect under which it 
apprehends its object, or, we may say, that which it 
studies primarily and intrinsically and in reference to 
which it studies everything else ^ that which philo-sophy 
; 
studies in this formal sense in things, and the 
standpoint from which it studies everything else, is 
the first causes or highest principles of things in so far 
as these causes or principles belong to the natural 
order. 
The material object of a faculty, science, art, or 
virtue, is simply the thing or subject-matter—without 
further quahfication—with which that faculty, science, 
art, or virtue, deals. For instance, the material object 
of chemistry is inorganic bodies ; of the faculty of sight, 
objects within our range of vision. But this does not 
enable us to distinguish between chemistry and 
physics, which is also concerned with inorganic bodies, 
or between sight and touch. To obtain an exact 
definition of chemistry we must define its object as 
the intrinsic or substantial changes of inorganic bodies, 
and similarly the object of sight as colour. We have 
now defined the formal object {objectum formate quod), 
that is to say, that which immediately and of its very 
nature, or intrinsically and directly, or again neces-sarily 
and primarily (these expressions are equivalent 
renderings of the Latin formula per se primo) , is appre-hended 
or studied in things by a particular science, 
1 Qtod per se primo haec scientia consideral et sub cujus ratione caetera omnia 
cogncscil. 
1 06
DEFINITION OF PHILOSOPHY 
art, or faculty, and in reference to which it apprehends 
or studies everything else. 
Thus philosophy, alone among the branches of 
human knowledge, has for its object everything which 
is. But in everything which is it investigates only the 
first causes. The other sciences, on the contrary, have 
for their object some particular province of being, of 
which they investigate only the secondary causes or 
proximate principles. That is to say, of all branches 
of human knowledge philosophy is the most sublime. 
It follows further that philosophy is in strictest truth 
wisdom, for it is the province of wisdom to study the 
highest causes : sapientis est altissimas causas considerare. 
It thus grasps the entire universe in a small number 
of principles and enriches the intellect without 
burdening it. 
The account we have just given is applicable in an 
unquaUfied sense only to the first philosophy or meta-physics, 
but may be extended to philosophy in general, 
if it is regarded as a body of which metaphysics is the 
head.^ We shall then define philosophy in general 
as a universal body of sciences * whose formal stand- 
^ The ancients understood by the term philosophy the sum-total of 
the main branches of scientific study {physics, or the science of nature ; 
mathematics, or the sciences of proportion ; metaphysics, or the science 
of being as such ; logic ; and ethics). There could therefore be no 
question of distinguishing between philosophy and the sciences. The 
one question with which they were concerned was how to distinguish 
the first philosophy, or metaphysics, from the other sciences. We, on 
the contrary, since the enormous development of the special sciences, 
must distinguish from them not only metaphysics (the science of 
absolutely first principles) but the study of the first principles in a 
particular order (for instance, the mathematical or the physical) ; and 
the entire body of these constitutes what we call philosophy. 
2 Only metaphysics and logic constitute a universal science speci-fically 
one. 
107
. 
AN INTRODUCTION TO PHILOSOPHY 
— 
point ^ is first causes (whether absolutely first causes 
or principles, the formal object of metaphysics, or the 
first causes in a particular order, the formal object of 
the other branches of philosophy). And it follows 
that metaphysics alone deserves the name of wisdom 
absolutely speaking {simpliciter) , the remaining branches 
of philosophy only relatively or from a particular 
point of view {secundum quid) 
Conclusion I.—Philosophy is the science which 
by the natural light of reason studies the first 
causes or highest principles of all things—is, in 
other words, the science of things in their first 
causes, in so far as these belong to the natural 
order. 
The difficulty of such a science is proportionate to 
its elevation. That is why the philosopher, just 
because the object of his studies is the most sublime, 
should personally be the humblest of students, a 
humility, however, which should not prevent his 
defending, as it is his duty to do, the sovereign dignity 
of wisdom as the queen of sciences. 
The perception that the sphere of philosophy is 
1 Strictly speaking, there is no one formal object of philosophy, 
since philosophy as a whole is not simply one, but a compound of several 
distinct sciences {logic, natural philosophy, metaphysics, etc.), each 
specified by a distinct formal object {ens rationis logicum, ens mobile, ens 
in quantum ens—cf. Part II). But between the formal objects of the 
different philosophic sciences there is something analogously common 
the fact that they study, each in its own order, the highest and most 
universal causes, and treat their subject-matter from the standpoint of 
these causes. We may therefore say that the highest causes constitute 
the final object or the formal standpoint analogously common of 
philosophy taken as a whole, 
1 08
DEFINITION OF PHILOSOPHY 
universal led Descartes (seventeenth century) to regard 
philosophy as the sole science ^ of which the others 
were but parts ; Auguste Comte, on the contrary, 
and the positivists generally (nineteenth century), 
sought to absorb it in the other sciences, as being 
merely their " systematisation." It is evident that 
the cause of both errors was the failure to distinguish 
between the material and formal object of philosophy. 
Philosophy of Aristotle and St. Thomas 
Philosophy and the corpus of other sciences have the 
same material object (everything knowable). But the 
formal object of philosophy is first causes, of the other 
sciences secondary causes. 
Descartes Auguste Comte 
Philosophy absorbs the The sciences absorb 
other sciences—is the whole philosophy—there is no 
of science. philosophy. 
We said above that philosophy is a science, and 
that it attains certain knowledge. By this we would 
not be understood to claim that philosophy provides 
certain solutions for every question that can be asked 
within its domain. On many points the philosopher 
1 Descartes used the term " philosophy " in its ancient sense. For 
the ancients, as for Descartes, the word denoted the entire body of 
scientific knowledge. But the ancients divided philosophy thus under-stood 
into several distinct sciences, among which metaphysics was 
distinguished as in the fullest sense philosophy. Descartes, on the 
contrary, regarded philosophy, still understood as the entire body of 
scientific knowledge, as a science specifically one (of which meta-physics, 
physics, mechanics, medicine, and ethics were the principal 
subdi.isions). He thus recognised only one science. In our view 
philosophy is a body of sciences which owes its imity and distinction 
from the other sciences to its formal standpoint (first causes). The 
leading member of this body of sciences is metaphysics, a science 
specifically one, whose formal object is luiiversal (being qua being). 
109
AN INTRODUCTION TO PHILOSOPHY 
must be content with probable solutions, either 
because the question goes beyond the actual scope of 
his science, for example in many sections of natural 
philosophy and psychology, or because of its nature 
it admits only of a probable answer, for example the 
application of moral rules to individual cases. But 
this element of mere probability is accidental to science 
as such. And philosophy yields a greater number of 
certain conclusions, and of those many more perfect, 
namely, the conclusions of metaphysics, than any 
other purely human science. 
no
VI 
PHILOSOPHY AND THE SPECIAL SCIENCES 
W^E have now to define the relationship between 
philosophy (particularly the first philosophy or meta-physics) 
and the other sciences. 
Every science is mistress in her own house, inas-much 
as every science possesses the indispensable and 
sufficient means of attaining truth within its own 
sphere and no one is entitled to deny the truths thus 
proved. 
A science, however, or rather a scientist, may 
happen to make a mistake in its own domain. In 
such a case the science in question is no doubt com-petent 
to judge and correct itself, but it is obvious 
that a superior science has also the right to judge and 
correct it, if the mistake should contradict one of its 
own results and thus come under its jurisdiction. But 
philosophy, and especially philosophy in the highest 
sense, that is metaphysics, is the sovereign science. 
Therefore it is competent to judge every other human 
science, rejecting as false every scientific hypothesis 
which contradicts its own results. 
Take for example an hypothesis of physics which 
appears to contradict a truth of philosophy.^ Physics 
is competent to judge that hypothesis by the laws of 
1 It may, for instance, be questioned whether the law of inertia, as 
formulated since Galileo and Descartes, can be reconciled with tlic 
axiom of philosophy : quidquid movetur ab alio movetur. 
Ill
AN INTRODUCTION TO PHILOSOPHY 
physics, but philosophy is also competent to judge it 
by the principles of philosophy, determining whether 
and how far it really contradicts the philosophic truth 
in question. (If the contradiction is real, it is evident 
that the hypothesis of physics in question must be false, 
for one truth cannot contradict another. The physicist 
must therefore bow to the verdict of philosophy, revise 
his arguments and make further experiments.^) 
Let us now take a conclusion of philosophy which 
appears to contradict a truth estabhshed by physics.^ 
It is for philosophy to judge that conclusion in accord-ance 
with the principles of philosophy, to decide 
whether and how far it is really in conflict with the 
physical truth in question. But physics is incompe-tent 
to determine the question by the principles and 
data of physics. (If the contradiction is real, it is 
obvious that the alleged conclusion of philosophy is 
false, for one truth cannot contradict another. The 
philosopher will therefore bow, not indeed to the 
verdict of physics, but to the verdict of philosophy 
judging itself by means of physics, and will revise his 
arguments accordingly.) 
* It is true, no doubt, that we have actually to do, not with philosophy, 
but with philosophers, and that philosophers are fallible, and aphilosopher 
may therefore be mistaken in judging an hypothesis of physics, but 
this does not prove that he has no right to judge it. 
A physicist may therefore be justified in a particular case in 
maintaining an hypothesis of physics agidnst a philosopher who asserts 
that it contradicts a truth of philosophy. But that is because the 
evidence he possesses in support of his hypothesis convinces him that 
the philosopher is mistaken in his verdict—in other words, has not 
given it ut philosophus, as a mouthpiece of philosophy. But he would 
not therefore be justified in denying the philosopher's competence as 
such to determine the question. 
2 For example, when the philosophical doctrine oijree will appears to 
the mechanists to contradict the physical law of the conservation ofenergy. 
112
PHILOSOPHY AND THE SPECIAL SCIENCES 
Moreover, since the laws of one science are subordi-nate 
to the laws of a superior science, it clearly follows 
that it is the office of the superior science to govern 
the inferior. But since the principles of philosophy 
(the first philosophy or metaphysics) are the absolutely 
first principles of all human knowledge, they possess 
an authority over the principles of all other human 
sciences, which are in a certain sense dependent upon 
them. That is to say, philosophy (the first philosophy 
or metaphysics) governs the other sciences. 
Since the principles of philosophy (the first philo-sophy 
or metaphysics) are the absolutely first prin-ciples 
of all human knowledge, the principles or 
postulates of all human sciences are in a certain sense 
dependent upon them. 
They do not, it is true, depend directly on the 
principles of metaphysics, as the truth of a conclusion 
depends on the truth of its premisses. They are self-evident 
by the light of natural reason [principia per se 
nota) . But they are not absolutely speaking {simpliciter) 
first principles. Therefore, although they carry con-viction 
independently of metaphysics, nevertheless 
they presuppose in fact the principles of metaphysics 
and can be resolved into them. They can be known 
without an exphcit knowledge of the principles of 
metaphysics, but they could not be true, unless the 
latter were true. And in this sense they are indirectly 
subordinate to the latter. For instance, the mathe-matical 
axiom, two quantities which are equal to a third 
quantity are equal to one another^ can be resolved into the 
metaphysical axiom of which it is a special case : two 
beings identical with a third are identical with one another. 
It is for this resison that all the sciences are said to 
113
AN INTRODUCTION TO PHILOSOPHY 
be indirectly subordinate to metaphysics. Moreover, 
they are obliged on occasion to employ the universally 
valid principles of metaphysics. In this sense they 
are said to be subordinate to metaphysics in a particular 
aspect or relatively {secundum quid). 
To govern or direct anything is to prescribe its 
object or end. The sciences are not directed by philo-sophy 
to their end, in the sense that they cannot 
attain it without the aid of philosophy. Arithmetic, 
for example, has no need of philosophy to investigate 
the numerical truths which it investigates of its very 
nature. Philosophy, however, assigns the distinctive 
ends of the special sciences in the sense that it deter-mines 
speculatively the distinctive object of each, 
and what constitutes their specific unity and differen-tiation 
from the rest (classification of the sciences).^ 
And so doing it assigns the order in which they stand 
one to another. Thus all the sciences are ordered 
by wisdom : sapientis est ordinare. If therefore a 
science, or more correctly a scientist, should happen 
to lose sight of its true object by encroaching on the 
domain and rights of another science,* it is the duty 
of philosophy to redress the disorder involved. In 
this capacity philosophy governs or directs the sciences 
* A problem discussed in major logic. 
2 Such divagations and encroachments are only too frequent. 
For example, the non-EucIidcan geometries may be so treated as to 
divert mathematics from its proper end. On the other hand, since 
Descartes, mathematics has usurped the domain of all the sciences, 
and in our time physics and chemistry are constantly encroaching upon 
the domain of biology, medicine upon psychology ; while the incursions 
made by physics or biology into the province of philosophy itself are 
innumerable : for example, the pseudo-scientific theories of the non-existence 
affinal causes, the unreality of qualities, determinism, atomism, or 
the biological dogmas of fransformism and mechanism. 
114
PHILOSOPHY AND THE SPECIAL SCIENCES 
(to these distinctive ends), not by positive prescription, 
but by setting them right, if they transgress their 
boundaries. 
On the other hand, that knowledge in which the 
mind attains its ultimate good, that is to say, the 
highest knowledge, may be regarded as the common 
transcendent goal towards which all the special sciences 
converge. But this knowledge is bestowed by philo-sophy— 
the science of first causes—which in this 
capacity governs or directs the sciences in view of the 
common end to which their particular objects are 
subordinate. All the sciences are thus directed to 
wisdom. 
From all we have just said it follows that to be 
proficient in the sciences it is not necessary to be a 
philosopher or to base one's work on a philosophy ; 
neither need the scientist while engaged in his special 
task seek advice from the philosopher or attempt to 
play the philosopher himself; but " philosophy alone 
enables the man of science to understand the position 
and bearings of his special science in the sum-total of 
human knowledge " or "to acquire a notion either of 
the principles implicit in all experimental knowledge 
or the true foundations of the special sciences." ^ It 
follows, further, that a period in the history of human 
culture in which philosophy is not allowed her rightful 
suzerainty over the sciences as scientia rectrix ° inevitably 
ends in a condition of intellectual chaos and a general 
weakening of the reasoning faculty. 
Descartes, just because he absorbed all the sciences 
1 T. Richard, Philosophic du raisonnement dans la science d'apris Saint 
Thomas, p. 14.. 
' St. Thomas, In Metaph., Inlroduction.
AN INTRODUCTION TO PHILOSOPHY 
in philosophy, and regarded science as absolutely 
and without quaUfication one, believed that the prin-ciples 
of all the sciences depend directly on the principles 
of the first philosophy (metaphysics). In consequence 
he held that the study of the sciences and of philosophy 
itself must begin with metaphysics, that is to say, 
with the coping-stone of the entire edifice. 
The contrary error is the belief that the principles 
of science are absolutely independent of the principles of 
philosophy. There is therefore no place for a scientia 
rectrix, and the sciences are no longer a structure built 
on a definite plan, but a formless agglomeration. It 
is surprising that Auguste Gomte, who wished to 
reduce philosophy to a mere systematisation of the 
special sciences, failed to see that this very function of 
classifying and systematising the sciences (in what he 
terms their objective synthesis) is only possible if philo-sophy 
is a distinct science of a higher order on which 
the others are in a certain respect dependent.^ 
Philosophy of Aristotle and St. Tfiomas 
The principles of the special sciences are subordinate 
to the principles of philosophy, but only indirectly. 
Philosophy therefore governs the other sciences, but its 
government is such that it may be termed constitutional. 
(The special sciences are autonomous.) The study of 
the first philosophy (metaphysics) should be undertaken, 
not at the beginning, but at the end of intellectual research. 
Philosophy of Philosophy of those who reject 
Descartes Philosophy 
The principles of the The principles of the 
special sciences arc directly special sciences are not 
subordinate to those of subordinate to the prin- 
* For Gomte, indeed, sociology takes the position of scientia rectrix, 
only, however, as ordering the sciences in reference to the human 
subject, not in themselves. {Subjective synthesis.) 
1x6
PHILOSOPHY AND THE SPECIAL SCIENCES 
philosophy. The latter ciples of any science of a 
therefore exercises over the higher order. These sci-other 
sciences a govern- ences therefore are in no 
ment v^hich may be termed sense governed but are in 
despotic. a condition which may be 
The study of the first termed anarchy. There is 
philosophy (metaphysics) no supreme science or first 
should be undertaken at philosophy (metaphysics), 
the beginning of intellec-tual 
research. 
Finally, if a science bases its demonstrations on 
certain postulates or data, which it can neither explain 
nor defend, there must be a superior science whose 
function it is to defend these postulates or data. In 
this sense the science of architecture defends that of 
building. It is, however, obvious that every science, 
except the highest, bases its demonstrations on postu-lates 
or data it is incapable of explaining or defending. 
For instance, mathematics does not inquire what is 
the nature of quantity, number, or extension, nor 
physics what is the nature of matter. And if an 
objector should deny that the sensible world exists, that 
two quantities equal to a third are equal to one another, or 
that space has three dimensions, neither physics nor 
mathematics can refute his objection, since they on 
the contrary assume these postulates or data. There-fore 
it must be the function of philosophy (the first 
philosophy or metaphysics) to defend against every 
possible objection the postulates of all the human 
sciences. 
It is from common sense, or from the natural 
evidence of the intellect and experience, that the 
sciences derive their postulates. This is no doubt their 
sufficient warrant to build on these postulates, but it 
is insufficient to safeguard and protect them against 
117
AN INTRODUCTION TO PHILOSOPHY 
errors which call the postulates in question. And it is 
also insufficient to provide for the perfecting stability 
and essential needs of human knowledge. Human 
knowledge would remain excessively imperfect and 
weak, and would fail to reach its final end, if the 
postulates of the sciences were not scientifically 
explained, discussed, and defended. 
Philosophy, therefore, and particularly the first 
philosophy or metaphysics, because it is wisdom and 
the supreme science, judges, governs, and defends the 
other sciences. But the ruler is certainly not dependent 
upon those whom he governs. We therefore conclude 
that philosophy is independent of the inferior sciences, 
or at any rate depends on them only in the sense that 
a superior, when he is not strong enough to be self-sufficient, 
depends on the servants or instruments 
which he employs. It was for this reason that 
Aristotle regarded philosophy as the science pre-eminently 
free. 
Philosophy appeals to the facts, the data of experi-ence. 
To obtain the necessary materials it uses as 
instruments the truths provided by the evidence of 
the senses and the conclusions proved by the sciences. 
It depends on both, as a superior who cannot do his 
own work depends on the servants he employs. 
A dependence of this kind is a purely material 
dependence, since the superior depends on the inferior 
to be served by him, not to do him service. He 
therefore judges by his own light of whatever his 
servants bring him to supply his needs. For example, 
one of the most successful students of bees, Francois 
1x8
PHILOSOPHY AND THE SPECIAL SCIENCES 
Huber, who was blind, interpreted by the light of his 
intellect the facts seen by his servants' eyes. 
But further, this purely material dependence of 
philosophy, though absolutely necessary in respect of 
the evidence of the senses, is relative and contingent 
in respect to the conclusions of the sciences. It is in 
fact from the evidence of the senses that philosophy 
derives the fundamental principles which—interpreted 
by its own Hght—it employs as premisses in its demon-strations 
and as the means to prove its special truths. 
For instance, the truth, perceived by the senses and 
interpreted by the light of philosophy, there is motion 
in the universe, served Aristotle as the premiss from 
which he proved that being is divided into act and 
potentiality, and that there is a first mover that is 
pure act (God). It is obvious that philosophy is 
absolutely unable to dispense with data of this kind, 
and that the data thus employed as premisses must 
be absolutely true. Otherwise the conclusions which 
philosophy deduces from them would be uncertain. 
But it is otherwise with the propositions and conclusions 
of the sciences. No doubt these conclusions, if true, 
contribute to the store of materials utilised by philo-sophy— 
but philosophy (and particularly the first 
philosophy or metaphysics) is under no necessity to 
employ them, indeed ought not to employ them to 
establish its own conclusions, at any rate not its certain 
conclusions, though it may make use of them as 
confirmatory evidence. It must certainly have at 
its disposal some scientific conclusions, indeed as 
plentiful a supply as possible, because it cannot 
develop its principles satisfactorily until it sees them 
embodied, so to speak, in concrete examples which 
119
AN INTRODUCTION TO PHILOSOPHY 
the senses can perceive. But it does not need one 
particular scientific proposition rather than any other, 
provided, that is to say, that, true to its own nature 
and maintaining the Hberty due to a superior science, 
it draws its proofs from its own principles and from 
the fundamental truths supplied by the evidence of 
the senses and not from the conclusions supplied by 
the sciences. These latter, in fact, should not be 
premisses but simply illustrations which assist philo-sophy 
to attain its own truths. A sound philosophy 
can therefore dispense with the particular system of 
scientific explanations of which it makes use in accord-ance 
with the state of science at a particular epoch, 
and if that system were one day proved false the 
truth of that philosophy would not be affected. Only 
its language and the sensible illustrations with which 
it clothed its truths would require modification. 
These remarks are important. They show how the 
datum of experience on which philosophy is primarily 
based suflSces for the requirements of a supreme and 
universal science. This datum is provided by an 
instrument—the evidence of the senses—earlier than 
scientific observation, infinitely more certain than the 
inductions of the sciences, and placed by nature at 
the disposal of every man, and consists of truths so 
simple that they are universally and absolutely valid, 
so immediate and evident that their certziinty exceeds 
that of the best established scientific conclusions.^ 
^ To this fundamental datum we may add—but as secondary matter 
and at times valuable confirmation—the facts of a more special 
description discovered, controlled, and measured by the observations 
and experiments of science. We should bear in mind that the 
absolutely evident truths which constitute the primitive and funda-mental 
datum of pliilosophy must be carefully distinguished from 
120
PHILOSOPHY AND THE SPECIAL SCIENCES 
From what has been said we can also understand 
why the purely scientific mistakes to be found in 
older statements of AristoteUan and Thomistic philo-sophy, 
statements which inevitably bear the stamp 
of the scientific beliefs of their period, do nothing to 
discredit the truth of that philosophy. For no philo-sophy 
has observed more faithfully than that of 
Aristotle and St. Thomas the laws of thought which 
guarantee its purity and freedom. 
On the other hand philosophy, though distinct 
from the special sciences, is not unrelated to or 
isolated from them. On the contrary, it possesses the 
duty to exercise its office as scientia rectrix by con-stantly 
throwing its light on the discoveries and 
hypotheses, the unceasing activity and the development 
of the sciences, and one of the most essential requisites 
for its life and progress in the world is to maintain an 
intimate contact with the lower branches of study 
whose data it interprets and renders fruitful. 
To the extent to which philosophy thus concerns 
itself with interpreting by the aid of its own truths the 
facts or hypotheses which positive science regards as 
proved, the errors or lacunae of positive science may 
introduce accidentally into a true system of pliilosophy 
elements of error which are, so to speak, the token and 
price of the human development of philosophy—but 
certain interpretations of experience drawn from unscientific 
observation which are nothing more than pseudo-scientific statements. 
If, for example, in natural philosophy, to prove the reality of 
substantial change, it were argued that whereas water is a liquid body, 
hydrogen and oxygen are gaseous bodies, the argument would rest, not 
on a truth attested by the senses, but on a scientific error, for in reality 
the same inorganic bodies are found in the three states solid, liquid, 
and gaseous. Obviously an adequate scientific training helps the 
philosopher to avoid pitfalls of this kind. 
121
AN INTRODUCTION TO PHILOSOPHY 
they can only falsify a philosophy itself to the extent 
to which it is untrue to its own nature and enslaves 
itself to the lower branches of study.^ 
It is clear from everything which has been said that 
the nature and needs of philosophy make it incumbent 
upon the philosopher to keep himself as fully ac-quainted 
as he can with the scientific knowledge of his 
period, provided, however, that he preserves intact 
the freedom of philosophic truth. For though the 
philosopher as such need not use the affirmations of 
the special sciences to establish his own truths, he 
ought to make use of them, (i) to illustrate aptly his 
principles, (ii) to confirm his conclusions, (iii) to 
interpret, throw Hght upon, and assimilate, the 
assured results of the sciences so far as questions of 
philosophy are involved. And finally he should use 
the affirmations of science (iv) to refute objections and 
errors which claim support from its results. 
From yet another point of view the study of the 
sciences is necessary for the philosopher : his own 
education must of necessity, owing to the very condi-tions 
ofhuman nature, be a progress from the imperfect 
to the perfect, so that before undertaking the study of 
wisdom he should undergo the training of the sciences.* 
1 The '* crime " of the decadent Scholastics of the sixteenth and 
seventeenth centuries was that they beUeved, and made others believe, 
that the philosophy of Aristotle and St. Thomas was in this sense bound 
up with the mistakes of ancient science, of which it is in reality wholly 
independent. 
2 The philosopher also requires a scientific training to be in a position 
to distinguish readily between the primary evidence of experience and 
certain popular but really pseudo-scientific interpretations of experience, 
such, for example, as the hypothesis of the sun's motion around the 
earth or the over-hasty belief that a particular inorganic body is 
essentially liquid and another essentially solid or gaseous. 
122
PHILOSOPHY AND THE SPECIAL SCIENCES 
It is not therefore surprising that all the great 
philosophers have been thoroughly acquainted with 
contemporary science. Some have even been great 
scientists (for example Aristotle, Albertus Magnus, and 
Leibniz), and several scientific discoveries of the first 
magnitude has been made by philosophers, for instarrce 
the mathematical discoveries of Pythagoras, Descartes, 
and Leibniz. 
In this connection we may observe that a profound 
and practical knowledge of a single science with which 
the student is directly acquainted contributes far more 
to a philosophic training than a superficial and second-hand 
knowledge of a large number. Though owing 
to the degree to which specialisation has been carried 
in modern times he cannot hope ever to possess that 
complete knowledge of all the sciences which is 
possessed by the scientist in his particular department, 
the philosopher should nevertheless aim at acquiring 
a sufficiently thorough knowledge of the entire body 
of the sciences, an ideal in itself not beyond the bounds 
of possibility, as is proved by the example of several 
powerful minds. 
Conclusion II.—Philosophy is the highest of all 
branches of human knowledge and is in the true 
sense wisdom. The other (human) sciences are 
subject to philosophy, in the sense that it judges 
and governs them and defends their postulates. 
Philosophy on the other hand is free in relation 
to the sciences, and only depends on them as the 
instruments which it employs. 
123
VII 
PHILOSOPHY AND THEOLOGY 
— 
Philosophy is the highest of the human sciences, that 
is, of the sciences which know things by the natural 
light of reason. But there is a science above it. For 
if there be a science whch is a participation by man 
of the knowledge proper to God himself, obviously 
that science will be superior to the highest human 
science. Such a science, however, exists ; it is 
theology. 
The word theology means the science of God. The 
science or knowledge of God which we can attain 
naturally by the unassisted powers of reason, and 
which enables us to know God by means of creatures 
as the author of the natural order, is a philosophic 
science—the supreme department of metaphysics 
and is known as theodicy or natural theology. The know-ledge 
or science of God which is unattainable naturally 
by the unassisted powers of reason, and is possible 
only if God has informed men about himself by a 
revelation from which our reason, enlightened by 
faith, subsequently draws the implicit conclusions, is 
supernatural theology or simply theology. It is of this 
science that we are now speaking. 
Its object is something wholly inaccessible to the 
natural apprehension of any creature whatsoever, 
namely, God known in himself, in his own divine life, 
or in technical language sub ratione Deitatis, not, as in 
124
PHILOSOPIiY AND THEOLOGY 
natural theology, God as the first cause of creatures 
and the author of the natural order. And all theo-logical 
knowledge is knowledge in terms of God thus 
apprehended, whereas all metaphysical knowledge, 
including the metaphysical knowledge of God, is 
knowledge in terms of being in general. 
The premisses of theology are the truths formally 
revealed by God {dogmas or articles of faith) , and its 
primary criterion of truth the authority of God who 
reveals it. 
Its light is no longer the more natural light of 
reason, but the Ught of reason illuminated by faith, 
virtual revelation in the language of theology, that is to 
say, revelation in so far as it implicitly (virtually) 
contains whatever conclusions reason can draw 
from it. 
Alike by the sublimity of its object, the certainty 
of its premisses, and the excellence of its light, theology 
is above all merely human sciences. And although it 
is unable to perceive the truth of its premisses, which 
the theologian beHeves, whereas the premisses of 
philosophy are seen by the philosopher, it is never-theless 
a science superior to philosophy. Though, as 
St. Thomas points out, the argument from authority 
is the weakest of all, where human authority is con-cerned, 
the argument from the authority of God, 
the revealer, is more solid and powerful than any 
other.^ 
And finally as the object of theology is he who is 
above all causes, it claims with a far better title than 
* Licet locus ab auctoritate, quae fundaiur super ratione humana, sit 
infirmissimus, locus tarrum ab auctoritate quae fundatur super revelatione divina 
est efficacissimus. St. Thomas, Sum. Theol. i, q. i, a 8, ad 2.
AN INTRODUCTION TO PHILOSOPHY 
metaphysics the name of wisdom. It is wisdom par 
excellence.^ What relations, then, must obtain between 
philosophy and theology ? 
As the superior science, theology jW^^j philosophy 
in the same sense that philosophy judges the sciences.^ 
It therefore exercises in respect of the latter a function 
of guidance or government, though a negative govern-ment, 
which consists in rejecting as false any philosophic 
affirmation which contradicts a theological truth. In 
this sense theology controls and exercises jurisdiction 
over the conclusions maintained by philosophers. 
Thej&r^mfjj^j^ of philosophy, however, are independent 
of theology, being those primary truths which are 
self-evident to the understanding, whereas the premisses 
of theology are the truths revealed by God. The 
premisses of philosophy are self-supported and are not 
derived from those of theology. Similarly the light 
by which philosophy knows its object is independent 
* Theology is theoretical wisdom, par excellence the wisdom which 
knows God by the intellect and its ideas, that is to say, by the normal 
processes of human knowledge. There is another wisdom of a still 
higher order which is a gift of the Holy Spirit, and enables lis to know 
God experimentally and by means of charity. It enables xis to judge 
of divine things instinctively, as the virtuous man judges of virtue 
{per modum inclinationis) , not scientifically as the moralist judges of 
virtue {per modum cognitionis) . Cf. St. Thomas, Sum. Theol. i, q. i, a 6, 
a</ 3. 
2 See above, pp. 1 1 1 sqq. The philosopher and the scientist are never 
entitled to deny the rights which theology possesses over philosophy 
and the sciences. They may, however, be justified in rejecting in a 
particular instance, not indeed the authority of the Church, but the 
judgment of an individual theologian, since the individual theologian 
does not necessarily speak as the mouthpiece of theology, and may 
therefore be mistaken. 
126
PHILOSOPHY AND THEOLOGY 
of theology, since its light is the light of reason, which 
is its own guarantee.^ For these reasons philosophy 
is not positively governed by theology,^ nor has it 
any need of theology to defend its premisses (whereas 
it defends those of the other sciences). It develops 
its principles autonomously within its own sphere, 
though subject to the external control and negative 
regulation of theology. 
It is therefore plain that philosophy and theology 
are entirely distinct, and that it would be as absurd 
for a philosopher to invoke the authority of revelation 
to prove a philosophical thesis as for a geometrician 
to attempt to prove a theorem by the aid of physics, 
for example, by weighing the figures he is comparing. 
But if pliilosophy and theology are entirely distinct, 
they are not therefore unrelated, and although philo-sophy 
is of all the human sciences pre-eminently the 
free science, in the sense that it proceeds by means 
of premisses and laws which depend on no science 
superior to itself, its freedom—that is, its freedom to 
err—is Umited in so far as it is subject to theology, 
which controls it externally. 
I In the seventeenth century the Cartesian reform 
^ This light is its own evidence and in philosophy is sufficient of 
itself. But this does not prevent it serving also—in theology, however, 
not in philosophy—as the instrument of a superior light ; neither, of 
course, does it imply that human reason is not subordinate in its very 
principles to the First Intellect. 
2 Theology can turn the investigations of philosophy in one direction 
rather than in another, in which case it may be said to regulate 
philosophy positively by accident {per accidens) . But absolutely speaking 
theology can regulate philosophy only negatively, as has been explained 
above. Positively it does not regulate it either directly, by furnishing 
its proofs (as faith for apologetics), or indirectly, by classifying its 
divisions (as philosophy itself classifies the sciences). 
127
AN INTRODUCTION TO PHILOSOPHY 
resulted in the severance of philosophy from theology,^ 
the refusal to recognise the rightful control of theology 
and its function as a negative rule in respect of philo-sophy. 
This was tantamount to denying that theology 
is a science, or anything more than a mere practical 
discipline, and to claiming that philosophy, or human 
wisdom, is the absolutely sovereign science, which 
admits no other superior to itself. Thus, in spite of 
the religious beliefs of Descartes himself, Cartesianism 
introduced the principle of rationalist philosophy, 
which denies God the right to make known by revela-tion 
truths which exceed the natural scope of reason. 
For if God has indeed revealed truths of this kind, 
human reason enlightened by faith will inevitably 
employ them as premisses from which to obtain further 
knowledge and thus form a science, theology. And if 
theology is a science, it must exercise in respect of 
philosophy the function of a negative rule, since the 
same proposition cannot be true in philosophy, false 
in theology.^ 
* It may, it is true, be replied that Descartes's intention was simply 
to emancipate philosophy from the authority of a particular theological 
system—Scholasticism—which he regarded as worthless, because it 
took its philosophic or metaphysical principles from Aristotle. 
In reality, however, it was with theology itself that he broke, when he 
broke with Scholasticism, which is the traditional theology of the 
Church. And moreover his conception of science implied the denial 
of his scientific value of theology. In any case the result of his reform 
was the assertion of the absolute independence of philosophy in relation 
to theology. {Cf. Blondel, " Le Christianisme de Descartes." Revtie 
de Mitaph. et de Morale, 1896.) 
2 The theory of a double truth, by which the same thing may be 
true in philosophy, but fake in theology, was invented by the mediaeval 
Averroists, who sought in this way to evade the censures of the Church. 
In various forms it has been revived in modem times by all who, like the 
modernists, wish to keep the name of Catholics while freely professing 
in philosophy opinions destructive of some particular dogmatic truth. 
128
PHILOSOPHY AND THEOLOGY 
On the other hand, philosophy renders to theology 
services of the greatest value where it is employed by 
the latter. For in fact theology employs in its 
demonstrations truths proved by philosophy. Philo-sophy 
thus becomes the instrument of theology, and 
it is in this respect and in so far as it serves theological 
argument that it is called ancilla theologiae. In itself, 
however, and when it is proving its own conclusions, 
it is not a bond-servant but free, subject only to the 
external control and negative ruUng of theology. 
As was shown above, philosophy is from the very 
nature of things obhged to employ as an instrument 
the evidence of the senses, and even, in a certain 
fashion, the conclusions of the special sciences. Theo-logy* 
considered in itself as a science subordinate to 
the knowledge of God and the blessed, is not in this 
way obliged to make use of philosophy, but is absolutely 
independent. 
In practice, however, on account of the nature of 
its possessor, that is to say, on account of the weakness 
of the human understanding, which can reason about 
the things of God only by analogy with creatures, it 
cannot be developed without the assistance of philo-sophy. 
But the theologian does not stand in the same 
relation to philosophy as the philosopher to the 
sciences.^ We have seen above that the philosopher 
^ This distinction between the relationship of theology to philosophy 
and that of philosophy to the special sciences derives from the fact that, 
since theology is a participation of the divine wisdom, the human 
subject is too weak for its unaided study and to draw conclusions from 
it is comj)elled to employ as premisses conclusions established by an 
inferior discipline. 
Since, however, philosophy is a human wisdom, at which reason can 
129
AN INTRODUCTION TO PHILOSOPHY 
should employ the propositions or conclusions which 
he borrows from the sciences, not to estabUsh his own 
conclusions (at any rate not conclusions for which 
metaphysical certainty is claimed), but merely to 
illustrate his principles, and therefore that the truth 
of a metaphysical system does not depend on the 
truth of the scientific material it employs. The 
theologian, on the contrary, makes use at every turn 
of philosophic propositions to prove his own conclu-sions. 
Therefore a system of theology could not 
possibly be true if the metaphysics which it employed 
were false. It is indeed an absolute necessity tiiat 
the theologian should have at his disposal a true 
philosophy in conformity with the common sense of 
mankind. 
Philosophy taken in itself normally precedes theo-logy. 
Certain fundamental truths of the natural order 
are indeed what we may term the introduction to the 
faith {praeambula Jidei). These truths, which are 
naturally known to all men by the light of common 
sense, are known and proved scientifically by philo-sophy. 
Theology', being the science of faith, pre-supposes 
the philosophical knowledge of these same 
truths. 
Philosophy considered as the instrument of theology 
serves the latter, principally in three ways. In the 
first place theology employs philosophy to prove the 
truths which support the foundations of the faith in 
that department of theology which is termed apolo-arrive, 
though with difficulty, by its unassisted natviral power, the 
human mind should be able to draw from it certain conclusions 
(especially metaphysically certain conclusions) without employing as 
premisses the conclusions of sciences to which it is superior in d^[nity 
and in certainty. 
130
PHILOSOPHY AND THEOLOGY 
1 
getics,^ which shows, for example, how miracles prove 
the divine mission of the Church ; secondarily to 
impart some notion of the mysteries of faith by the 
aid of analogies drawn from creatures—as for instance 
when theology uses the philosophic conception of 
verbum mentale, the mental word,' to illustrate the 
dogma of the Trinity ; and finally to refute the 
adversaries of the faith—as when theology shows by 
means of the philosophic theory of quantity * that the 
mystery of the Eucharist is in no way opposed to 
reason. 
We must not forget that, if philosophy serves 
theology, it receives in return valuable assistance 
firom the latter. 
In the first place, so far as it is of its nature subject 
to the external control and negative ruhng of theology, 
it is protected from a host of errors ; and if its freedom 
to err is thus restricted, its freedom to attain truth is 
correspondingly safeguarded.* 
In the second place, in so far as it is the instrument 
of theology, it is led to define more precisely and 
with more subtle refinements important concepts and 
theories which, left to itself, it would be in danger of 
neglecting. For example, it was under the influence 
See Garrigou-Lagrange, De Revelatione, i, 2, 19 18. 
2 A theory studied in psychology. 
' An explanation given by cosmology. 
* Unassisted reason can indeed avoid error on any particular 
point whatsoever within the sphere of philosophy, but in view of the 
weakness of human nattire it is unable without the assistance of grace 
to avoid error on some point or other ; that is to say, without a special 
grace or the negative control of revelation and theology it cannot 
achieve a perfect human wisdom. {Cf. St. Thomas, Sum. Theol., i, q. i, 
a. I ; Sum. contra Cent., i, 4 ; Garrigou-Lagrange, De Rev., i, pp. 41 
sqq.)
AN INTRODUCTION TO PHILOSOPHY 
of theology that Thomism elaborated the theory of 
nature and personality, and perfected the theory of the 
habitus, habits, etc. 
Conclusion III.—Theology, or the science of 
God so far as He has been made known to us by 
revelation, is superior to philosophy. Philosophy 
is subject to it, neither in its premisses nor in its 
method, but in its conclusions, over which 
theology exercises a control, thereby constituting 
itself a negative rule of philosophy. 
132
VIII 
PHILOSOPHY AND COMMON SENSE 
Before wc know things with a scientific or perfect 
knowledge by reflecting upon them and by their 
causes, we know them imperfectly {unscientific know-ledge, 
the knowledge of everyday life). We must 
remember that we are obliged not only to begin with 
this unscientific knowledge of everyday life ; we must 
be content with it to the end, improving it more or 
less by study and reading, in that enormous number 
of cases where science in the strict sense is unattain-able. 
For, so far as the knowledge of secondary causes is 
concerned, no man can possibly attain, with the per-fection 
required of the genuine scientist, universal 
knowledge ; in other words, he cannot specialise in 
all branches of science, a contradiction in terms. He 
is fortunate, indeed, if he can make himself master of 
a single science. For all the others he must be satisfied 
with a knowledge which, however enriched and 
improved it may be in the case of what is known as 
a cultivated man, that is to say, a man well acquainted 
with the scientific knowledge of other people, is always 
inferior to science in the strict sense. But in the 
domain of first causes, the science of all things is within 
a man's grasp, for it is precisely the distinguishing 
character of the science called philosophy to know 
133
AN INTRODUCTION TO PHILOSOPHY 
all things by their first causes/ and it is to the philo-sopher 
or the sage, the wise man, that we have the 
right to apply Leonardo da Vinci's aphorism : facile 
cosa e farsi universale ; it is easy for a man to make 
himself universal. 
Ordinary knowledge- consists for the most part of 
mere opinions or beliefs, more or less well founded. 
But it implies a solid kernel of genuine certainties in 
which the philosopher recognises in the first place data 
of the senses (for example, that bodies possess length, 
breadth, and height), secondly, self-evident axioms (for 
example, the whole is greater than the part, every event has 
a cause, etc.), and thirdly, consequences immediately 
deducible from these axioms (proximate conclusions). 
These certainties which arise spontaneously in the 
mind when we first come to the use of reason are thus 
the work of nature in us, and may therefore be called 
an endowment of nature * as proceeding from the 
natural perception, consent, instinct, or natural sense 
of the intellect. Since their source is human nature 
itself, they will be found in all men aUke ; in other 
words, they are common to all men. They may 
therefore be said to belong to the common perception, 
consent, or instinct, or to the common sense of mankind. 
The great truths without which man's moral life is 
impossible—for example, knowledge of God's existence, 
the freedom of the will, etc.—belong to this domain of 
^ It is therefore obvious what a stupendous delusion is involved in 
the positivist view of philosophy. Were philosophy merely the 
co-ordination or systematisation of the sciences, its attainment would 
presuppose a perfect mastery of all the sciences, that is to say, specialisa-tion 
in every science, which amounts to saying that philosophy is beyond 
the reach of man. 
* Kleutgen, La philosophic scolastique, i, p. 439.
PHILOSOPHY AND COMMON SENSE 
common sense, as consequences immediately deducible 
(proximate conclusions) from primary data appre-hended 
by observation and first principles apprehended 
by the intellect. All men, unless spoiled by a faulty 
education or by some intellectual vice, possess a 
natural certainty of these truths. But those whose 
understanding has never been cultivated are not 
able to give any account or at least any satisfactory 
account of their convictions ; that is to say, they 
cannot explain why they possess them. 
These certainties of common sense, conclusions of 
an implicit reasoning, are as well founded as the 
certainties of science. But their possessor has no 
knowledge, or an imperfect knowledge, of the grounds 
on which he bases them. They are therefore imperfect 
not in their value as truth but in the mode or condition 
under which they exist in the mind. 
Of the self-evident truths {the whole is greater than 
the part, every event has a cause, etc.) which are the object 
ofwhat is termed the understanding ofprinciples, and whose 
certainty is superior to that of any conclusion of science, 
common sense possesses a knowledge whose mode is 
equally imperfect, because it is confused and implicit. 
Common sense therefore may be regarded as the 
natural and primitive judgment of human reason, 
infalUble, but imperfect in its mode. 
The wholly spontaneous character of common 
sense, and its inability to give an account of its con-victions, 
have led certain philosophers to regard it 
as a special faculty purely instinctive and unrelated 
to the intellect (the Scottish school, end of eighteenth 
and beginning of nineteenth century ; Reid, Dugald 
Stewart, and in France, JoufFroy), or as a sentiment 
135
AN INTRODUCTION TO PHILOSOPHY 
. 
distinct from and superior to reason (the intuitive or 
sentimentalist school ; for instance, Rousseau, Jacobi, 
and in our own time Bergson). But in that case it 
would necessarily be blind, for we possess no other 
light than that of the intellect or reason. The light 
of common sense is fundamentally the same light as 
that of science, that is to say, the natural light of the 
intellect. But in common sense this light does not return 
upon itself by critical reflection, and is not perfected by 
what we shall learn to know as a scientific habit {habitus) 
We must now define the relations which obtain 
between philosophy and common sense. 
Philosophy cannot, as the Scottish school main-tained, 
be founded on the authority of common sense 
understood simply as the common consent or universal 
witness of mankind, or as an instinct which in fact 
compels our assent. For it is in fact founded on 
evidence, not on authority of any kind. 
But if by common sense we understand only the 
immediate apprehension of self-evident first principles, which 
is one of its constituents, we may say with truth that 
it is the source of the whole of philosophy. For the 
premisses of philosophy are indeed the evident axioms 
which in virtue of its natural constitution implant in 
the mind its primary certainties. 
It is important to be quite clear that, if philosophy 
finds its premisses already enunciated by common 
sense, it accepts them not because they are enunciated 
by common sense, or on the authority of common 
sense understood as the universal consent or common 
instinct of mankind, but entirely and solely on the 
authority of the evidence. 
136
PHILOSOPHY AND COMMON SENSE 
Finally, if we take into account the entire body of 
truths (premisses and conclusions) known by common 
sense with certainty but in an imperfect mode, we 
must conclude that philosophy is superior to common 
sense, as the perfect stage of anything (in this case the 
scientific stage of knowledge) is superior to the im-perfect 
or rudimentary stage of the same thing (in 
this case the pre-scientific stage of the same knowledge, 
which is yet true and certain at both stages). 
If in common sense we consider not the conclusions 
which it reaches but the premisses alone, it is still 
inferior to philosophy in respect of its mode of know-ledge, 
but superior alike to philosophy and to all the 
sciences in respect oi its object and of the light in which 
it knows. For, as we have said above, philosophy and 
all the sciences are ultimately founded on the natural 
evidence of first principles (to which philosophy 
returns—in criticism—to study them scientifically, 
whereas the other sciences are content to accept them 
from nature). 
Philosophy studies scientifically the three categories 
of truths to which common sense bears instructive 
witness : (i) the truths of fact which represent the 
evidence of the senses ; (ii) the self-evident first prin-ciples 
of the understanding, in as much as it clears up 
their meaning by critical reflection and defends them 
rationally ; (iii) the consequences immediately de-ducible 
(proximate conclusions) from these first 
principles, inasmuch as it provides a rational proof 
of them. And, further, where common sense yields 
to the mere opinions of popular belief, philosophy 
continues to extend indefinitely the domain of scientific 
certainty. Thus philosophy justifies and continues 
137
AN INTRODUCTION TO PHILOSOPHY 
common sense, as, for instance, the art of poetry justi-fies 
and continues the natural rhythms of language. 
It is also the province of philosophy to decide what 
are the genuine certainties affirmed by common sense, 
and what is their true significance ; a function which 
common sense is incapable of performing, for the very 
reason that it does not understand, or does not under-stand 
clearly, the grounds ofits knowledge. In this sense 
philosophy controls common sense, as, for example, the 
art of poetry controls the natural rhythms of language. 
Nevertheless, common sense has the right and duty 
to reject any philosophic teaching which denies a 
truth of which it possesses natural certainty, as the 
inferior has the right and duty to oppose a superior 
who acts in a manner evidently unjust. For as soon 
as a truth becomes known to us, by whatever channel, 
it is a sin not to accept it. Common sense may there-fore 
accidentally judge philosophy. 
It is related of Diogenes that when Zeno the Eleatic 
was arguing in his presence against the possibility of 
motion, his sole reply was to get up and walk. Simi-larly, 
when Descartes taught that motion is relative 
or " reciprocal," so that it makes no difference whether 
you say the moving object is moving towards the 
goal or the goal towards the moving object, the 
English philosopher Henry More retorted that when 
a man runs towards a goal panting and tiring himself,^ 
he has no doubt which of the two, the moving object 
or the goal, is in motion. 
These protests of common sense bzised on the 
evidence of the senses were perfectly justified. It 
must, however, be added that they were insufficient 
1 Letter of March 5, 1649. 
138
PHILOSOPHY AND COMMON SENSE 
—not indeed to confute the respective theses of Zeno 
and Descartes but to confute them as errors in philo-sophy. 
That would have demanded a philosophic 
refutation of the arguments adduced by these philo-sophers, 
and explanations showing why and at what 
point they went wrong. 
It must be observed that though in itself and in 
order to establish its demonstrations philosophy does 
not depend upon the authority of common sense, 
understood as the universal consent or common instinct 
of mankind, nevertheless it is dependent upon it in a 
certain sense {materially, or in respect of the subject), 
in its origin as a human activity and in its development 
in the mind of philosophers. From this point of view 
philosophy may be compared to a building, and the 
great pre-scientific conclusions of common sense (the 
existence of God, the freedom of the will, etc.) to the 
scaffolding which nature has erected beforehand. 
Once the edifice has been completed it supports itself 
on its rock-bed, the natural self-evidence of its first 
principles, and has no need of scaffolding. But 
without the scaffolding it could not have been built. 
It is now evident how unreasonable that philosophy 
is, which priding itself on its scientific knowledge of 
things despises common sense a priori and on principle, 
and cuts itself off from its natural convictions. 
Descartes (who in other respects and in his very 
conception of science concedes too much to common 
sense) began this divorce, on the one hand, by 
admitting as the only certain truths those scientifically 
established, thus denying the intrinsic value of the 
convictions of common sense, and on the other hand, 
by professing as part of his system several doctrines 
139
AN INTRODUCTION TO PHILOSOPHY 
incompatible with those convictions. His disciple 
Malebranche, and above all the critical philosophers 
of the Kantian school, as also certain modernist 
philosophers, have carried this tendency to its extreme, 
until for some of these philosophers it is suflSdent that 
a proposition should be acceptable to common sense 
for it to be questioned or denied by science, which 
would be contaminated by the " credulity " of the 
common herd, unless it taught the contrary of what 
mankind at large believes to be true. 
Yet the greater the natural strength of a man's 
intelligence, the stronger should be his grasp of these 
natural certainties. He therefore who professes to 
condemn common sense shows not the strength but 
the weakness of his understanding. 
It is now obvious that in its attitude to common 
sense, as in its solution of the majority of the great 
philosophic problems, Thomism keeps the golden mean 
between two opposing errors Hke a mountain summit 
between two valleys. 
Philosophy of Aristotle and St. Thomas 
The convictions of common sense are valid, and 
science is untrue to itself if it rejects them. But the 
basis of philosophy is the natural witness of the intellect, 
not the authority of common sense. 
Rationalist, Critical, and 
Scottish School Modernist Schools 
Not only are the convic- Not only is the authority 
tions of common sense of common sense incapable 
valid, but the authority of of furnishing the basis of 
common sense imposing philosophy, but the con-itself 
as a blind instinct on victions of common sense 
the mind is the foundation are destitute of any specu-on 
which philosophy should lative value, 
be based. 
140
PHILOSOPHY AND COMMON SENSE 
From all that has been said it is evident what an 
important part the certainties of common sense play 
as an introduction to philosophy. Those who are 
beginning the study of philosophy and about to 
acquaint themselves with the most recent problems, 
and even perhaps the most misleading systems, ought 
to repose an absolute trust in the convictions of 
common sense of which they find their minds 
already possessed, for they will help them to rise to 
a higher and more perfect knowledge, conclusions 
scientifically established. 
Conclusion IV.—Philosophy is not based upon 
the authority of conunon sense understood as 
the universal consent or common instinct of 
mankind ; it is nevertheless derived fi-om common 
sense considered as the imderstanding of self-evident 
first principles. 
It is superior to common sense as the perfect 
or " scientific " stage of knowledge is superior 
to the imperfect or ordinary stage of the same 
knowledge. Neveitheless philosophy may be 
accidentally judged by common sense. 
For the purposes of this present outline we need 
only add that philosophy is not constructed a priori on 
the basis of some particular fact selected by the 
philosopher (Descartes's cogito), or principle arbitrarily 
laid down by him (Spinoza's substance, Fichte's pure 
ego, SchelUng's absolute, Hegel's idea) whose conse-quences 
he ingeniously develops. Its formal principles 
are the first principles apprehended in the concept of 
being, whose cogency consists wholly in their evidence 
141
AN INTRODUCTION TO PHILOSOPHY 
for the intellect,^ and on the other hand its matter 
is experience, and its facts * the simplest and most 
obvious facts—the starting-point from which it rises 
to the causes and grounds which constitute the 
ultimate explanation. Not a whimsy spun out of his 
own brain, but the entire universe with its enormous 
multitude and variety of data must be the philosopher's 
teacher. 
And he must always bear in mind that, if philosophy 
enables the human intellect to apprehend with absolute 
certainty the highest and most profound reaUties of 
the natural order, it cannot therefore claim to exhaust 
those realities by making them known to the utmost 
extent of their inteUigibiUty. From this point of 
view science does not destroy the mystery of things, 
that in them which is still unknown and unfathomed, 
but on the contrary recognises and deUmits it ; 
' even 
what it knows it never knows completely. The wise 
man knows all things, inasmuch as he knows them in 
their ultimate causes, but he does not know, is infinitely 
removed from knowing, everything about everything. 
Ignorance, however, is not the same as error. It is 
sufficient for the philosopher that he knows with 
This is what the Positivists fail to see. 
2 This is what the pure intellectttalists—from Parmenides to Hegel 
— 
who construct their metaphysics wholly a priori, have failed to grasp. 
' Aristotle {Metapk., i, 2) remarks that the occasional cause of 
philosophy is ro Oavfidi^eiv, admiratio, by which he means wonder 
mingled with dread, in other words awe, a wonder which knowledge 
tends to remove. But we must be careful to understand his meaning 
of the wonder which does not understand, not of the admiration, indeed 
the awe, bom of understanding. The wise man is astonished at 
nothing because he knows the ultimate causes of all things, but he 
admires far more than the ignorant man. Cf, De Part. Arum., i, 5, 645 a 
16 : ^c waffL rois <pvaiKoU ivearl u Oav/xaffrdv. 
142
PHILOSOPHY AND COMMON SENSE 
certainty what it is his province to know and what 
it is of the first importance for us to know. Indeed, it 
is better not to know things which divert the mind 
from the highest knowledge, as Tacitus remarks : 
nescire guaedam, magna pars sapientiae. 
We have considered the nature of philosophy ; it 
remains to distinguish its departments. We shall 
thus obtain a clear notion of its sphere, and at the same 
time become acquainted with its principal problems. 
143
Introduction of philosophy
PART TU'O 
THE CLASSIFICATION OF PHILOSOPHY
Introduction of philosophy
THE MAIN DIVISIONS OF PHILOSOPHY 
When a man undertakes a work, he begins by 
testing his tool in various ways to learn the use he 
can and should make of it. 
The philosopher's work is to acquire knowledge ; 
his tool, reason. Therefore the philosopher before he 
begins his work must examine reason to discover the 
use he should make of it. 
The study of reason as an instrument of acquiring 
knowledge or means of discovering truth is called logic. 
Logic is therefore, strictly speaking, not so much a 
department of philosophy as a science or art, of which 
philosophy (and indeed all the sciences) makes use, 
and the introduction to philosophy. It is a propae-deutic 
to science.^ The other sciences are dependent 
upon logic inasmuch as it teaches the method of 
procedure in the acquisition of knowledge, and we are 
obUged to possess the means or tools of knowledge 
before we can acquire knowledge itself. 
It is thus evident that the study of philosophy must 
1 " Res autem de quibus est logica, non quaeruntur ad cognoscendum propter 
seipsas, sed ut adminiculum quoddam ad alias scientias. Et idea logica non 
continetur sub philosophia speculativa quasi principalis pars, sed quasi quoddam 
reductum ad earn, prout ministrat speculationi sua instrumenta, scilicet 
syllogismos et dejinitiones, et alia hujusmodi, quibus in speculativis scientiis 
indigemus. Unde et secundum Boethium in Comment, sup. Porphyrium, 
non tarn est scientia quam scientiae instrumentum." St. Thomas, Sup. 
Boet. de Trin., q. 5, a. i, a^ 2. It is therefore only reductively that 
logic belongs to theoretical philosophy.
AN INTRODUCTION TO PHILOSOPHY 
from the very nature of things begin with logic, 
although on account of its difficulty and extremely 
abstract character logic usually repels rather than 
attracts beginners.^ A few of the moderns rebel 
against this order of study, and maintain that logic 
should be studied only in the course of learning the 
other branches of philosophy, or after they have been 
learned. But this is like arguing that the surgeon 
should only study anatomy by the practice or after the 
practice of his art upon the sick, it is absurd, Aristotle 
remarks, to study at the same time a science and its 
conditions or method of procedure. axoTxov a^xa 
J^YITetv £7ti,cJTYi[i.ir)v xal XpOTTOV erutcTTYifXYjc;'.'' 
When by the study of logic he has made himself 
master of his tool, the philosopher can set to work. 
What that work is we know already : to acquire a 
knowledge of things by their first principles. 
If, however, we consider the aim of learning, there 
are two distinct types of knowledge. We can, for 
example, make use of our eyes simply in order to see 
• Cf. St. Thomas, Sup. Boet. de Trin., q. 6, a. i, ad 3. " Dicendum quod 
in addiscendo incipimus ab eo, quod est magisfacile, nisi necessitas aliud requirat. 
Quandoque enim necesse est in addiscendo non incipere ab eo quod estfacilius, sed 
ab eo a cujus cognitione cognitio sequentium dependet. Et hac positione oportet 
in addiscendo incipere a logica, non quia ipsa sit facilior scientiis ceteris ; habet 
enim maximam difficultatem, cum sit de secundo intellectis ; sed quia aliae 
scientiae ab ipsa dependent, in quantum ipsa docet modum procedendi in omnibus 
scientiis. Oportet enim primum scire modum scientiae quam scientiam ipsam, ut 
dicitur II. Metaph. 
2 Metaph., ii, 995 a 12. St. Thomas, In II Metaph., 1. 5. "Quia 
enim diversi secundum diversos modos veritatem inquirunt, idea oportet quod homo 
instruatur per quern modum in singulis scientiis sint recipienda ea quae dicuntur. 
Et quia non est facile, quod homo simul duo capiat, sed dum ad duo attendit, 
neutrum capere potest : absurdum est, quod homo simul quaerat scientiam et 
modum qui corwenit scientiae. Et propter hoc debet prius addiscere logicam quam 
alias scientias, quia logica tradit communem modum procedendi in omnibus 
scientiis." 
148
THE MAIN DIVISIONS OF PHILOSOPHY 
and enjoy the sight of things, and we can also make 
use of them for the practical purposes of life. 
In the same way we can employ our reason 
scientifically, solely for the pleasure of knowledge. 
The sciences thus acquired exist solely for the sake 
of knowledge (the theoretical sciences). And if there 
be a theoretical science which seeks to account for 
things by their first principles, its object will be that 
which is the first principle in the theoretical order, 
namely, the first causes of everything which exist (that 
is to say, the first causes naturally knowable). That 
science is theoretical philosophy. We can, on the other 
hand, employ our reason scientifically for our profit 
and the improvement of our hfe ; the sciences thus 
acquired exist to procure by some kind of activity the good 
of man (the practical sciences). And if there be a 
practical science which seeks to regulate human acts 
by first principles, its object will be that which is the 
first principle in the practical order, namely, the absolute 
good of man (the absolute good naturally knowable).^ 
Such a science is practical philosophy—otherwise termed 
moral science or ethics.^ 
There are, indeed, other practical sciences besides 
ethics ; for example, medicine, which seeks to procure 
the health of man. But the object of these sciences 
is not good, pure and simple (the sovereign good), 
but some particular human good. They do not, 
* That is to say, the sovereign good of man as it would be, if his end 
were simply natural happiness. See below, pp. 265-267. 
* Observe that this division of philosophy into theoretical and 
practical relates to the end, the aim, not to the object itself of the science, 
which as such is always necessarily theoretical. Therefore it does not 
enter into the specification in the strict sense of the philosophic sciences. 
See below, p. 271.
AN INTRODUCTION TO PHILOSOPHY 
therefore, refer in the practical order to which they 
belong to the first principle of action, and for that 
reason are not philosophies. Ethics, or moral 
science, is thus the only practical science which 
deserves the name of philosophy.^ 
We must bear in mind that although the object of 
ethics is to procure a good which is not solely a good 
of the intellect, that is to say, does not consist in know-ledge 
alone, its rule of truth is that which is, and it 
proceeds by way of demonstration, resolving conclusions 
into their premisses. In other words it is practical 
in virtue of its object (to know in order to procure the 
good of man in the ordering of his acts) , but as science 
in the strict sense it is theoretical knowledge.* 
We must further remark that the practical sciences 
are obviously subordinate to the theoretical, (i) as 
presupposing (if not in the order of their origin in 
1 It may be added that of the practical sciences, only one, ethics, is 
in fact vere et proprie scieniia, that is to say, proceeds by way of demon-stration, 
in a necessary matter, and imparts a truth, which consists in 
knowing things in conformity with what is, and not in properly directing 
a contingent action. The other practical sciences (medicine, architec-ture, 
strategy, etc.) are arts, not sciences in the strict sense. {Cf. John 
of St. Thomas, Cursus Philos., i. Log. ii, q. i, a. 5). 
But though ethics is in the strict sense a science, it is for that very 
reason only in an indirect sense practical : for its procedure consists 
in providing knowledge {speculabiliter) not in producing action {opera-biliUr), 
and though it certainly supplies rules immediately applicable 
to concrete cases, the right application and good use of these rules in 
practice is the effect not of ethics but of the virtue of prudence. 
On the other hand, as we shall see later, the philosophy of art is also, 
in a sense, a practical philosophy. But it is very far from being a 
practical science, even hke ethics in the indirect sense, for it treats only 
of principles and is unable to descend to the rules immediately applicable 
to the concrete work to be executed. 
* Hence even practical philosophy is a theoretical wisdom which 
proceeds by way of knowledge (see above, p. 102). 
150
THE MAIN DIVISIONS OF PHILOSOPHY 
time, at least in the nature of things) the truths proved 
by these sciences, which they apply for the benefit 
of man—for example, medicine, as the science of 
heaUng, presupposes anatomy (ii) as sciences inferior 
; 
in dignity to the theoretical sciences. For the latter 
are studied for their own sake, and are therefore good 
in themselves, whereas the practical sciences are 
studied for the good or utiUty of man, and are therefore 
good only in relation to that good or utility. It 
follows that philosophy in the strictest sense is 
theoretical philosophy (especially the first philosophy 
or metaphysics), logic being the science introductory 
to it, and ethics the science detached from it to treat 
specially of that which concerns the good of man. 
We are now in a position to define more exactly 
the object of these three main divisions of philosophy. 
A science which seeks to procure man's sovereign 
good must before all else treat of those things which 
are the indispensable conditions of its attainment. 
But these are the actions which man performs in the 
free exercise of his faculties ; in other words, human 
acts as such. We may therefore say that human 
acts are the formal object (subject-matter) of moral 
philosophy. 
A science whose aim is to know things by their first 
causes must treat primarily of that in things which 
depends immediately upon those causes. But that 
in things which depends immediately upon the first 
or highest causes is that which is most essential in 
them, their being, and that which is the most widely, 
indeed universally, distributed, being, which everything 
whatsoever possesses. We therefore conclude that the 
151
AN INTRODUCTION TO PHILOSOPHY 
formal object of theoretical philosophy is the being 
of things. 
Now theoretical philosophy studies the being of 
things in different fashions and from higher or lower 
points of view (degrees of abstraction) . It may study 
the being of things with their sensible properties {ens 
mobile) , or the being of things with the sole properties 
of quantity {ens quantum) , or the being of things with 
the sole properties of being (being qua being, ens in 
quantum ens) . Hence arise the three principal divisions 
of theoretical philosophy (see below, Ghs. Ill & IV). 
Finally, a science which studies reason as the tool 
for the attainment of truth must treat before all else 
of that which we handle or manipulate when we 
reason. But that which we handle or manipulate when 
we rezison is the things themselves. For example, when 
we affirm that man is superior to the other animals because 
he possesses intellect, it is indeed the thing itself, man, 
which we hold in our mind and to which we join or 
attribute those other things, intellect and superiority. 
But the man that we thus handle in our mind is 
obviously not the man as he exists, or can exist in 
reality ; there is no question of seizing some man 
who passes in the street to stick an attribute on to his 
back. That our mind may work on them, things 
possess in the mind a manner of being which they do 
not and cannot jjossess in reality. They exist there 
so far as they are known. Predicated one of another 
divided, reunited, linked together according to the 
necessities of knowledge, they lead there a distinct 
life, with its own laws. It is this life and its laws, 
the order to which things must submit so far as they 
are objects of knowledge, if they are to guide the mind 
152
THE MAIN DIVISIONS OF PHILOSOPHY 
to truth, that logic primarily studies, and since it is 
concerned with something which exists and can exist 
only in the mind or with what philosophers term an 
ens raiionis, a conceptual being, we may say that the 
formal object of logic is that conceptual being, 
ens rationis (the order which should prevail among 
conceptual objects) which directs the mind to truth. 
As opposed to conceptual being, the ens rationisy which 
can exist only in the mind—for example, the genus 
animal or the species man {the genus animal comprises 
mankind and the brutes, man is the species of Peter)—we 
term real being, ens reale, that which exists or can exist 
in reality—for example, animals, man, human nature 
{all animals are mortal, human nature is fallible). 
Conclusion V.—Philosophy is divided into three 
principal parts : (i) logic, which is the introduc-tion 
to philosophy in the strict sense, and which 
studies the conceptual being {ens rationis) which 
directs the mind to truth ; 
(ii) theoretical 
philosophy or simply philosophy, which studies the 
being of things (real being, ens reale) ; 
(iii) practical 
philosophy or ethics, which studies human acts. 
153
II 
LOGIC 
Logic studies reason as the tool of knowledge. To 
study any complicated machine, a reaper for instance, 
we must begin by making it work in the void, while we 
learn how to use it correctly and without damaging it. 
In the same way we must first of all learn how to use 
reason correctly, that is to say, in conformity with the 
nature of ratiocination, and without damaging it. 
Hence arises our first problem : What are the rules 
which we must obey in order to reason correctly ? 
We should next study our reaper no longer in the 
void, but as applied to the actual material with which 
it was designed to deal, learning how to use it, not 
only correctly, but profitably and efficiently. In the same 
way we must study reasoning as applied to facts, 
asking ourselves under what conditions reasoning is 
not only correct but also true and conclusive, and 
productive of knowledge. It is in this department of 
logic that we study the methods employed by the 
different sciences. But before this a far graver 
problem will arise and demand solution. 
It is by our ideas that things are presented to the 
mind, so that we can reason about them and acquire 
knowledge. 
Everybody knows by experience what an idea is. 
It is sufficient for a man to reflect on what is in 
154
LOGIC 
a 
; 
his mind, when he makes a judgment of any kind. 
For instance, philosophers have made many mistakes 
philosophers^ mistakes, have made, many—all these are 
present in the mind as so many ideas. Nevertheless 
to clear up any possible obscurity, we will try to 
describe what everybody means by the word. We 
will, for instance, define ideas as images or interior 
reproductions of things, by which the latter are pre-sented 
to us in such a way that we can reason about 
them (and thus acquire knowledge). 
No doubt the words we employ express our ideas. 
But they bring with them something besides. If, for 
example, I pronounce the word angel, I have in my 
consciousness two images of the being in question. 
In the first place an idea, in virtue of which, strictly 
speaking, I know that particular being (the idea of a 
pure spirit), but in addition a sensible representation 
(the image of some figure more or less nebulous and 
winged) which possesses no Ukeness whatever to the 
being in question, for a purely spiritual being is 
invisible. 
If, again, I pronounce the word square I have in 
my consciousness the idea of the square, by means 
of which I can reason about the thing concerned (the 
idea of a rectangular polygon offour equal sides) and at 
the same time a sensible representation—which in this 
case adequately represents the thing in question— 
particular figure which I imagine drawn in chalk on 
the blackboard. The idea and the representation are 
quite distinct, as is shown by the fact that I can vary 
the latter in a host of different ways (the imaginary 
drawing can be larger or smaller, white, red, yellow, 
etc.) without any variation of the former. Moreover, 
155
— 
AN INTRODUCTION TO PHILOSOPHY 
if I were to pronounce, for example, the word myriagon 
instead of square^ I should possess as definite and as 
clear an idea of it as I had of the square (the idea of 
a polygon of ten thousand sides) ^ whereas the only sensible 
representation I could form of it would be extremely 
vague and confused. 
It is evident that if the sensible representations 
assist me to reason, they are not the instrument with 
which I reason to acquire a knowledge of things. 
For I can reason as accurately about the angel or the 
myriagon as I can about the square. And my reasoning 
is in no way dependent on the thousand alterations 
I can make in my sensible representations of an angel, 
myriagon, or square. 
From this we conclude that things are presented to 
our consciousness in two different fashions, either by an 
idea or by a sensible representation. 
By the first we think (intelligimus) the thing, by the 
second we imagine it. The representation is simply a 
species of phantom, an image of what we have pre-viously 
seen, heard, touched, etc., in short, of what 
has been originally made known to us by a sensation. 
Formerly called a phantasm, it is now called simply an 
image. In future, then, we shall reserve to denote 
it the term image, whose meaning we accordingly 
restrict. (But we must no longer use the same word 
to denote an idea.) We conclude, therefore, that 
Conclusion VI.—Ideas are the internal likenesses 
of things by which the latter are presented in such 
a way that we can reason about them (and thus 
acquire knowledge) ; images are the internal 
likenesses of things by which the latter are pre- 
156
LOGIC 
sented to us as our sensations have first made them 
known to us. Words directly signify ideas, at the 
same time evoking images. 
If now we compare objects as they are presented 
by ideas and as they are presented by sensations or 
images, we see at once that they are distinguished 
by a character of the very first importance. If, for 
example, I call up before my mind the image of a 
man, I see present in my imagination with outlines 
more or less vague and more or less simplified some 
particular man. He is fair or dark, tall or short, 
white or black, etc. But if I form the idea of man, 
as, for instance, when I state the proposition man is 
superior to the irrational animals, or whites and blacks are 
alike men, that idea does not bring before me any 
man in particular. It leaves out of account all the 
individual characteristics which distinguish one man 
from another ; in the language of philosophy it 
abstracts from them. 
This is proved by the fact that while remaining 
absolutely the same and without need of any modifica-tion 
whatsoever, it can be applied to the most dis-similar 
individuals ; Sancho Panza is just as much a 
man as Don Quixote. Moreover, when we cast our 
mind over the different sciences, that is to say, the 
different systems of ideas by which we know reality, 
we find that none of them is concerned with the 
individual as such. Chemistry, for instance, only 
studies in chlorine or nitrogen what is common to all 
the individual molecules of chlorine or nitrogen. 
And this must necessarily be the case. The individual 
as such explains nothing (for, since it represents only 
157
AN INTRODUCTION TO PHILOSOPHY 
itself, it obviously cannot account for anything else).^ 
Again, we have only to take any idea whatsoever and 
fix our attention on what it presents to us, comparing 
it with the images which form and dissolve around it, 
to perceive at once the abstract nature of the idea. 
In the transition from the image to the idea whatever 
is individual evaporates, so to speak, slips between 
our fingers, and vanishes. Take, for example, the 
idea of weapon which I employ when I state that 
man is the only animal obliged to manufacture its weapons. 
As I pronounced the word weapons, I was no doubt 
conscious of a halo, so to speak, of fluctuating images 
surrounding the idea thus expressed, to any of which 
I can at pleasure give a more definite shape, di javelin 
very shadowy no doubt, a flint axe^ a cross-bow, a 
gun. . . . But of the individual characteristics of the 
particular javelin, axe, bow, or gun as they appear 
in my imagination with their distinctive form, colour, 
and dimensions, nothing whatever remains in my idea 
of weapon. Everything of the sort has disappeared. 
Though what I apprehend by the idea is certainly 
some thing, that some thing is of an entirely different 
order (immaterial), it is simply a certain determination 
of being, a certain nature, an instrument of attack or 
defence ; and that is devoid of any individual character. 
That is to say, objects as presented to us by our 
sensations and images are presented in a state which is 
individual, or in technical language singular. On the 
contrary, objects as presented to us by our ideas, by 
the internal likenesses which enable us to reason about 
them, are presented in a state which is non-individual, 
abstract, or in technical language universal. 
1 Cf. T. Richard, op. cit., p. 21. 
158
LOGIC 
— 
We call universal that which is the same in a multi-tude 
of individuals, one in many, unum in multis.) We 
shall therefore hold as an established truth that 
Conclusion VII.—Our sensations and images 
present to us directly or by themselves the indi-vidual, 
our ideas directly and by themselves the 
universal. 
But the question at once arises : Since real objects 
are individual or singular, how can the knowledge 
we obtain by means of our ideas be true, since our 
ideas directly present only the universal ? 
This problem, which will compel us to investigate 
carefully in what exactly consists the universality of 
that which our ideas present, is, not indeed in itself, 
but at any rate for us men, the first and most im-portant 
of philosophic problems.^ For it is concerned 
with the nature of the intellect itself and of our ideas, 
that is to say, with the instrument by which all our 
knowledge is obtained ; and the solution propounded 
by different philosophers dominates their entire 
system. 
From this point of view, and taking no account of 
many differences of secondary importance, we may 
classify philosophers in three great schools : 
{a) The nominalist school, for which universzds have 
1 Does the problem of universals belong to logic, psychology, or 
metaphysics (criticism or epistemology) ? To all three, in fact, 
according as it is studied from three different standpoints. We may 
inquire what constitutes the nature of a universal (standpoint of the 
formal cause), or the manner in which a universal is formed in the mind 
(standpoint of the efficient cause), or the epistemological value of the 
universal (standpoint oi the final cause).
AN INTRODUCTION TO PHILOSOPHY 
no existence except as names or ideas with which 
nothing in reality corresponds ; for instance, there is 
nothing in the reality of human nature which is equally 
present in Peter^ Paul, and John. This position amounts 
to sheer negation of the possibility of intellectual 
knowledge, and reduces science to a figment of the 
mind. The most typical representatives of this school 
are, in antiquity the sophists and the sceptics, in modem 
times the leading English philosophers, William of 
Occam in the fourteenth century, Hobbes and Locke 
in the seventeenth, Berkeley and Hume in the eigh-teenth, 
John Stuart Mill and Spencer in the nine-teenth. 
It may be added that the majority of modem 
philosophers (that is to say, of those who ignore or 
oppose the scholastic tradition) are more or less 
deeply, and more or less consciously, imbued with 
nominalism. 
(b) The realist school (" absolute realism ") for which 
the universal as such, the universal taken separately, 
as it exists in thought, constitutes the reality of things. 
This position reduces sense-knowledge to mere illusion. 
That which is real is, for example, a human nature 
existing in itself and separately outside the mind, a man in 
himself (Platonism), or a universal being existing as such 
outside the mind and regarded as the sole and unique substance 
(doctrine of Parmenides, Vedantism) . The systems of 
certain modern philosophers (Spinoza, Hegel) approxi-mate 
more or less closely to realism.^ 
{c) The school which is usually called that of 
> It mast be borne in mind that realism understood in this sense, 
far from being incompatible with idealism, is essentially an idealist 
doctrine. For realism of this type regards as the reality of things 
that which is distinctive of our ideas as such. Plato is thus at once 
the most typical representative both of idealism and absolute realism. 
i6o
LOGIC 
moderate realism. (Its doctrine, however, is in the most 
strict sense original, and keeps the just mean between 
realism and nominalism, not by watering down or 
modifying absolute realism, but by a view of things 
which transcends the opposing errors.) This school, 
distinguishing between the thing itself and its mode 
of existence, the condition in which it is presented, 
teaches that a thing exists in the mind as a universal, 
in reality as an individual. Therefore that which we 
apprehend by our ideas as a universal does indeed 
really exist, but only in the objects themselves and 
therefore individuated—not as a universal. For 
example, the human nature found alike in Peter, Paul, 
and John really exists, but it has no existence outside the 
mind, except in these individual subjects and as identical 
with them ; it has no separate existence, does not exist in 
itself. This moderate realism is the doctrine of 
Aristotle and St. Thomas. 
Philosophy of Aristotle 
and St. Thomas {Moderate Realism) 
That which our ideas That which our ideas 
present to us as a universal present to us as a universal 
does not exist outside the exists outside the mind 
mind as a universal. individuated. 
Nominalism Realism 
That which our ideas That which our ideas 
present to us as a universal present to us as a universal 
has no real existence what- really exists as a universal, 
soever. 
It is impossible to over-emphasise the importance 
of the problem of universals. It is for want of atten-tion 
to it that so many philosophers and scientists of 
modem times cling to the naive belief that science 
i6i
AN INTRODUCTION TO PHILOSOPHY 
must be a copy pure and simple, a tracing of the 
individual reality ; serve up the stock arguments 
of ignorance against abstraction, the essential pre-condition 
of all human knowledge ; and when treating 
of the principles of the sciences, especially of mathe-matics, 
spin elaborate theories, devoid of solid founda-tion, 
whose sole result is to render knowledge totally 
impossible. 
-i 
162
Ill 
THE PHILOSOPHY OF MATHEMATICS 
THE PHILOSOPHY OF NATURE 
The distinctive object of theoretical philosophy is 
the being of things. The things which are immedi-ately 
observed are corporeal things, bodies. But the 
term body may be taken in two distinct senses. It 
may mean a mathematical body, or a natural or physical 
body. A mathematical body is simply that which 
possesses three-dimensional extension, breadth, length, 
and height. A natural or physical body is that 
which is perceived by the senses as possessing certain 
active and passive properties. 
THE PHILOSOPHY OF MATHEMATICS 
If the philosophy of mathematics studies the being of 
bodies in the first sense of the term body, it is obvious 
that the first problem it must consider is in what does 
the primary object of mathematics consist ; in other 
words, what is the nature of quantity, extension, and 
number ? ^ 
1 Questions relating to the philosophy of mathematics are usually 
treated in natural philosophy or in metaphysics. We believe, however, 
that if classification is to be scientific, we are obliged to maintain in 
what is now known as philosophy (scientific knowledge of things by 
their first causes) the fundamental division of the sciences (the whole 
group of which constituted for the ancients theoretical philosophy) 
into three parts : physica, mathematica, metaphysica, corresponding to 
163
AN INTRODUCTION TO PHILOSOPHY 
The enormous progress made by modern mathe-matics 
has rendered more indispensable than ever 
before the philosopliic study of the first principles of 
the mathematical sciences, which alone can provide 
a rational account of the true nature of mathematical 
abstraction and the mentzd objects which it considers, 
the properties- and mutual relationships of the con' 
tinuous and the discontinuous, the real meaning of surds 
and transjinite numbers, the infinitesimal, non-Euclidean 
space, etc., and finally of the validity of mathematical 
transcripts of physical reahty, and of such hypotheses, 
for example, as the theory of relativity. 
THE PHILOSOPHY OF NATURE * 
Since the philosophy of sensible nature studies the being 
of bodies in the second sense of the term body, it 
the three grades of abstraction (see p. 1 52) . Cf. Aristotle, Metaph., vi, i, 
1026 a 18. Tjoeii iv eUv <pio(Xo<plai dewptjTiKal, ixad7)/j,aTiKT^, <(>v<nKi/i, 
deciXoyiKTi. 
It is true, as we shall see later, that the philosophy of mathematics, 
for the very reason that it studies the essence of quantity and is thus 
at least reductively metaphysical, transcends the strict sphere of the 
mathematical sciences and is specifically distinct firom them. This, 
however, does not alter the fact that it is concerned with the second 
degree of abstraction and must therefore be studied as a separate 
branch of philosophy. 
1 In the logical order of the sciences, the natural sciences which 
correspond to the first degree of abstraction (see p. 152) precede the 
mathematical sciences, which correspond to the second, so that in 
accordance with this order we should be obliged to divide theoretical 
philosophy into (i) The philosophy of nature (corresponding to the first 
degree of abstraction), (ii) The philosophy of mathematics (corresponding 
to the second degree), (iii) metaphysics (corresponding to the third 
degree). 
Nevertheless the philosophy of mathematics should precede natural 
philosophy for two reasons. 
On the one hand, truths of the mathematical order are easier to 
164
MATHEMATICS AND NATURE 
must deal with a large number of problems. We can, 
however, pick out the most important of these for 
mention here. 
The most universal and obvious characteristic of 
the corporeal world, which is involved in every 
physical event, is change. Philosophers, in whose 
vocabulary change of every description is termed 
motion, must therefore inquire what motion is. 
It is at once obvious that if motion exists, something 
must be moved, namely, bodies. Further, certain 
changes seem to affect the very substance of bodies ; 
as, for instance, when the chemical combination of 
hydrogen and oxygen produces a new body, water. 
How is this possible ? We are compelled to ask 
what corporeal substance is. 
(a) The mechanists—whether in their doctrine of 
the human soul they are materialists (Democritus, 
Epicurus, Lucretius, among the ancients, Hobbes in 
the seventeenth century, etc.) or spiritualists Uke 
apprehend than truths of the natural order, which presuppose 
experience. For this reason children should be taught the elements 
of mathematics before the natural sciences, the study of whic^equires 
a more advanced age. {Cf. Aristotle, Mc. Eth., vi ; St. T'homas, 
Sup. Boet. de Trm., q. 5, a. i, a// 3.) 
We should therefore follow the same order in philosophy and lead the 
mind up to the study of natural philosophy by the study of the 
philosophy of mathematics. 
On the other hand, and this is the more important consideration, 
natural philosophy with the last and highest of its sub-divisions, 
namely psychology, touches the frontier of metaphysics. It would be 
a breach of continuity to insert the philosophy of mathematics between 
natural philosophy and metaphysics. 
In the seventeenth century Sylvester Maurus maintained—and in 
so doing was faithful to the Aristotelian tradition—that the natural 
order of study is as follows : (i) logic, (ii) mathematics, (iii) physics, 
(iv) metaphysics. {Quaest. philos., i, q. vii.) 
165
AN INTRODUCTION TO PHILOSOPHY 
Descartes—reduce corporeal substance to matter, which 
in turn they confuse with quantity or geometrical 
extension. They can therefore admit no essential 
or specific difference among bodies, which are all 
modifications of one single substance. Moreover, the 
physical universe is for them devoid of quality and 
energy, since space and local motion alone are real, 
and the union of matter and spirit in a being such as 
man becomes absolutely unintelligible. 
{b) Another school {dynamism) tends on the contrary 
to get rid of matter as a constituent of bodies. It 
culminates in the system of Leibniz, who reduced 
corporeal substance to units of a spiritual character 
{monads) analogous to souls. For Leibniz extension, 
indeed sensible reaUty as a whole, is nothing more 
than an appearance or a symbol, and the corporeal 
world as such is absorbed in the spiritual. The 
dynamism of Boscovich (eighteenth century), who 
reduced corporeal substance to points offorce, and the 
modern physical theory which claims to explain 
everything in the physical universe as manifestations 
of one sole reality, energy (of which, however, its ex-ponents 
fail to give a philosophic definition), may be 
regarded as degradations and materialisations of 
Leibniz's doctrine. 
{c) The Aristotelian philosophy recognises in 
corporeal substance two substantial principles : 
(i) matter {first matter, materia prima), which, however, 
in no way represents, as in the conception of the 
mechanists, the imaginable notion of extension, but 
the idea of matter (that of which something else is 
made) in its utmost purity—it is what Plato called a 
sort of non-entity, simply that of which things are 
i66
MATHEMATICS AND NATURE 
made, which in itself is nothing actual, a principle 
wholly indeterminate, incapable of separate existence, 
but capable of existing in conjunction with something 
else (the form) ; (ii) an active principle, which is 
so to speak, the living idea or soul of the thing, and 
which determines the purely passive first matter, 
somewhat as the form imposed upon it by the sculptor 
determines the clay, constituting with it one single 
thing actually existent, one single corporeal substance, 
which owes to it both that it is this or that kind of 
thing, that is to say, its specific nature, and its existence, 
somewhat as the form imposed by the sculptor makes 
a statue what it is. On account of this analogy with 
the external form of a statue (its accidental form) 
Aristotle gave the name of/orm {substantialform) , which 
must be understood in a sense altogether special and 
technical, to this internal principle of which we are 
speaking, which determines the very being of corporeal 
substance. 
The Aristotelian doctrine, which regards a body as 
a compound o{ matter [xikri) andform ([xopcprj), is known 
as hylomorphism. It accepts the reahty, on the one hand, 
of matter, the corporeal world, and extension,^ on 
the other of physical qualities,' also a distinction of 
nature or essence between the bodies which we regard 
as belonging to different species. It reveals the 
presence, even in inanimate bodies and Hving things 
devoid of reason, of a substantial principle, immaterial 
in its nature, which, however, differs from spirits in 
^ Extension of quantity is not, as the mechanists hold, the substance 
of bodies, but their first accident. 
* Qualities are also accidents of corporeal substance. (See below, 
pp. 217-232 (substance and accident).) 
167
AN INTRODUCTION TO PHILOSOPHY 
the Strict sense, in its incapacity to exist apart from 
matter. And it renders intelligible the union in 
the human being of matter and a spiritual soul which 
is the form of the human body, but differs from the 
Other substantial forms inasmuch as it can exist apart 
from matter. 
Ptdlosophy of Aristotle and St. Thomas {Hylomorphism) 
Every corporeal substance is a compound of two 
substantia! and complementary parts, one passive and 
in itself wholly indeterminate [matter), the other active 
and the principle of determination {form). 
Mechanism Dynamism 
Ck>rporeal substance is Corporeal substance is 
regarded as something explained either as units 
simple, a matter itself belonging to the category 
identified with geometrical of pure forms and spirits 
extension. (Leibnizian monadism) or 
as a manifestation of force 
or energfy. 
We have now to consider a class of bodies which 
possess a peculiar interest for us, and seem to be superior 
to all the others. They are living bodies, from the 
lowliest micro-organism to the human organism. 
The property which distinguishes them from all other 
bodies is self-movement. On that account common sense 
recognises in them a soul or principle of life, irreducible 
to any combination of physico-chemical factors or 
elements. If this is indeed the case, we must inquire 
whether there are not different kinds of soul, whether 
vegetables and animals possess a soul, etc. On the 
other hand certain philosophers (known by the 
general appellation of mechanists) claim that science 
will one day explain all the phenomena of life by 
the forces of lifeless matter, that is to say, that the living 
i68
MATHEMATICS AND NATURE 
organism is simply a very complicated physico-chemical 
machine. This involves a problem of the 
first importance. What is Ufe ? What are the first 
principles which constitute the living organism ? 
But of all living things which possess a body the 
highest is man. Man is as it were a world apart, 
for the study of which we are in a peculiarly favourable 
position, because we know him from within by what is 
called self-consciousness. His most distinctive charac-teristic 
is the possession of intelligence or reason. If, 
however, intelligence is indeed something wholly 
immaterial, it follows that the science which studies 
man, though a branch of natural philosophy which 
treats of moving or sensible being, is in a sense 
intermediate between this department of philosophy 
and metaphysics which treats of the wholly immaterial.^ 
If it is the possession of intelligence or reason which 
makes man man, the problems which relate to his 
intellectual activity must, it would seem, dominate 
the entire science of man.^ And in fact the 
1 The science of man occupies therefore a singular position (due to 
the very nature of its object) astride two distinct sciences, nattiral 
philosophy and metaphysics. It is for this reason that all questions 
involving the intellect and the strictly spiritual portion of psychology 
display in the case of man such extreme complexity and are, so to 
speak, overshadowed by matter. It is not surprising, therefore, that 
when the Thomists wished to investigate these questions in their purity 
they studied not man but the angels. Hence the extreme importance 
of the treatise De Angelis, not for theology alone, but also for philosophy 
and metaphysics. 
' Observe that psychology as understood by the moderns does not 
correspond exactly to the ancients' treatment of the soul. Aristotle's 
Trepl fvxv^> De Anima, studies not only the human soul, but also the soul 
in general as the principle of life, whether vegetative, sensitive, or 
intellectual. Such a treatise therefore belongs to what we now call 
biology as well as to psychology. 
169
AN INTRODUCTION TO PHILOSOPHY 
fundamental problem of psychology is that of the origin, 
of ideas : how we are to explain the presence in us of 
the ideas which enable us to reason about things 
and which present things to us as universals. 
At this point we have been brought back by a 
different approach to the problem of universals which 
we have lately considered. We then noted that what 
our ideas immediately present to us is something 
non-individual or universal. We have now to ask 
how this knowledge of the universal is acquired by 
our minds. 
We saw above that things as they are known by 
the senses and tha imagination are presented as 
individuals. It is this particular man that I see, with 
this particular appearance actually impressing itself 
on my retina and distinguishing him from the other 
man I see beside him. Sense-knowledge is thus 
knowledge of the individual alone. The object as 
object of sensation or the object reproduced by an 
image is the object apprehended in its individuality. 
Since, therefore, what we know immediately by our 
ideas is not individual, the reason must be that our 
ideas are in fact extracted by us from our sensations and 
images, but in such a fashion that there enters into 
them nothing whatsoever of the object as it exists 
as an object of sensation or reproduced by an image (that is 
to say, as we shall see later, as the object of a know-ledge 
steeped in materiality). Arising from images, 
but of a higher order than the image, and apprehending 
nothing of the object as reproduced by the image, our 
ideas must necessarily be unable to give us any know-ledge 
of the object in its individuality. 
170
MATHEMATICS AND NATURE 
Moreover, we could not possibly derive our ideas 
from things, except by way of our senses, which are 
in immediate contact with things. And we have 
only to observe the mental development of a child to 
be convinced that all our knowledge begins with the 
senses. Therefore intellectual knowledge (knowledge 
by means of ideas) must undoubtedly be derived from 
sense-knowledge. 
On the other hand, since everything apprehended 
by sensations and images is characterised by individu-ality, 
our ideas must be extracted from images in such 
a way that nothing of the image, as such, enters into 
the idea. 
But how is this process of extraction conceivable ? 
If nothing whatever of the object as it is reproduced 
by the image enters into the object as it is apprehended 
by the idea, it is obvious that the idea is not the result 
of any combination or distillation of sensations or 
images. We are therefore compelled to postulate an 
agent of a higher order, the voii<; 7roL7]Ti.x6(;, as the 
Peripatetics termed it, the intellectus agens—a kind of 
intellectual light (we may perhaps compare it to 
X-rays) which, when applied to the object presented 
to us by the image, draws out of it for our understanding 
something already contained in it but hidden, which 
the image by itself could never reveal. The some-thing 
thus extracted and Uberated from that which 
constitutes the individuality of the object (because 
it is liberated from that which constitutes the materi-ality 
of sense-knowledge) is the form or intelligible 
likeness of the object, which is, so to speak, imprinted 
on the intellect to determine it to know, by making 
it produce within itself by a vital reaction the idea 
171
AN INTRODUCTION TO PHILOSOPHY 
in which it apprehends the object as a universal ; for 
example, the idea of man or living being, of Aryan or 
Semite. 
We must, however, remember that what our ideas 
present thus as a universal is in itself (abstracting 
fix)m its existence either in things or in the mind) 
neither individual nor universal, being purely and 
simply that .which the thing is.^ 
We must also bear in mind that, if our intellect 
does not directly know the individual as such, it 
knows it indirectly. For, at the very moment when 
it thinks of an object by means of an idea, it turns to 
the images from which the idea has been drawn, 
which present the thing as an individual. And by 
thus reflecting on the images it apprehends, though 
indirectly and in a manner wholly superficial and 
totally inexpressible, the individuality of the thing. 
Conclusion VIII.—Our ideas are extracted {ab^ 
straded) from the sensible datum by the activity 
of a special faculty (the intellectus agens or active 
intellect) which entirely transcends the sensible 
order and is, as it were, the light of our under-standing. 
Philosophers term abstraction the operation by which 
we thus extract our ideas from the store of images 
accumulated by sense experience, ideas which repre-sent 
that which the thing is, abstracting from its 
individuality. 
Here we may add that abstraction admits of lesser 
and greater degrees. For instance, though the idea 
1 I.e. the nature, essence, or quiddity, of the object. See pp. 20i, 206. 
172
MATHEMATICS AND NATURE 
of horse is, like every idea, abstract, when we think of 
horse we can at the same time see or imagine horses, 
and thus know in the sensible order what we know 
at the same time by means of our idea in the intelligible 
order. If, on the other hand, we think of angel or 
spiritJ the sole function of the more or less vague images 
which accompany the thought is, as we have already 
observed, to assist the intellect to function. In their 
own order they have no value as knowledge, for they 
tell us nothing. We can neither see nor imagine an 
angel or a spirit ; in this case, therefore, we cannot 
at the same time know by our senses the thing we 
know by our intellect. 
It is important to bear in mind that the things 
with which philosophy is primarily concerned belong 
to this second category. They cannot be known 
either by the senses or the imagination, but solely by 
the intellect. 
It is to this higher degree of abstraction that the 
study of philosophy owes its special difficulty. Be-ginners 
are often perplexed when they suddenly 
exchange the literary studies on which they have been 
engaged hitherto, studies in which the imagination 
was employed equally with the intellect, for an exercise 
wholly intellectual. But this difficulty will soon pass, 
if they will not try to represent by the imagination 
objects of pure thought which are entirely unimagin-able, 
such, for instance, as essence, substance, accident, 
potentiality, and act ; a chimerical attempt, which will 
only cause needless headaches and effectually prevent 
them from understanding anything of philosophy. 
If abstraction is indeed such an operation as we 
have described, it follows in the first place that man 
173
AN INTRODUCTION TO PHILOSOPHY 
is endowed with a spiritual soul, the first principle 
of this function (for its result, our ideas, is incommensur-able 
with sensations and images and of a purely 
immaterial order) ; and on the other hand, that it is 
of the very nature of this spiritual soul to be united 
to a body (for our ideas cannot be formed except by 
means of sensations and images, which in turn neces-sarily 
suppose bodily organs). We thus perceive how 
the problem of abstraction, or the origin of ideas, is 
bound up with another fundamental problem of 
psychology, which concerns the very essence of man : 
in what does the human being consist ? Does man 
possess a spiritual soul, wholly different from that of 
the beasts ? And if so what is the relationship between 
this soul and the human body ? 
On the problem of the origin of ideas philosophers 
may be divided roughly into three main groups : 
(a) The sensualists, who hold that ideas are derived 
from the senses, but reduce ideas to sensations, (b) The 
partisans of innate ideas,^ who recognise the essential 
distinction between ideas and sensations or images, 
but deny that we extract our ideas fi-om the sense 
datum, (c) The school of Aristotle and St. Thomas^ 
which holds that our ideas differ essentially from 
sensations and images, but that they are extracted 
^ We may so term this second group in default of a more suitable title, 
but only if we considerably widen its meaning. For in this class of 
philosophers we must include not only those who teach that our ideas 
exist in our minds from birth in the same way as our soul exists (the 
doctrine of innate ideas in the strict sense), but those who hold that 
they are immediately implanted in us by God or are seen by us in GJod 
(Berkeley, Malebranche), or are the arbitrary product of our mind 
imposing its laws on things (Kant).
MATHEMATICS AND NATURE 
from them by the operation of the spiritual light in us 
(voiji; noirixiy(.6c„ intellectus agens). 
The principal representatives of sensualism are 
Locke (seventeenth century) and John Stuart Mill 
(nineteenth century) in England, and Condillac 
(eighteenth century) in France. The sensualists are, 
as a rule, also nominalists, but the converse does not 
hold, and many philosophers whom we class here 
among the defenders of innate ideas betray, in modem 
times at least, the influence of nominalism. In the 
second class (the defenders of innate ideas) we must 
reckon Plato among the ancients, Descartes (seven-teenth 
century) and Leibniz (seventeenth to eighteenth 
century) among the modems. Though their explana-tions 
differ, all these hold that our ideas are innate. 
Kant (end of eighteenth century) also belongs to this 
group, though for him what is innate is not our ideas, 
but the categories, rules, or forms in accordance with 
which our mind manufactures the objects of knowledge. 
Philosophy of AristotU and St. Thomas 
Our ideas are derived Our ideas are essentially 
from the senses (and there- different from sensations 
fore from things) but by and images, but are ab-the 
operation of a spiritual stracUd from them by the 
faculty, and are essentially operation of a spiritual 
different from sensations faculty, 
and images. 
Sensualism Doctrine of Innate Ideas 
Our ideas are derived Ideas differ essentially 
from the senses, which are from sensations and images 
sufficient to produce them, and are not derived from 
and do not differ essenti- the senses (nor therefore 
ally from images and sen- from things, with which 
sations. our senses alone are in 
immediate contact).
. 
AN INTRODUCTION TO PHILOSOPHY 
The answers which philosophers have given to the 
problem of human nature correspond strictly with the 
position they adopt towards the problem of abstrac-tion. 
The sensualists, at least so far as they are faithful 
to the logic of their doctrine (Gondillac, for example, 
was not), deny either that the soul existe {materialists), 
or that we can in any case know its existence {pheno-menalists) 
. The defenders of innate ideas, on the other 
hand, tend to regard man as a pure spirit which 
happens to be joined to a body—how, they find it 
difficult to explain {dualism or exaggerated spiritualism).^ 
Finally, the school of Aristotle and St. Thomas teaches 
that man is a composite of two substantial principles, 
each incomplete in itself and the complement of the 
other, one of which is a spiritual and immortal soul 
{animism) 
Philosophy of Aristotle and St. Thomas (Animism) 
Two principles each incomplete in itself, one of which 
(the rational soul) is spiritual, form together a single 
substance (the human composite). 
Error of Defect Error of Excess 
The human soul does not Man is a spirit acci-exist 
{materialism) or is dentally united to a body 
unknowable {phenomenal- {exaggerated spiritualism) : 
ism). the soul and the body are 
two substances each com-plete 
in itself {dualism) 
. 
We should remark further that the position adopted 
by philosophers towards the origin of ideas also deter-mines 
their attitude to the general problem of the 
* This tendency recurs even in Kant (especially in ethics), though 
he, like the phenomenalists, denies that reason can demonstrate the / 
existence of the soul. >i 
176
MATHEMATICS AND NATURE 
. 
existence of things known by the senses (the sensible 
or corporeal world) and of things invisible and 
spiritual, accessible to reason alone. 
Philosophy of Aristotle and St. Thomas 
{also of common sense) 
It is impossible without absurdity to doubt either the 
existence of corporeal objects (attested by the senses) 
or the existence of spiritual objects (proved by reason) 
Systems more or less 
Materialist 
Nothing exists which is 
not perceptible by the 
senses and material {abso-lute 
materialism) ; or at 
least its existence is un-knowable 
{phetwmenalist 
materialism and positivism). 
Systems more or less 
Idealist 
The world perceived by 
the senses has no real 
existence {absolute idealism); 
or at least its existence is 
unknowable and doubtful 
{phenomenalist idealism). 
177
IV 
CRITIGISM (ePISTEMOLOGY) 
In studying man philosophy is dealing with an 
object which already by an entire portion of itself 
transcends the corporeal world, that is to say, the world 
of sensible nature. But it has the power and duty to 
go further, and since its distinctive object is the being 
of things, it must study that being no longer as 
corporeal, sensible, or moving (the subject-matter of 
the philosophy of sensible nature) , but simply as being ; 
consequently it must study being under an aspect 
absolutely universal, and as it is present not only in 
visible things but also in things which possess no 
corporeal, sensible, or mobile being ; that is to say, 
in things which are purely spiritual. This is the 
object of that branch of philosophy which is philosophy 
or wisdom par excellence, and is known as the first 
philosophy or metaphysics.^ 
CRITICISM (ePISTEMOLOGY) 
But before undertaking this study, the philosopher 
must secure against all possible attack or distortion the 
1 The name metaphysics originated in the fact that in the catalogue 
of Aristotle's works drawn up by Andronicus of Rhodes, the treatise 
dealing with the first philosophy (Ilepi rrjs irpuTtp <t>ioao<f>la^, the title 
probably which Aristotle himself would have given to it) comes after the 
books which treat of naturi (Meri tA tpvaiKi). It would seem, however, 
that chronologically Aristotle followed the same order in the actual 
comp>osition of his works. 
178
CRITICISM ( EPISTEMOLOGY) 
principles of this sovereign science, which are also the 
principles of all human knowledge. For it is the 
office of wisdom to defend its own principles and those 
of the other sciences. 
It will therefore be necessary, before studying being 
in itself, as such, to study the relation ofhuman thought 
to being. This is the object of a special department of 
metaphysics, known as criticism, because it has the 
function of judging knowledge itself Logic shows 
how and in accordance with what rules reason attains 
truth and acquires knowledge ; this in turn pre-supposes 
the pKDSsibility of true knowledge (a possibility 
attested by common sense and evident by the light of 
nature). Criticism submits this presupposition to 
scientific treatment, showing in what the truth of 
knowledge consists, and establishing by a reflex 
argument that true, certain, and scientific knowledge 
is undoubtedly attainable.^ 
What is the truth of knowledge, and is it possible to 
refiite those who question the veracity of our organs of 
knowledge, particularly of the intellect or reason ? 
This clearly is the double question which arises at the 
outset. The answer, however, is sufficiently plain. 
As to the first question, there is no difficulty in under-standing 
what is meant by the notion of truth. What 
is a true or truthful word ? A word which expresses, 
as it really is, the speaker's thought ; a word in 
1 By thus distinguishing criticism (epistcmology) from logic, and 
making it the first part, special introduction, or if you prefer, 
apologetic introduction, to metaphysics, we are faithful to the 
arrangement and divisions of Aristotle himself, who discusses 
criticism briefly {Metaph., iv) before studying the great problems of 
being as such.
AN INTRODUCTION TO PHILOSOPHY 
conformity with that thought. What, then, is a true 
thought ? A thought which represents, as it really is, 
the thing to which it refers ; a thought in conformity 
with that thing. We therefore conclude that truth 
in the mind consists in its conformity with the thing. 
It is impossible to define truth otherwise without 
lying to ourselves, without falsifying the notion of 
truth of which in practice we make use, in the living 
exercise of our intelligence, each time that we think. 
We may further remark that a thought false in all 
its constituents is an impossibility for, being in 
conformity with nothing whatsoever, it would be the 
zero of thought. If, for instance, I affirm that stones 
have a soul^ this is undoubtedly a complete error. But 
it is true that stones exist, true also that certain beings 
have a soul ; that is to say, all the constituents which 
compose this false thought are not false. Therefore 
error itself presupposes truth. ^ 
We may also observe that if man were really and 
seriously to doubt the veracity of his organs of 
knowledge he simply could not live. Since every 
action or abstention from action is an act of trust in 
that veracity, action and inaction would alike become 
impossible. A man therefore who attempted to carry 
out in his Ufe the thought truth is impossiblefor me would 
inevitably lose his reason. Nietzsche, who was a great 
poet but regarded belief in truth as the ultimate 
bondage from which the world should be deUvered, 
made the experiment to his cost. 
i Cf. Sum. TheoL, ii-ii, q. 172, a. 6 : Sicut se habet bonum in rebus, 
ita verum in cognilione. ImpossibiU est autem inveniri aliquid in rebus, 
quod totaliter bono privetur : unde etiam impossibile est esse aliquam cogrdtionem 
quae totaliter sit falsa absque admixtione alicujus veritatis. 
180
CRITICISM (ePISTEMOLOGY) 
As for the sceptics, who doubt, at least theoretically 
and in words, the reliability of our organs of knowledge, 
especially of the intellect or reason, it would obviously 
be waste of breath to attempt to demonstrate its 
reliability to them. For every demonstration rests on 
some previously admitted certainty, and it is their very 
profession to admit of none. To defend human 
knowledge against their attack it is sufficient (i) to 
show in what that knowledge consists and how it is 
attained ; 
(ii) to refute the arguments they adduce ; 
(iii) to make a reductio ad absurdum. When they say 
that they do not know whether any proposition is 
true, either they know that this proposition at any rate 
is true, in which case they obviously contradict 
themselves, or they do not know whether it is true, 
in which case they are either saying nothing whatever, 
or do not know what they say. The sole philosophy 
open to those who doubt the possibility of truth is 
absolute silence—even mental. That is to say, as 
Aristotle points out, such men must make themselves 
vegetables. No doubt reason often errs, especially in 
the highest matters, and, as Cicero said long ago, there 
is no nonsense in the world which has not found some 
philosopher to maintain it, so difficult is it to attain 
truth. But it is the error of cowards to mistake a 
difficulty for an impossibihty. 
Conclusion IX.—The truth of knowledge consists 
in the conformity of the mind with the thing. 
It is absurd to doubt the reliability of our organs 
of knowledge. 
On this question of the reliability of our organs of 
i8i
AN INTRODUCTION TO PHILOSOPHY 
knowledge philosophers may again be divided 
— 
roughly—into three groups : 
{a) The sceptics, who, impressed by the enormous 
number of errors put forward by men, and especially 
by philosophers, doubt the trustworthiness of reason, 
and affirm that truth is impossible of attainment. 
The principal representatives of scepticism are, among 
the ancients, Pyrrho (360-270), the neo-Academics 
(Arcesilas 315-241 ; Cameades, 214-129) and the later 
Greek sceptics (Aenesidemus, first century a.d., and 
Sextus Empiricus, end of the second century) ; in modern 
times Montaigne and Sanchez in the sixteenth century, 
and pre-eminently David Hume in the eighteenth. 
The philosophers called anti-intellectualists, because 
they despair of intellect and reason, and look for truth 
to the will, to instinct, feeling, or action (Rousseau, 
Fichte, Schopenhauer, Bergson, William James, the 
modernist and pragmatist school), must be classified with 
the sceptics, because, although they do not, like the 
sceptics strictly so called, declare truth unattainable, 
they maintain that it is unattainable by the organ whose 
distinctive nature it is to discover truth, and because 
by rejecting the intellect and reeison they effectually 
deprive man of his sole normal means of attaining it. 
(b) The rationalists, on the contrary, are of opinion 
that truth is easy to attain, and therefore undertake to 
bring all things within the compass of reason, a human 
reason which has no need to submit humbly and 
patiently to the discipline, whether of reaUty itself, a 
teacher, or God. In the first case they tend to 
subjectivism, which takes as its criterion of truth the 
knowing subject, not the object to be known ; a 
position which is the dissolution of knowledge. In the 
182
. 
CRITICISM ( EPISTEMO logy) 
second they tend to individualism, which calls upon each 
philosopher to work out a philosophy entirely his own, 
and create an original and novel view of the universe 
{Weltanschauung). In the third, they tend to naturalism^ 
which claims to attain to a perfect wisdom by the 
unassisted powers of nature, and rejects all divine 
teaching.^ 
The father of modem rationalism was Descartes 
(seventeenth century) , to whom Malebranche, Spinoza, 
and Leibniz traced more or less directly their 
philosophic pedigree. But its first principles and true 
spirit were revealed by Kant (end of eighteenth 
century), who completed the Cartesian revolution ; 
his pantheistic successors, Fichte, Schelling, and Hegel, 
deified the human subject of knowledge. Through 
Kant, and the subjectivist philosophy which traces its 
origin to him, rationaUsm, as before in the era of the 
sophists, has joined hands with its opposite (scepti-cism), 
and become absorbed in the anti-intellectualism 
of the modernists (end of the nineteenth and beginning 
of the twentieth century) 
{c) The school of Aristotle and St. Thomas teaches 
that truth is neither impossible nor easy, but difficult 
for man to attain. 
It is thus radically opposed alike to scepticism and 
to rationalism. It sees in the multitude of errors put 
1 Naturalism rejects divine teaching in these two different ways : 
( 
I ) It denies God the right to teach men truths in themselves inaccessible 
to the unassisted reason {supernatural mysteries). (2) It also denies him 
the right to teach men by revelation truths in themselves accessible to 
unassisted reason (truths of the natural order, philosophic truths—for 
example, the immortality of the human soul) which reason can indeed 
discover by its unaided powers, but always with the risk of mingling 
error with truth, whereas revelation brings them within the reach of 
all, easily and without any admixture of error. 
183
AN INTRODUCTION TO PHILOSOPHY 
forward by men and particularly by philosophers a 
sign indeed of the weakness of the human under-standing, 
but a reason to prize the intellect the more 
dearly and to embrace truth the more ardently, and 
an instrument for the advancement of knowledge by 
the refutations and explanations which these errors 
call forth. And, on the other hand, it recognises that 
reason is our sole natural means of attaining truth, but 
only when formed and disciplined, in the first place 
and pre-eminently by reality itself (for our mind is 
not the measure of things, but things the measure of 
our mind), secondly by teachers (for science is a 
collective, not an individual, achievement, and can 
be built up only by a continuous living tradition), and 
finally by God, if he should please to instruct mankind 
and bestow upon philosophers the negative rule of 
faith and theology.* 
Philosophy of Aristotle and St, Thomas 
{Moderate Intellectualism) 
That which really is the cause of truth in the mind. 
Reason is capable of attaining with complete certainty 
the most sublime truths of the natural order, but with 
difficulty and only when duly disciplined. 
Error by Defect Error by Excess 
Reason is incapable of Reason attains truth in 
attaining truth, which every sphere easily and 
either is wholly inaccessible without any need of sub-to 
man (scepticism) or must mitting to any external 
be sought otherwise than discipline (rationalism). 
by the intellect (anti-intel-lectualism). 
Synthesis of these Two Errors 
The mind of man makes the truth of that which he 
knows (namely, phenomena), and that which really is, 
the thing in itself, is unknowable by reason (criticism 
or Kantian agnosticism). 
1 See above, p. 124. 
184
CRITICISM (EPISTEMOLOGY) 
Another question, among those with which criticism 
is concerned, demands consideration here. The 
intellect or reason being the instrument of philosophy, 
what is the formal object of the intellect, to which 
intellectual knowledge relates directly and in itself? 
To answer this question it is suflficient to ask oneself 
whether there does not exist an object which is always 
present to the mind when the intellect functions ? 
Such an object does exist. Whatever I know by my 
intellect, there is always some being or mode of being 
present to my mind. There is, however, nothing else 
except being which is always present in this way. 
If, for example, I think of a quality, a magnitude, or a 
substance, in all these cases alike I think of some being 
or mode of being ; but there is nothing except being 
which is common to these three objects of thought, 
and therefore present in all three alike. We therefore 
conclude that being is the formal object of intellect, 
that is to say, the object which it apprehends primarily 
and in itself (per se primo) and in function of which it 
apprehends everything else. 
To know the cause of a thing, its purpose, origin, 
properties, and relations with other things, is in these 
various ways to know what it is, to apprehend its 
being under those different aspects. To use the 
understanding without the notion of being arising is 
an impossibility. 
The intellect, moreover, is able to apprehend the 
being of bodies in their sensible appearances (pheno-mena). 
It is thus, for example, that in physiology it 
studies the properties of living organisms in reference 
to causes which themselves belong to the sensible 
order. Of this nature are the sciences of secondary 
185
AN INTRODUCTION TO PHILOSOPHY 
causes or the sciences of phenomena. But the intellect 
can also apprehend the being of things in their first 
principles. This is the function of philosophy as a 
whole, which in turn is subdivided into natural 
philosophy and metaphysics^ according as the being 
apprehended in its first principles by the intellect is 
the being of bodies as such or being simply as being. 
Psychology indeed deals with this question of the 
formal object of the intellect. But the distinctive 
function of criticism is to make clear that the being 
with which we are here concerned is indeed the actual 
being of things, which exists in them independently of 
the knowing mind. To maintain on the contrary 
that the object of our intellect is not the being of 
things but the idea of being which it forms in itself, 
or more generally that we apprehend immediately 
only our ideas,' is to deliver oneself bound hand and 
foot to scepticism. For if that were the case, it would 
be impossible for our mind under any circumstances 
to conform itself to that which really is, and truth 
would therefore be unattainable. Moreover, the 
intellect would stand convicted of falsehood, for what 
the intellect professes to know is what things arc, not 
what its ideas are. In reality ideas, as the conscious-ness 
of every man witnesses immediately, are our 
instruments of knowledge. If, therefore, knowledge did 
not apprehend the things themselves, knowing would 
be an operation or activity without end or object, 
which is absurd. For to form an idea or judgment is 
to know, just as to make use of a knife is to cut. And, 
just as it is impossible to cut without cutting something 
1 The doctrine of Descartes and after him of all subjective 
philosophy. 
1 86
— 
CRITICISM (e PISTEMO LO G y) 
—the end or object of the act of cutting, which is not 
the knife, but the thing cut by it—so it is impossible 
to know without knowing something—the end or 
object of the act of knowing, which is not the idea, 
but the thing known by it.^ 
Conclusion X.—The formal object of the intellect 
is being. What it apprehends of its very nature 
is what things are independently of us. 
From the two truths just enunciated, the intellect is 
a truthful faculty, and being is the necessary and immediate 
object of the intellect, there arises as a corollary a funda-mental 
truth. 
By intelligible we mean knowable by the intellect. But 
to affirm that being is the necessary and immediate 
object of the intellect, and that the intellect attains 
true knowledge, amounts to saying that being, as 
such, is an object of which the intellect possesses true 
knowledge ; that is to say, that it is intelligible. And 
to say that being as such is intelligible is to say that 
intelligibihty accompanies being, so that everything 
is inteUigible in exact proportion to its being. We 
therefore conclude 
Conclusion XL—Being as such is intelligible. 
Everything is intelligible in exact proportion to 
its being. 
* Intellectual knowledge comes into existence by means of ideas. 
But ideas are simply that by means of which {id quo), not that which 
{id quod) we know directly, a pure medium of knowledge, not (unless 
reflexively) an object or term known. This is why we say that the being 
of things is the immediate object of our intellectual knowledge (by 
immediate we mean known without the intermediary of another term or object 
pTiviously known). 
187
AN INTRODUCTION TO PHILOSOPHY 
It must be borne in mind that when we affirm 
that everything is intelligible in exact proportion to 
its being we mean intelligible in itself, to intellect, not 
intelUgible to us, to our intellect. If, indeed, as a 
result of the inferiority of human nature, our intellect 
is disproportioned to a being which exceeds it because 
it is superior to man, that being, though in itself more 
intelhgible, will be less intelUgible to us. This, how-ever, 
is the case with all wholly spiritual natures, and 
pre-eminently of God. In himself he is the most 
intelligible of beings, but his intellect alone is pro-portionate 
to this supreme intelligibility. 
1 88
ONTOLOGY : ESSENCE 
Having in criticism examined and defended the 
principles of knowledge in general, whether scientific 
or philosophic, we can proceed to the study of meta-physics 
in the strict sense, that is to say, the science 
of being qua being. This is the very heart of philo-sophy. 
We have now to consider being as such, and 
the great truths it contains in itself; to inquire how 
it enters into all things without being exhausted by 
any ; to study its inseparable properties, unity, truth, 
and goodness, to which we may add beauty ; and 
finally to treat of it in its activity, and attempt to 
penetrate the nature and modes of causation. 
We must also examine how throughout the entire 
created universe being is divided, whether we consider 
the constitution of all created being (division of being 
into potentiality and act^ essence and existence) or the 
different kinds of created beings (division of being 
into substance and accident). We shall then realise that 
the concepts elucidated by ontology are the key to 
everything else. Certain among them are indeed so 
indispensable that we must consider them here, for 
indeed at every turn we are obliged to invoke the 
primary concepts of essence, of substance and accident, 
oi potentiality and act. Though it is obviously impos-sible 
in a mere introduction to give an analysis and 
complete defence of these concepts, we shall try to 
189
AN INTRODUCTION TO PHILOSOPHY 
establish them with all due care, employing, it is true, 
examples rather than developed arguments, and 
simplifying matters considerably, but following never-theless 
the order demanded by a strictly scientific study. 
Although the notion of being, since it is the first 
and best known of all, is evidently too clear in itself 
to admit of definition in the strict sense, the first task 
incumbent on a man who wishes to think seriously 
is to clarify this notion in his mind, and with that 
object to discover the primary conceptions into which 
it is divided.' We shall therefore begin by asking the 
following question : What are the objects of thought 
which inevitably and from the very outset impose 
themselves upon the intellect when it considers being 
as such, or to put it in another way, since being is the 
primary object of intellect, what are absolutely the 
first data of the intellect ? * 
We shall see that this one fundamental question 
admits of three answers according as we adopt the 
1 Cf. Aristotle, Metapk., v. 
' The notions explained in pp. 1 9 1 sqq. present some difficulty to 
beginners on account of their extremely abstract character. It is, 
however, impossible to omit them, for they are literally of primary 
importance. And in particular we are convinced of the urgent 
necessity to define with the utmost care, from the very outset, the 
fundamental concept of essence. Materials for the study of this concept 
are scattered in different places, but is it not because we have forgotten 
to collect them that the term essence, when we meet it on the threshold of 
metaphysics, arouses to-day such suspicion, and, even if it forces itself 
on our acceptance, leaves such vagueness in the mind ? 
The student must therefore devote particular attention to the study 
of the notions here explained, without, however, attemjiting to 
comprehend them perfectly. For the moment it will be sufficient to 
make his first acquaintance with them. Later when he meets them 
again in ontology, after he has become more familiar with philosophy, 
they will seem much easier. 
190
ontology: essence 
. 
standpoint of intelligibility, of existence, or of action. 
The consideration of the first of these standpoints will 
lead us to determine what is meant by essence, of the 
second to determine what is meant by substance (as 
opposed to accident), of the third to determine what 
is meant by act (as opposed to potentiality) 
ESSENCE 
We shall first consider being from the standpoint 
of intelligibility ; that is to say, we shall consider 
being so far as it is adapted to enter the mind, or is 
capable of being apprehended by the intellect. This 
is the most universal standpoint we can adopt, for 
we have seen that being as such is intelligible and 
hence that intelligibility is co-extensive with that 
which is or can be. To the primary datum of the intellect 
from this point of view we shall give the name of 
essence. 
(a) To consider being from the standpoint of 
intelUgibiUty or as it is capable of being apprehended 
by the intellect is in the first place to consider it so far 
as it can be simply presented to the mind without 
affirmation or negation—in so far as it can be the 
object of simple apprehension, as we shall term it later. 
Triangle, polygon, seated, this man, are so many objects 
simply presented to the mind without being affirmed 
or denied. 
From this point of view the primary datum of the 
intellect is quite simply that which is placed at tlie 
outset before our mind when we form the conception 
of anything, or conceive the idea of it. Since we have 
agreed to use the term essence in this sense, we conclude 
191
AN INTRODUCTION TO PHILOSOPHY 
that an essence is that which in any object of thought 
whatsoever is immediately and primarily {per se primo) 
presented to the intellect : id quod in aliqua re per se 
primo intelligitur. 
Every idea whatsoever, unless it be, like the idea of 
a square circle, a pseudo-idea involving a contradiction, 
brings immediately before the mind something. 
The something thus immediately presented to the 
mind is an essence or a nature. When I think of man, 
humanity, animal, goodness, white, whiteness, seated, triangle, 
etc., each of the objects thus immediately presented 
to my mind, each of these intelligible units is by 
definition an essence in the wide sense of the word.^ 
An essence therefore is simply an object of thought 
as such. Every essence, however, possesses its intelli-gible 
constitution which distinguishes it from others 
and involves certain attributes. 
Here, however, an important observation must be 
made. If I consider the triangle with its properties, 
man, humanity, etc., they remain exactly what they are 
as objects of thought, whether I suppose them actually 
to exist or not. The fact of existence does not in any 
way affect essences as such. To conceive them I 
1 We have already seen that the individual as such is not directly 
apprehended by our intellect. When indirectly, by a reflection on 
the images (see above, p. 172), we form an individual concept, the 
object presented to our mind by this concept, Peter, this man, this tree, 
is abo, so far as it is an object of thought, an essence in the wide sense 
of the term. That is to say, the concept of essence in the wide sense 
must be extended even to individual objects of thought. As for those 
conceptual beings {blindness, for example, or nothingness) which present 
nothing that really exists, the name essence is inapplicable to them for 
the reason that a privation as such has obviously no essence. (See 
St. Thomas, De Ente et Essentia, i.) Nevertheless, from our present 
standpoint, we may call them improperly essences, in the wide sense. 
192
ontology: essence 
abstract from the fact that they do or do not actually 
exist. 
We thus perceive that being in the sense of existence 
and being in the sense of essence belong to two distinct 
categories.^ The term being has two wholly different 
meanings. For example, in the quotation "to be 
or not to be, that is the question," being means existence^ 
but on the contrary in the phrase a living being it means 
essence. In the first case the term being signifies the 
act of being, the act, if I may so put it, which posits 
a thing outside nonentity, and outside its causes 
{extra nihil, extra causas) ; and in the second case it 
signifies that which is or may be, that which corresponds 
to some existence actual or possible. We may therefore 
say that being is didded into essence and existence. 
BEING (that which is : essence in the wide sense (essentia) 
(ens)  
{entitas) act of being : existence (existentia) 
The relationship which obtains between these two 
terms is a problem which we shall study later ; it is, 
beyond question, not simply with reference to ourselves, 
Uke the problem of universals, but in itself the 
fundamental problem of philosophy : are essence 
and existence really distinct in all things except God ? 
Actual existence, the fact of existing actually, is not 
^ Observe that in existence itself we may distinguish two things : 
existence as the fact of existing {existere in actu exercito) and existence 
as an object of thought {existentia ut quod quid est) . Regarded from the 
latter point of view existence itself assumes the objective status of every 
object of thought and confronts the intellect as a particular essence or 
quiddity. Esse dupliciter sumi potest, scilicet in actu exercito ipsius existentiae, 
et per modum quidditatis ; et ut exercet existentiam, addit supra seipsum ut quod 
quid est ; et consequenter ut objectutji intellectus est abstractius {quam ut objectum 
voluntatis) : quia est objectum voluntatis secundum quod stat in actu exercito 
existentiae, intellectus autem secundum quod rationem habet quidditatis cujusdam 
in seipso. Cajetan, in I, q. 8a, a. 3.
AN INTRODUCTION TO PHILOSOPHY 
included in the object of any of our ideas as such. 
Our intellect can ascribe actual existence to a particular 
object of thought only by basing itself directly or 
indirectly by means of ratiocination on the witness 
of our senses (or reflexively of our consciousness). 
Thus it immediately judges sensible objects exist, I 
exist, and demonstrates the existence of God by 
arguing, for example, from the observed fact of motion. 
It cannot by itself alone attain the actual existence of 
the objects of its thought. 
Those essences, on the other hand (such as triangle, 
even numbers, humanity), which of their nature connote 
only a possible existence (for which reason they are 
also called possibles) , are data furnished immediately by 
our intellect and ideas. 
We must now examine more closely this notion of 
essence, or being understood as that which is or can be. 
We have just defined an essence : that which in any 
object of thought whatsoever is immediately and 
primarily presented to the intellect : id quod in aliqua 
re per se primo intelligitur. Let us see whether this 
extremely wide concept (for it is applicable to any 
object of thought) may not be subdivided and 
qualified in such a way that the same definition taken 
in a more restricted sense shall henceforward be 
applicable, in each particular instance, only to a 
particular object of thought. 
{b) The mere presentation to the mind of an object 
of thought {man, white) is but the beginning of 
intellectual knowledge, which is perfect only in the 
judgment by which the mind affirms or denies this 
object of thought in reference to another {Peter is a 
man, this flower is white). If then we would consider 
194
ontology: essence 
being from the standpoint of intelligibility, to discover 
what is from this point of view the absolutely primary 
datum of the intellect, we must consider objects of 
thought so far as they can be apprehended by the 
intellect when it judges, for example when it affirms 
that Peter is a man. From this point of view which, 
among the various objects of thought which can be 
realised in a given subject, is that which the intellect 
apprehends immediately and before everything else ? 
We shall call it essence in the strict sense of the term. 
Consider any object of thought, for example, Peter, 
Paul, this dog, this bird: Peter is tall, Paul is laughing 
and moving, this dog is barking, this bird is flying. Each 
of these is a particular whole, individual, concrete, 
and independent, completely equipped for existence 
and action. 
It is individual subjects of this kind that our mind 
apprehends before anything else (from the standpoint 
of existence) when we think of that which is. When 
applied to objects of this kind the expression that which 
is acquires a more definite and special force. It no 
longer simply means that which corresponds to some 
actual or possible existence, but that which fulfils in 
the strictest sense and before everything else the act 
of being. These objects are all, though in very 
diverse respects, actors in the drama of the universe. 
When, however, we adopt the standpoint of 
intelligibility, our mind does not among the different 
objects of thought which things can present apprehend 
in the first place these individual subjects as such. 
On the contrary the individual, as we have seen above, 
escapes the direct grasp of the intellect. What I 
know of Peter is what I know he is—for example, a 
195
— 
AN INTRODUCTION TO PHILOSOPHY 
man. It is such objects of thought as man or humanity 
which it perceives in Peter, or such as white or whiteness 
which it perceives in thisflower, it is what a thing is that, 
from this point of view, our mind primarily apprehends, 
and it is therefore in tliis direction that we must look 
for the absolutely first datum of the intellect in relation 
to intelligibility (essence in the strict sense). 
The concept of essence in the wide sense has thus 
been subdivided into two. There is in the first place 
that which in the strict sense is, that which. And in 
the second place what a thing is, what. 
{that which is : essence j what 
in the wide sense  that which 
act of being : existence 
That which is in the strict sense we shall entitle the 
primary subject of existence and action. It is what 
philosophers also term suppositum and person. For the 
moment we may neglect it, for, as we shall see, it 
does not concern our present inquiry. 
Let us, on the contrary, consider what a thing is. In 
the notion of what a thing is there are further distinc-tions 
and exclusions to be made, to determine more 
precisely what is actually the absolutely primary 
datum of the intellect from the standpoint of intelHgi-bility 
and therefore deserves to be entitled essence in 
the strict sense of the term ; what, for example, is 
the essence of Peter. Peter is seated. Peter is capable of 
laughter. Peter is a man. Is what is here predicated 
of Peter 
seated, capable of laughter, man—in each of 
these three cases, or in one alone, the being which 
the intellect apprehends in Peter immediately and 
primarily from the standpoint of intelligibility ? We 
said above that every object of thought is, as such, 
196
ontology: essence 
an essence {essence in the wide sense). Now we are 
studying what Peter is, and inquiring what is the 
object of thought which constitutes the essence of 
Peter {essence in the strict sense). 
The following are the characteristics of the object of 
thought thus defined, that is, of the being primarily 
apprehended by the intellect when it considers what 
a thing is. 
It is at once plain that the being to which the 
intellect is directed in the first place when it thinks 
what a thing is, is a being which the intellect cannot 
conceive that thing lacking or deprived of. 
It is in fact in terms of that being that the intellect 
immediately conceives, apprehend?, grasps, sets before 
itself and names the object in question. To deprive 
the thing of that being, or to alter its constitution in 
any way, would be to set before the intellect, by 
definition, a different thing. 
It is thus a being which that thing so far as it exists 
cannot lack or be deprived of (otherwise the intellect 
would not be truthful). For example, Peter, so far as 
he exists, cannot be other than a man ; on the other 
hand, he can be not seated. 
The being in question is therefore a being which 
the thing considered by the intellect is necessarily and 
immutably. 
Moreover, it is obviously the being which in the 
thing possesses primary importance for the intellect, 
since it is that to which the intellect is first directed. 
It is thus the being which before anything else ^ the 
It Is plain that the word before denotes in this connection a priority 
of nature, not of time.
AN INTRODUCTION TO PHILOSOPHY 
object is, and is, so to speak, the ground of what the 
object is in other respects. It is the first being of the 
thing. Peter, for example, is a man before he is 
capable of laughter or mortal. 
We conclude that the being to which the intellect is 
in the first place directed when it thinks what a thing 
is, is its necessary and first being, or, in short, the being 
which constitutes the thing, what it necessarily and 
primarily is. 
This is the first characteristic of what we h^ve 
agreed to call essence in the strict sense. 
There is a second. It was the standpoint of intelligi-bility, 
it will be remembered, which we adopted 
when we undertook this study of essence. Peter is a 
man {rational animal) before being mortal. That is, 
man includes animal, and in the notion of animal the 
intellect finds the necessary characteristic, mortal. 
The characteristics mortal and capable of laughter— 
necessarily possessed by Peter—have in him a principle 
and ground, which by its very notion, or by what 
it is, or its own intelligibiUty, compels the intellect to 
posit these characteristics, and this principle or 
ground is one of the elements or aspects which con-stitute 
the being man. It is from the standpoint of 
intelligibility that Peter is a man before being mortal 
or capable of laughter. 
Thus if the being man is, as we have said, first, it is 
in the order of intelligibility that it is first. In other 
words, it is in Peter the first principle of intelligibility.^ 
However long our formula, we must say, if we would 
express this truth exactly, that the being man is in 
1 Non enim res intelligibUis est nisi per suam definitionem et essentiam, 
St. Thomas, De Ente et Essentia, i. 
198
ontology: essence 
virtue of its constituent elements or aspects the root 
of all the characteristics necessarily possessed by Peter ^ 
which have in Peter a principle which by its very notion 
requires them. 
This, then, is the second characteristic of what we 
have agreed to call essence in the strict sense ; of the 
being to which the intellect is directed in the first 
place when it considers what things are. It is in the 
thing the first principle of intelligibility. 
Our intellect apprehends this being which is the 
first principle of intelligibility in two ways, one 
imperfect, the other perfect. 
If, for example, we know that an object is a man, 
without, however, being able to state what man is, we 
possess a confused knowledge of the being in question. 
Our intellect grasps that being, has truly apprehended 
it, and really perceives it, but, so to speak, after the 
fashion in which our eyes see an opaque object. 
If, however, we know this same object and are able 
to define what it is {an animal endowed with reason) we 
now possess a distinct knowledge of the being in 
question. 
Our intellect not only perceives it, but also perceives 
its principles or constituent aspects. 
In the first instance the being in question is pre-sented 
to us imperfectly, in the second perfectly, with 
the perfection demanded by science, so that we can 
employ it as a first principle of intelligibility. (For 
example, from the knowledge that this thing is endowed 
with reason I can deduce that it is capable of speaking, 
laughing, worshipping God, etc.) But in both instances 
it is obviously the same being which is presented to 
1 These characteristics are termed properties.
AN INTRODUCTION TO PHILOSOPHY 
US. Therefore, though I do not yet know or even 
can never know this particular being distinctly as a 
rational animal, in itself it will be none the less (though 
in this case I do not know how) in virtue of its 
constituent elements the root of all the characteristics 
which possess in Peter a principle requiring them by 
its very notion ; it will be none the less in itself the 
primary being of the thing as the first principle of its 
intelligibility. 
We now know what are the characteristics of essence 
in the strict sense and are in a position to define it 
as follows : Essence is the necessary and primary 
being of a thing as the first principle of intelligibility, 
or, in other words, what a thing necessarily and 
primarily is as intelligible, in short, the primary 
intelligible being of a thing : id quod per se primo 
intelligitur in aliqua re.^ 
fthat which is : (what a thing is primarily 
i 
essence in the 
essence in the - as intelligible i strict sense 
BEING-! wide sense that which : ihc sih}tc. oi 
action {suppositum, person) 
act of being : existence 
1 Essence considered as an attribute of the thing (for example, man 
when we say Peter is a man) is strictly what the thing is necessarily and 
primarily as intelligible. Essence considered separately and in the 
pure state (for example, when we speak of humanity or the being man, we 
cannot say Peter is humanity or Peter is tlie being man) is strictly that in 
virtue of which a thing is what it is necessarily and primarily as intelligible, 
or, to put it in another way, that in virtue of which it is constituted 
in a determinate degree of primarily intelligible being. If therefore we 
consider essence in the pure state, we must substitute in our synopses 
for the expression what the expression that in virtue of which : 
(that which is : (that in virtue of which a thing sessence in the 
essence in the ! what it is primarily as intelligible/ strict sense 
wide sense that which : the subject of 
action {suppositum, person) 
nct of being : existence 
200
ontology: essence 
Conclusion XII.—The essence of a thing is what 
that thing is necessarily and primarily as the 
first principle of its intelUgibility. 
This primary datum of the intellect is termed by 
philosophers not only the essence but also the quiddity 
and nature. It is what Aristotle and the schoolmen 
called the to xt, ^v elvat, the quod quid est,^ and which 
they defined as id quod per se primo intelligitur in aliqua re,'^ 
a definition with which we were already acquainted, 
but to which we have now attached a completely 
definite sense. 
The definition, when used of essence in the wide 
sense, meant what a particular idea first presents to the 
intellect. When employed of essence in the strict sense, 
it means what a particular ?,ih]tct primarily isfor the intellect. 
Observe that every object of thought, every essence 
whatsoever [essence in the wide sense) is in fact the 
essence of something [essence in the strict sense) appre-hended 
more or less completely (in some or other of 
its properties) . When I think of animal, I apprehend 
the essence of^ Peter in one part of its properties. When 
I think of man I apprehend it as a whole. When I 
think of Aryan, Breton, or Peter I apprehend it as a 
whole with the addition of certain characteristics 
or attributes derived from the matter (see below, 
pp. 207-2 1 6) . When I think of a living body endowed with 
sensibility I apprehend the entire essence of the subject 
1 The Latin equivalent of the Greek term is quod quid erat esse—as 
St. Thomas explains {De Ente et Essentia, i), id est Iwc per quod aliquid habet 
esse quid, that which makes any object of thought this or that particular thing. 
2 Or stated more fully : id quod primo in re concipitur, sine quo res esse 
non potest, esique fundamentum et causa ceterorum quae sunt in eadein re : ut 
animal rationale est hominis essentia. 
201
AN INTRODUCTION TO PHILOSOPHY 
animal (and at the same time the essence of the subject 
Peter in one part of its properties) . When I think of 
white or prudence, I perceive the essence of a particular 
quality. When I think of goodness, unity, being, I 
apprehend a certain created participation of the 
Divine Essence (or I apprehend, by analogy, if I think 
of subsistent goodness, etc., the Divine Essence itself). 
Observe, further, that every subject capable of 
forming part of any proposition whatsoever ^ has an 
essence distinctively its own, whether it be an individual 
subject such as Peter {substantia prima, subject par 
excellence), an abstract and universal subject {sub-stantia 
secunda) such as animal, an accident, for example, 
a particular colour or virtue, or a transcendental, for 
instance the one, the good, etc. 
The primary intelligible being of a thing is called 
essence {essentia) because since the intellect is modelled 
on being, what a thing primarily is for the intellect 
must be that which is of primary importance in it 
from the standpoint of being itself ; in fact, as we shall 
see later, it is by and in its essence that a thing possesses 
being or existence {esse).^ It is called quiddity {quidditas) 
because it is that which the definition expresses and 
declares, which in turn answers the question quid est 
hoc ? What is this ? It is called nature {natura) because 
it is the first principle of the operations for the per-formance 
ofwhich the thing has come into being {nata) .* 
1 With the exception of conceptual beings, which do not, strictly 
speaking, possess an essence (see above, p. 192, note). 
2 Essentia dicitur secundum quod per earn et in ea res habet esse. St. 
Thomas, De Ente et Essentia, i. 
3 Qjtidditas est ipsa rei entitas considerata in ordine ad definitionem explicantem 
quid ilia sit. Entitas vera rei considerata in ordiru ad esse, dicitur essentia ; 
in ordine ad operationem dicitur natura. 
202
— 
ontology: essence 
The formal object of the intellect is being. On the 
other hand, what we have agreed to term essence is 
nothing but the primary intelligible being of a thing. 
Our intellect can, therefore, really apprehend the essences of 
things.^ To deny this would be to deny the intellect 
itself, and to say that it is bound to miss what is 
peculiarly its object. 
Moreover, our intellect claims to give us knowledge 
of the essences of things. The sciences which are its 
work have no other aim than to grasp these essences, 
either distinctly, to deduce from them the properties 
of a thing (as when we know that a particular figure is 
a right-angled triangle, or that Peter is a rational animal), 
or confusedly, simply in order to place a thing in its 
species and describe it (as when we know that a 
particular body is sulphurid acid, a particular plant 
alisma plantago). If therefore our intellect were 
incapable of really attaining the essences of things, 
it would deceive us. It is therefore an absolutely 
necessary consequence of the fundamental axiom that 
our intellect is trustworthy, that 
Conclusion XIII.—Our intellect is capable of 
knowing the essences of things. 
We do not maintain that the intellect always knows 
the essences of things (in the totaUty of their 
properties). The specific essences of things are often 
1 It can also attain directly (by an appropriate concept) these 
complete, i.e. completely determined, essences, at least in the case of 
things immediately accessible to us, namely, bodies. (We attain, for 
example, the complete, i.e. completely determined, essence of Peter, 
when we know Peter not only as a living being or as an animal, but as 
a man.) 
203
AN INTRODUCTION TO PHILOSOPHY 
unknown to us, and undefined. This is due to the 
imperfection of the human intellect. But we maintain 
that our intellect is capable of knowing them—and 
therefore does actually know them in many cases. 
Neither do we maintain that the intellect can always 
know the essences of things perfectly, that is distinctly.^ 
That it can often know them only confusedly * matters 
Uttle. What is certain is that it is capable of appre-hending 
them. The eye, to take a parzdlel case, sees 
the coloured objects within its compass with more or 
less distinct detail ; it may require the assistance of a 
magnifying glass, but it can see them. 
It is important to bear in mind that the experimental 
sciences are very far from being able to know perfectly 
the essence of the things which they study. They are, 
in fact, unable to attain a truly distinct notion of their 
essences, and never possess more than a confused or 
purely descriptive notion of them. They know them, 
so to speak, after the fashion of a blind man by means of 
indirect signs. 
For example, we know distinctly the essence or 
nature man when we distinguish man from the other 
animals by the specific difference endowed with reason. 
But we cannot know in the same fashion how the dog, 
for example, differs from the lion ; we know it only by 
differences of a purely descriptive kind. Often even 
when we have before us a series of concepts of diminish-ing 
generality, for instance, a living body, animal, 
irrational animal, vertebrate, mammal, canine, dog, poodle, 
1 Even so it can determine them only by means of an evident 
character previously known by us (for instance, the faculty of reason in 
human nature) which it perceives to be a necessary factor of its 
constitution. 
* See above, p. J99. 
204
ontology: essence 
etc., ending with Gyp or Fido, we may not know what 
concept {canine ? dog ? poodle ?) designates (in the 
totality of its properties) the essence of Gyp or Fido. 
This, however, does not alter the fact that somewhere 
in the series of concepts in the Ust just given, and any 
others which might be inserted among them, there 
must necessarily be a concept which designates that 
essence. (In fact, in the example we have chosen, 
it is the concept dog, as zoology discovers by indirect 
signs, and without being able to give us a truly distinct 
knowledge of the essence thus apprehended.) 
When we think of man, for example, or any other 
object directly presented to the mind by a human idea 
(an abstract idea), we put before ourselves something 
stripped of individuality, something which, being 
apprehended by a single concept, constitutes in our 
mind a single, and solitary object of thought—which 
is therefore in our mind something belonging to one 
{man) and capable of existing in many (in all men), that 
is something universal.^ Thus everything directly 
apprehended by an idea of our intellect—and con-sequently 
the essence of a thing—is in our mind as 
a universal. 
No doubt, considered as it exists in reality, the 
essence is individuated, for it is then identical with a 
subject, Peter for example, who is himself individual.* 
1 See above, p. 159. 
2 If, for instance, I can say Peter is a man, it is because the thing (the 
material object) apprehended under the object of thought man is 
identical with the thing apprehended under the object of thought 
Peter. When I thus proceed from the existence of things in my mind 
to their existence in reality, I must say that the object of thought man, 
single in my mind, is multiplied in all tlie individuals in which it is 
realised and is identical with each. 
205
AN INTRODUCTION TO PHILOSOPHY 
But this condition of individuality is no part of the 
very nature or inmost constitution of the essence, does 
not belong to the essence of Peter as such, to its 
character as an essence. If indeed the essence 
considered in itself {secundum se) were individual, our 
intellect could never know it, for everything directly 
apprehended by an idea of the intellect is apprehended 
as a universal. 
Considered in itself {secundum se) the essence is 
neither universal nor individual. It abstracts from 
every condition and mode of existence, being purely 
and simply what the object is primarily as intelligible 
and what the definition expresses.. Thus it is equally 
present in the actual thing, individuated (in order to 
exist) and, in our mind, universalised (in order to be 
known). For example, we see a man only in public, 
therefore in complete dress, whereas in his bedroom 
he wears pyjamas. Nevertheless the man we know, 
when we see him in the street, is the same man, because 
his pyjamas are no more part of his nature than liis 
suit of clothes neither belongs ; to the man considered 
in himself. Similarly considered in itself the essence 
is not universal, but neither is it individual ^ that ; 
is 
to say, the essence as such, the essence of Peter taken 
in itself, abstracts from all the characters which 
distinguish Peter from Paul or John.' 
1 Cf. Aristotle, Metaph., vii, 8, 1033 b 22 ; 10, 1035 b 14. Here wc 
are speaking only of corporeal things, which alone are immediately 
accessible to us (being connatural to the human intellect), consequently 
the only things whose essence is directly knowable (otherwise than by 
analogy) and can be known complete, i.e. completely determined. 
2 From all that has been said it follows that, when, for example, 
we say Peter and Paul possess the same essence or the same nature, the word 
same refers to the essence of Peter and Paul as it exists in the mind (for 
then it is one and the same object of thought), not as it exists in reality 
206
. 
ontology: essence 
Conclusion XIV.—The essences of things are 
universal in the mind, and considered in them-selves 
neither universal nor individual. 
This proposition is of the first importance. To deny 
it inevitably involves suspicion of the human intellect, 
which cannot directly apprehend in its concepts the 
individual as such ; 
^ we shall either demand from it 
what it cannot give, a knowledge strictly superhuman 
—intellectual intuition of the individual—or deny 
its objective reference and fall into subjectivism. 
We must therefore bear firmly in mind that to know 
the essence or nature of anything it is not necessary to 
know the principles which constitute its individuality,* 
since the essence, considered in itself, is, in fact, 
nothing individual. Misconception of this fundamental 
(for then it is identical with Peter and with Paul, two different 
individuals). But since the essence in question is not individual in 
itself {secundum se), in other words is not distinct in Peter and in Paul qua 
essence, it foUovre that it is in Peter and Paul such that it can be appre-hended 
by the mind in a single concept and constitute in the mind 
one and the same object of thought. This is expressed by the statement 
that the essence formally universal in the mind is fundamentally universal 
in things or in reality. (The nature of anything exists in the mind 
either in a condition of logical or formal universality, as, for example, 
when we say man is the species of Peter and Paul, or in a condition of 
metaphysical or fundamental universality, as when we say man is 
mortal. The term fundamental here refers to the proximate foundation 
of universality. When, on the other hand, we say that the nature or 
essence is fundamentally universal in reality, we are speaking of the 
remote foundation of universality) 
1 We are now speaking of things known by our intellect and do not 
deal with the question how the intellect knows its own individual and 
material act. 
2 Cf St. Thomas, De Verit., q. 2, a. ^, ad i ; Intellectus noster 
singularia non cognoscens propriam habet cognitionem de rebus, cognoscens eas 
secundum proprias rationes speciei. 
207
AN INTRODUCTION TO PHILOSOPHY 
truth is at the bottom of the errors of several great 
modern metaphysicians, Spinoza, for example, and 
Leibniz (exaggerated intellectuaiists) , also Bergson 
and the anti-intellectualists of the present day. 
(a) The essence of corporeal things is universal in 
the sense just explained. That is to say, in this category 
of being there are a multitude of individuals possessing 
the same essence. 
Individuals possessing the same essence, for example, 
Peter, Paul, and John, are on the same level in respect 
of primarily intelligible being ; they are essentially 
equal. 
Nevertheless these individuals differ one from 
another. Peter is fair, short, and sanguine, John dark, 
tall, and choleric, etc.^ Such characteristics peculiar to 
a particular individual are not derived from the 
essence. Otherwise they would be identical in all 
the individuals which by hypothesis possess the same 
essence. They are therefore non-essential characters. 
Nevertheless they are, in fact, unalterable and 
necessary. 2 If he were not fair, sanguine, etc., Peter 
would no doubt be a man, but he would not be Peter. 
We must therefore conclude that these characteristics 
have their ground in what the object is necessarily and 
primarily, but as an individual, or, in what we may 
term the individual nature of the thing. (By individual 
1 We are not now speaking of those purely contingent characteristics 
which distinguish one individual from another, for example Peter is in 
Paris, Paul in Rome, Peter is rich, Paul poor, etc. We are speaking of 
those characters which arise out of the constituent being of the individual, 
the innate characters, which are, radically at least, unalterable. 
* But in a fashion altogether different from the characters derived 
from the essence (properties). See below, p. 212, note. 
208
ontology: essence 
nature we mean incommunicable to any other object 
or, if you prefer, wholly circumscribed.) 
{by In this individual nature we find, as in the 
essence, the notes necessary and first being. But, on the 
other hand, and this is the important point, it is not 
the necessary and first being of the thing as first 
principle of intelligibility ; it is not the first principle 
of intelligibility. The individual characteristics such 
as fair, sanguine, etc., are not, as we pointed out, 
derived fi'om Peter's essence ; they are not required 
by it. That is to say, they do not possess in Peter a 
principle or ground which requires them by its ver/ 
notion, or in virtue of what it is, that is to say, of its 
own intelligibility (as, for example, rational requires 
capacity for laughter). Nevertheless, since they are 
necessarily possessed by Peter, they have their root 
in Peter, in Peter's individual nature ; they have 
there a principle. 
They must then have as their principle something 
which does not require them by its very notion, 
in virtue of its being or of its own intelligibility, 
something in the notion of which the intellect cannot 
discover a necessity for these characteristics rather 
than any others. Therefore, his principle is in itself 
wholly indeterminate. If neither by its notion nor 
in virtue of its being or its own intelligibility it requires 
this rather than that, it is because in itself it has no 
notion, being, or intelligibility. We are thus led to 
a principle which of itself is absolutely nothing 
conceivable, to first matter as Aristotle understood it, 
something which can enter into the constitution of 
a being, but is not itself a being. 
209
AN INTRODUCTION TO PHILOSOPHY 
If it be admitted that non-being of this sort is part of 
all corporeal things, and that when itself individuated 
by some qualification ^ it is the primary root of their 
individuaUty, it is easy to see that the characters 
which are derived from the individual nature of the 
thing, since their primary root is the individual 
matter with the dispositions it happens to possess at 
the moment when the thing comes into existence, 
have, as their first principle in the thing, a principle 
which does not require them by its very notion—for 
in itself it has neither notion nor intelligibility ; it 
requires them solely in virtue of the accidental 
dispositions it happens to possess at a particular 
moment. 
Thus, the individual nature is not the first principle 
of intelligibility, because it is by its matter that it is the 
principle of the individual characters.' 
{c) We have merely sought to indicate here how the 
obscure notion o^first matter, the study of which belongs 
In so far as it materia signata quantitate. Obviously beings wholly 
incorporeal or immaterial {pwre spirits) cannot derive their individuation 
from first matter. They must therefore be individuated by their 
essence itself, and each individual in consequence differs from the rest 
as a horsef for example, differs from a man, each being by himself a specific 
essence. For this reason in the order of pure spirits there are no two 
beings essentially equal. And consequently in the case of pure spirits 
(but only in their case) the essence is something individual and the 
concept of complete essence identical with that of individual nature. 
* To avoid any possible confusion be it observed that an individual 
nature is not unintelligible in itself. It is first matter that is unin-telligible 
in itself. Though the individual nature is not the first 
principle of intelligibility, is not the primarily' intelligible being of the 
thing, it is nevertheless the primary principle of its being, for it is the 
essence as individuated by matter, and is therefore intelligible in 
itself. That is why an intelligence more perfect than ours, the Divine 
Mind for example, can know it directly. 
210
ontology: essence 
— 
to natural philosophy, arises naturally in the mind 
as soon as it is understood that considered in itself 
the essence of corporeal objects is not individual, a proposition 
itself demanded by the fundamental axiom of the 
trustworthiness of the intellect. 
We may further point out that since matter, this 
species of non-being, is present as the ground of 
individuation (and consequently as the primary root 
of certain qualifications)^ only in the individual nature 
(in Peter's nature as such) and not in the essence 
{humanity), we may regard the essence, the primary 
intelligible being, as free from all the qualifications due 
to matter as their primary root or as immaterialised 
being, ^ in other words as the archetypal being of the thing,' 
1 That is the sense of St. Thomas's dictum : formae et perfectiones rerum 
per materiam determinantur {De Verit., q. 2, a. 2). 
2 Aristotle, Metaph., vii, 7, 1032 b 14 : Xtyu 5'ovalav (Lvev CXrji rb tI Jjv 
ilvai. 
^ Cf. St. Thomas, De Ente et Essentia, ii : Haec materia (signata) in 
definitione honiinis in quantum homo non ponitur ; sed poneretur in definitione 
Socratis, si Socrates definitionem haberet. In definitione autem hominis ponitur 
materia non signata . . . from this it follows : 
(i) That Socrates possesses his essence not precisely as Socrates, but 
as man, for the essence is that which the definition expresses {cf. De 
Verit., q. 2, a. 2, at/g) and Socrates, as Socrates, is indefinable, Socrates's 
individual nature is the essence of man individuated by the materia 
signata. 
(ii) That essence taken in the pure state or separately, as for instance 
when we speak of humanity or the being man, may be regarded as the 
immaterialised being (stripped of the qualities derived from materia 
signata), or as the formal being of the thing as a whole (comprising both 
matter—not individual—and form). It is in this sense that the ancients 
gave to the essence (itself comprising the matter—not individual 
and the form) the name oiform {forma totius) : Et ideo humanitas significatur 
ut forma quaedam. Et dicitur quod est forma totius . . . sed magis est 
forma quae est totum, scilicet formam complectens et materiam, cum praecisioru 
tamen eorum per quae materia est nata designari (St. Thomas, De Ente et 
Essentia, 3). It is important to observe that, although the individual 
matter (e.g. haec ossa, hae carnes) is no part of the essence or specific 
211
AN INTRODUCTION TO PHILOSOPHY 
an ideal being which in the pure state or separately 
has no existence except in the mind, and exists in 
reality only as individuated by matter, in the concrete 
state of the individual nature. 
We must therefore conclude that there is nothing 
more in the individual nature than in the essence from 
the standpoint of primarily intelligible or archetypal 
being.^ From this point of view all individuals of 
the same species are on the same level of being ; to 
nature, on the other hand the unindividuated or common matter 
{ossa, carnes) is part of it. What constitutes man is not the soul alone, 
but soul and body together. {Cf. Aristotle, Metaph., vii ; St. Thomas, In 
VII Metaph., 1. lo, 1492 and 1496, ed. Cathala). This unindividuated 
or common matter, taken simply as receiving the form and determined 
by it, and not as the primary root of certain characters (the individual 
characters) of the subject, is made known to us by the form : materia 
cognoscitur per formam, a qua sumitur ratio universalis (St. Thomas, loc. cit. 
Cf. De Verit., q. 10, a. 4 fi? 5, and is part of that which we have here 
termed the immaterialised being (archetypal being) or formal being of the 
thing (forma totius, seu potius forma quae est totum). 
^ The individual nature contains more (the qualities peculiar to the 
individual, for example, a particular temperament) than the essence, but 
only from the standpoint of matter, not from the standpoint of purely 
intelligible or immaterialised being. The individual characters are 
no part of that being and add nothing to it in its own order. 
Remark in this connection that the individual characteristics (fair, 
sanguine, etc.) from the very fact that they are derived from the matter, 
are necessary, and unalterable in a totally different sense than are the 
characters derived from the essence (properties) . The latter are necessary 
de jure, as derived from a prmciple constitutive of the essence which 
demands them in virtue of its very concept ; it is absolutely impossible 
that Peter should exist without being mortal. The individual characters, 
on the contrary, are only necessary de facto, as derived from particular 
dispositions of the matter which they presuppose. If it is impossible 
that Peter should exist without possessing a particular temperament, 
the existence of that characteristic presupposes certain material condi-tions 
in virtue of which Peter possesses a particular individual nature, 
but which are not themselves necessary. Hence these characters 
can be to a certain extent modified, and are unalterable only in their 
ground. 
212
ontology: essence 
know their (universal) essence is to know all there is to 
know in them, for the being of Peter as Peter is no more 
complete or determinate than the being of Peter as 
man. It is merely more closely circumscribed. 
We can now understand how, although the human 
intellect cannot directly know the being of objects 
in its individuality, its nature as an intellect is not 
frustrated on that account, nor does it miss its formal 
object, for it truly knows the being of things so far as 
it is primarily intelligible or archetypal being.^ Hence, 
though imperfect, it is neither useless nor untrust-worthy. 
{d) Be it observed that the synonyms essence, 
quiddity, and nature, all of which denote a universal, 
may be stretched to denote some:hing singular, 
when we consider the essence {humanity, for example) 
as individuated by matter (in Peter, for instance), or 
as it possesses in reality a singular mode of existence. 
Nevertheless, strictly speaking, the term nature alone 
is compatible with the predicate individual, whereas 
the expressions individual essence or individual quiddity 
are incorrect.^ 
In all this we have in mind things immediately accessible to us, 
namely corporeal things, which the human intellect cannot apprehend 
directly in their individuality, because, since it is obliged to abstract 
from images its wholly immaterial ideas, it is by that very fact compelled 
to abstract from that which constitutes the materiality of sense know-ledge, 
namely the individual m.atter. 
As regards immaterial things {pure spirits), our intellect is equally 
incapable of apprehending them in their individuality, but for an 
entirely different reason ; because pure spirits are not immediately 
"ccessible to us, and we can know them only by analogy, not in their 
essence, and are unable to apprehend their complete essence. 
^ So far at least as the order of corporeal things is concerned. In 
the order of pure spirits, on the contrary, the essence is individual (see 
213
AN INTRODUCTION TO PHILOSOPHY 
We have indeed seen that the terms essence and 
quiddity are used in reference both to the existence and 
the definition of anything. The definition, however, 
can express only the primarily intelligible being of the 
thing, for it states its constituent elements which are 
by their very notion principles of intelligibility in it. 
Hence the definition cannot express the material 
individuating principles of the thing, and for that 
reason the individual nature as such is indefinable. 
Therefore, since the quiddity, what the object is as definable, 
can only consist in the primarily intelligible being 
of the object, it must be universal. Similarly, that in 
virtue of which anything invites that supreme perfec-tion 
which consists in existence can clearly be nothing 
but its immateriaHsed being. For it is not in virtue 
of that in it of which the first principle is matter 
that it invites existence. Its individuality is merely a 
condition in which it must be in order to exist. And 
since the essence, what a thing is, taken precisely as that 
in virtue of which it receives existence, can consist 
in nothing but its immateriaUsed being, it must be 
universal. 
The term nature, on the contrary, is used in reference 
above, p. 2 1 o, note i ) . And if we know the essences of spiritual beings 
after the fashion of a universal, it is because we only know them 
inadequately and by analogy with the corporeal objects previously 
known. 
The expression individual nature is not uncommon in St. Thomas {cf. 
De Verit., q. 2, a. 5, nature singularis ; Sum. Theol., i-ii, q. 51,3, i, 
natura individui, etc.). He also uses, though exceptionally, the 
expression essentia singularis {cf. De Verit., q. 2, a. 7). Whatever may 
be thought of the propriety of the term, in any case St. Thomas 
understands by it simply the essence individuated by the matter (not 
in Spinoza's sense, the essence complete, as an essence, only in the in-dividual). 
214
ontology: essence 
to the operations which anything is adapted to 
perform. A thing, however, does not act solely in 
accordance with its archetypal or primarily intelligible 
being, but also as it is subject to particular material 
conditions and possesses a particular individuaUty. 
Nothing therefore prevents our diverting the term 
nature from its primary significance to denote 
secondarily what a thing is as individual. 
(what primarily as intelligible : essence, 
quiddity, nature 
what as wholly circumscribed (essence 
individuated by matter) : individual 
nature 
[that which is : 
essence in the 
wide sense 
BEING -. 
 that which : the subject of action 
{suppositum, person) 
^act of being : existence 
Finally we may remark that in a series of concepts 
such as substance, living body, animal, man, Aryan, Breton, 
etc., only the concept man, strictly speaking, denotes 
Peter's essence. The concepts substance, living body, 
animal, denote only certain elements or intelligible 
aspects which enter into the constitution of that 
essence ; in other words, they denote that essence only 
in one part of its qualifications, and the concepts 
Arjyan or Breton only as circumscribed and differentiated 
by certain additional notes arising from the dispositions 
of matter. Aryan and Breton are thus, like the essence 
man, universal objects of thought apprehended by the 
mind in the individual Peter and liberated by abstrac-tion 
from the conditions of individual matter ; but 
they are universals whose extension is less than that 
of the essence, and which belong to a particular class 
215
AN INTRODUCTION TO PHILOSOPHY 
(race) divided into a multitude of individuals possess-ing 
the same essence ; and, since they can be 
distinguished only by means of characters rooted 
in certain dispositions of matter they cannot be the 
subject of a notion strictly speaking distinct, or a 
true definition. 
-J 
216
VI 
ONTOLOGY : SUBSTANCE AND ACCIDENT 
Adopting the standpoint of intelligibility, we asked 
ourselves in the preceding paragraph what is the being 
primarily apprehended by the intellect from that 
point of view. We were thus brought to the notioi. 
of essence strictly so-called, or nature (the two terms 
may be regarded as synonymous) ; 
^ what an object 
primarily is as intelligible. 
Let us now consider the being of things no longer 
in reference to intelligibiHty but in reference to 
existence. 
What from this new standpoint is the being which 
immediately presents itself to the consideration of the 
intellect, that to which the latter is directed before 
anything else ? In other words, what is the being 
primarily apprehended by the intellect as existing ? 
We have already answered the question.* What the 
mind apprehends first of all as existing, is beings such 
as Peter, Paul, this man, this dog, this bird, individual 
concrete and independent subjects, fully equipped to 
be and to act, and which we have termed the primary 
subjects of action, supposita, or persons.^ It is they who 
primarily fulfil the act of being. 
1 See above, pp. 201, 213. 
^ See above, p. 195. 
2 The name person is reserved for supposita of an intellectual nature, 
who are therefore masters of their actions and possess the maximum 
of independence. 
217
— 
AN INTRODUCTION TO PHILOSOPHY 
The subject of action may be thus defined from the 
standpoint of existence. It exists wholly by itself alone 
or by its own means, not in the sense that it has no 
need of a cause [Peter has been engendered and many 
causes combine to keep him in being) but in the , sense 
that it is by itself sufficiently disposed to be drawn 
from nothingness by the causes of its being taken 
; separately it possesses in itself or in its own nature 
everything necessary to receive existence.^ In this 
sense we may say that it is a being existing by itself 
[per se) or in virtue of itself, in virtue of its own nature, 
ens per se existens. Since a being of this kind exists as 
a whole and in no wise as part of another being or 
subject in which it exists, we may also say that it exists 
in itself, in se. 
A being which exists per se ^ ,or rather a being 
immediately disposed to exist per se, is thus from the 
standpoint of existence the first datum of the intellect. 
Observe further, that when the intellect makes being 
of this kind its object, it transcends the limits which 
define the essence in the strict sense or the nature {what 
a thing is, or rather—if we take the pure essence, 
abstracted from the subject which possesses it 
that in 
virtue of which an object is what it is) 
. 
We are now concerned, as we have already 
* We are speaking here of created subjects. An uncreated (divine) 
Person possesses in himself everything necessary to exist with an 
underived existence. When we say that the suppositum is in no way a 
part of the whole in which it exists, the term whole obviously means 
a whole tliat is one in itself (see p. 250), not a collective whole, for example 
the universe. 
2 This formula is preferable, because existence itself cannot enter 
as a constituent part into the definition of anything created. See 
further St. Thomas, Qpodlib., 2, q. 2, a. 4, arf 2 : ipsttm esse non est de 
ratione suppositi. 
218
ontology: substance and accident 
hinted, with that which is in the strict sense, Peter for 
example, and not with that in virtue of which Peter 
is what he is {humanity, the property of Peter in 
virtue of which he is a man, or his individual nature, 
the Petrine humanity, so to speak, in virtue of which 
he is Peter). 
' that which is : ( what (or that in virtue of which) 
BEING. 
essence in the J 
wide sense I that which primarily exists : primary 
subject of action {suppositum, person) 
act of being : existence 
To be sure, that which is, Peter, possesses no distinctive 
characters other than those which constitute what he 
is or his individual nature. But when I say Peter I 
conceive this nature as constituting the whole which 
exists in nothing other than itself.^ When, on the other 
hand, I say Petefs nature, I conceive that nature as 
distinct from the whole which it serves to constitute 
and as existing in him, in that whole.* In short, the 
1 Because I conceive it as possessed of a certain mode or fashion of 
being which philosophers term subsistence or personality, and which 
terminates it, somewhat as a point terminates a line. 
In this introduction we make no claim to solve the problem 
of subsistence (the distinction between nature and person) which 
constitutes one of the most important problems of ontology. Adopting 
the pedagogical standpoint of coherent exposition, we have merely 
sought to present and classify notions, so that their sense and mutual 
relationship may be understood by a synopsis which is complete from 
the outset. 
3 I conceive it in abstraction from the modality called subsistence 
or personality which terminates it. Similarly I can consider a line 
abstracting from the point which terminates it, in which case the 
line thus considered is simply a part of the whole constituted by 
the line and point taken together, and exists in that whole. 
219
AN INTRODUCTION TO PHILOSOPHY 
subject of action possesses a nature or essence ; the 
concept of that nature or essence taken as such {what 
or that in virtue of which) is not the concept of the subject 
of action {that which) . 
We will now turn to this nature or essence of the 
subject of action. We have just observed that the 
subject of action exists (is capable of existence) in 
virtue of its own nature or its own essence,^ The 
nature or essence of the subject of action is there-fore 
that in virtue of which it is capable of existence 
pure and simple {simpliciter) ; the nature of Peter 
considered as the subject of action is that in virtue of 
which I can say simply Peter exists.^ 
Existence pure and simple is undoubtedly Peter's 
primary or tirst existence. But it is not his sole mode 
of existence. He is sad to-day, yesterday he was 
cheerful ; to-day he exists as sad, yesterday he existed 
as cheerful. He has lost the former existence and 
acquired the latter, but he has not therefore ceased 
to exist purely and simply. That is to say, he possesses 
a host of secondary qualifications in virtue of which 
he exists not only simply {simpliciter) but also under a 
particular aspect {secundum quid). It is thus that he is 
a musician or a philosopher, ill or in good health, happy or 
unhappy, etc. All these qualifications have accrued 
{accidere) to that which he is primarily as existing, are 
increments, or accretions, accidents. 
Philosophy, health, happiness, sorrow ; all these are so 
' In virtue of Its essence in the strict sense of the term in the case of 
a purely spiritual subject, in virtue of its nature in the sense of 
individual nature in the case of a corporeal subject. (See p. 235, 
note I.) 
* That is to say, Vk'ithout regard to any particular point of view, 
w^ithout modifying my thought by any addition. 
220
ontology: substance and accident 
many essences ^ to which our attention has not hitherto 
been directed, and which are not sell-subsistent in 
being, but on the contrary subsisting, so to speak, only 
as coverings of the subject of action. Employing the 
analogy of sensible objects we may say metaphorically 
that the latter exists beneath the accidents [substat) and 
supports them. From this point of view it may be 
termed a substance.* For example, we say that Peter is 
a substance. Moreover, since his nature considered 
precisely as such {what he is, that in virtue of which he 
is what he is, that in virtue of which he is capable of 
existence pure and simple), Uke himself exists beneath 
the accidents, it also is entitled to the name of 
substance, and we can speak of Peter's substance. We 
have now distinguished the notion of substance as 
opposed to that of accident.^ 
1 The definition of essence given above (p. 201) is applicable 
to accidents, if the subject is considered in a particular aspect. 
Understood in a concrete sense, as attributed to the object (for example 
sad when we say Peter is sad), the accident is what a thing primarily is as 
intelligible in a particular respect {being sad is the ground on which Peter 
possesses certain characteristics which necessarily follow from sadness). 
Understood abstractly, and separately or in the pure state (for instance 
when we speak of sorrow) the accident is that in virtue of which a thing 
primarily is what it is as intelligible in a certain respect. 
We may further use the term essence no longer in respect of the subject 
Peter, but in respect of the accidents themselves, and say that sorrow 
is that in virtue of which a particular passion primarily is what it is as 
intelligible. 
2 The subject of action is also called vwdcrraffis (hypo-stasis), -irpwrov 
viroKflnevov, primum subjectum attributiorns. 
3 Observe that the term substance {substantia) corresponds to the Greek 
word ovala understood in a restricted sense. The term ovcla primarily 
denotes essence or nature, but since substances are the first object which 
the intellect attains, when it considers what exists, they are also for 
that reason the first object which confronts the intellect with the notion 
of essence ; in other words they are the first to merit the denomination 
of essence or nature. Hence the term oiitria, which taken in its most 
221
AN INTRODUCTION TO PHILOSOPHY 
'thai which is : rwhat a thing is (essence or Tin a particu-essence 
in the nature) and in virtue of 
J lar aspect : accident 
wide sense - which it receives exist-j 
ence y absolutely : ^ 
Ahat which primarily . , 
/., , . ^ v ysubstance 
exists (the subject of 
| 
aiction,suppontum, person) j 
act of being : existence 
SUBSTANCE 
The name substance is given, as we have just seen, 
both to the subject of action itself {that which primarily 
exists) and to its nature considered precisely as a 
nature or essence {what a thing is, that in virtue of which 
the subject of action is what it is and claims existence 
pure and simple).^ What then shall be our definition 
general sense denotes essence, and is afterwards divided into substance 
ajid accident, has most naturally served to denote in a special sense the 
first member of the pair, substance. 
1 The subject of action (suppositum or person) is nothing but the 
substantial nature completed by a particular modality {subsistence or 
personality) which terminates it, as a point terminates a line (without 
adding anything to it in its order of nature) and renders it absolutely 
incommunicable. The term substance (corresponding to the Greek 
ovffia, which primarily denotes essence—see the preceding note) denotes 
the substantial nature without defining whether or no it is terminated by 
subsistence. It is therefore applicable alike to the nature (apprehended 
by the mind without the subsistence which terminates it) and to the 
subject of action (the terminated nature). But when we distinguish and 
contrast the nature (not terminated) and the subject of action, the term 
substance remains attached to the nature (not terminated) and is then 
contrasted with the subject of action taken as such. Thus when we 
speak of Peter's substance, we mean precisely the nature in virtue of which 
the subject of action Peter possesses primary being, and which is part of 
him. And theologians use the term in this sense when they teach that 
in the Divine Trinity the Father and the Son (two distinct Persons) 
possess the same substance, are consubstantial, ojuoovaioi. 
On the other hand, the Greek term vwixTTaai.^ {hypo-stasis, etymo-logically 
the same formation as sub'Stantia) after a certain vacillation in 
222
— 
ontology: substance and accident 
of substance ? Being receiving existence of itself {per 
se) or in virtue of itself in the unqualified sense adopted 
above ? No. For so understood the definition applies 
only to the subject of action ; absolutely speaking, it 
alone 
Peter for example—exists as a whole, and not 
as part of a being or subject in which it exists. Its 
nature, on the contrary, is part of itself and exists in 
itself. Peter's nature exists in Peter, and is part of 
Peter. It is true that since Peter is himself constituted 
by it, and exists in virtue of it {per earn), this nature 
does not exist in something previously existing which 
receives it (as, for example, sorrow exists in Peter, who 
was previously in existence). We can therefore say 
that it exists (is capable of existence) per se, in the 
precise sense that in order to exist it has no need to 
be part of another being previously existing which 
receives it in itself, but that on the contrary it constitutes 
the whole (the subject of action) which exists by itself. 
In this sense and, if we are careful to quaUfy and 
explain our meaning, the description ens per se existens 
is appUcable not only to the subject of action, but also 
to its nature, and may therefore serve as the definition 
of substance.^ 
(The same is true of the expression ens in se existens.) 
its employment, came finally to denote the subject of action taken 
as such [person) and is used exclusively in this sense. It is thus 
contrasted with substance understood as a nature not terminated by 
subsistence. 
Were we tempted to deny the vital importance of these abstract 
concepts and distinctions, we might recollect that for the word b/j-oovaios, 
on which the true understanding of the Trinity depends, but which 
differs only by an iota from the unorthodox term dfioiovcrioi, Catholics 
were willing, when the Arian heresy flourished, to suffer every kind of 
persecution, and sometimes even death. 
* Cf. John of St. Thomas, Cursus philos., i, Log. ii, q. 15, a. i. 
223
AN INTRODUCTION TO PHILOSOPHY 
We shall therefore ^ define substance as a thing or 
nature that can exist by itself or in virtue of itself {per 
se)—and not in another thing [in alio), that is to say, in 
a subject previously existing.^ 
(Alternatively we may define substance as a thing 
or nature whose property is to exist in itself.) 
Conclusion XV.—Substance is a thing or nature 
whose property is to exist by itself, or in virtue of 
itself {per se) and not in another thing. 
It is evident that the idea of substance represents 
something which really exists. If no substance 
existed, no ix.ture capable of existing in itself, all 
natures would be such as could exist only in something 
else. But in that case, since nature A could exist only 
in nature B and nature B in turn only in nature C, there 
would be an infinite regression which could never 
1 Existence itself cannot be a constituent part of any created nature. 
It is for that reason that substance must be defined as a thing or nature 
capable ofexisting per se or apt to exist per se. The same observation was 
made above in reference to the suppositum (p. 218, note 2). 
We must make the sense of the suggested definition clear. If per se 
(or in se) is understood in the restricted sense explained in the text, our 
definition will mean : substance is a nature apt to exist per se {or in se) qua 
nature or essence in the suppositum which it constitutes when terminated by the 
subsistence. If, however, per se (or in se) is understood in the absolute 
sense in which it was taken above (p. 218), the oposed definition will 
mean : substance is a nature apt to exist per se {or in be) qua subject of action 
(suppositum or person). 
' The term substance signifies a thing capable of existing in itself, or of 
subsisting ; that is to say, of being self-contained as an existent thing 
(its function subsistere), so that, once it exists, it sustains in being the 
additional qualities or accidents with which it is invested (its function 
substare). But it is only as a suppositum that substance is immediately 
capable of performing these two functions. Considered as a nature 
or essence it merely seeks to perform them. 
224
ontology: substance and accident 
reach a nature in which all these natures existed ; 
they therefore could not exist. 
Those philosophers who, like Fichte (nineteenth 
century), denounced the " dead substance of the 
Latins " to oppose to it " Teutonic action or 
becoming " were fighting against the intellect itself, 
which is simply unable to dispense with the notion 
of substance and imposes it upon us as an absolutely 
primary and immediate datum. Moreover, that which 
they took for substance and declared " dead," 
" inert," etc., was a mere figment of their own 
imagination. For substance is not an " empty 
receptacle," " an inert and dead support." It is the 
absolutely primal being of a thing, the radical 
principle of its activity and all its actuality. As 
Aristotle said, substantia est primum ens.^ But to 
perceive this a philosopher must make use of his 
intelligence, rise above the animal life of the senses, 
and not be content to show his skill iu handling words 
devoid of conceptual content and freighted only with 
material images. 
The substance of an object, so long as that object 
exists, is as such immutable.^ 
Peter's substance is that in virtue of which Peter exists 
purely and simply, that is to say, as Peter. So long as 
Peter exists, liis substance as such cannot change. And 
when Peter's substance does change (when Peter's 
body becomes a lifeless corpse) Peter exists no longer, he 
is dead. 
1 Metaph., vii, i. 
'' No doubt when Peter grows the change affects Peter's substanei 
itself, which increases, but solely in respect of its quantity. It does not 
affect it as substance. 
225
AN INTRODUCTION TO PHILOSOPHY 
Moreover, in itself substance is invisible, imperceptible 
by the senses. For the senses do not apprehend being 
as such, but present to us directly only the changing 
and the moving. In a certain sense, to be sure, it is 
indeed the substance of Peter that my eyes see, as it 
was truly Jesus whom the disciples saw at Emmaus, but 
my eyes thus apprehend the substance only in fact and 
materially, not formally. 
In other words the object seen or touched is some-thing 
which while seen or touched is at the same time 
also a substance ; but it is not seen or touched as a 
substance. As a substance it is conceived, not seen 
or touched, and so far as it is seen or touched it is 
coloured or exerting resistance, not being and 
substance. In the language of philosophy substance 
is intelligible in itself (per se) and sensible only accidentally 
(per accidens). That therefore in things which 
possesses most importance for us escapes the direct 
grasp of our senses and imagination, and is a pure 
object of the intellect, since the intellect alone appre-hends 
being as such (sub ratione entis). 
Observe that, if from the standpoint of existence 
substance is in things the being which is the primary 
and immediate object of the intellect, on the other 
hand to discover not only that a particular object 
possesses a substance, but also in what that substance 
consists, or what is its nature, we are obhged to take 
our stand upon that which reveals this nature to our 
senses, namely the operations, phenomena, or accidents, 
of the substance. In this sense we know the substance 
by the accidents. 
226
ontology: substance and accident 
ACCroENT 
Consider now such things as the laughter, movement, 
sorrow, joy, colour, etc., which I perceive in Peter, and 
which make Peter exist in certain aspects. These 
things are capable of existence. But they obviously 
do not exist after the same fashion as substance. 
To exist they must belong to another being previously 
existing.^ They exist as something which belongs to a 
being or subject already in existence. In this sense 
we say that they exist in something other than them-selves.- 
Conclusion XVI.—An accident is a nature or 
essence whose property is to exist in something 
else. 
1 Previously—if not in the order of time, at least in the order of 
nature. 
2 The accident of which we are sp>eaking is the predicamental accident 
which is contrasted with the substance. The term accident, when it is 
contrasted with property and signifies a predicate not derived from the 
essence (the predicable accident) has another meaning. 
If we are thinking of the predicamental accident, or of the contrast 
between substance and accident (a contrast between real beings) , we 
may say that the attribute relates to an accident (the intellect in virtue 
of which a man is capable of laughter is an accident really distinct from 
the substance) . If, on the other hand, we are thinking of the predicable 
accident, that is to say, of the contrast between those unreal beings of 
logic (the predicables) , genus, species, specific difference, property, and accident, 
it denotes not an accident but a property, an attribute predicated of the 
subject, not as something which helps to constitute his specific essence, 
but as arising necessarily from it. 
Conversely, if we are considering the predicamental accident, we 
must say that the individuating characteristics (the possession of a 
particular temperament, or heredity) belong, radically at least, to the 
substantial, not to the accidental order. If, on the other hand, we are 
considering the predicable accident, we must say that these character-istics 
are accidents (attributes predicated of the subject, neither as 
helping to constitute the specific essence nor as derived from it). 
227
AN INTRODUCTION TO PHILOSOPHY 
It follows that though an accident partakes, indeed, 
of being, it does not exist as a being ; it is essentially 
of a being, ens entis, and capable of existence only as 
the complement or perfection of a being. Thus the 
word being is predicable of the accident only in a 
secondary and indirect sense, and whereas being in 
the primary sense of the term is from the standpoint 
of existence the subject of action, so that our intellect 
apprehends immediately and of itself the subject of 
action, the substance, that which exists in itself, 
we find it difficult to arrive at a clear understanding 
of the accident. To succeed we are obliged to 
elaborate our notion of being, to make it more pliable, 
plane it down, bind it to the real, in short, to apprehend 
the accident by analogy with the substance which is 
contrasted with it. 
The mere fact that the term accident is a substantive 
involves us in the danger of regarding the accident as 
a substance, a piece of substance or a reduced 
substance. The imagination intervenes, and we depict 
to ourselves accidents and phenomena as fragments of 
matter inlaid in the suppositum, like a casing of mosaic 
or marquetry. Those for whom the words accident 
or phenomenon evoke images of this kind miss the notion 
of accident altogether. They conceive in fact only 
pseudo-substances and are incapable of advancing a 
step further in philosophy. An original effort of the 
intellect elaborating the notion of being is here the 
sole remedy. 
It is obvious that things such as an act of thought 
or a movement of emotion cannot be confused with 
our substance, because they come and go, and change 
within us, whereas our substance never ceases or
ontology: substance and accident 
changes, being as such immutable so long as we exist. 
Nevertheless these things are realities which affect us 
intrinsically. They are, therefore, really distinct from 
the substance in which they exist or, in technical 
language, inhere. There are thus contingent accidents 
(such that the subject can exist without them) real 
and really distinct from the substance. 
But if change, by showing that there are in a subject 
things which come and go, helps us to arrive at the 
notion of accident, it is very far from being a necessary 
attribute possessed by every accident. There are 
things without which a subject cannot exist, and which 
nevertheless are accidents, additional beings which 
complete the substance ; our understanding itself, 
for example, and our will are evidently something real 
in us. But they cannot be confused with our substance. 
For we possess a distinct notion of them wholly 
extrinsic to that of substance,^ which would be 
impossible if they were not essentially different from 
the latter.* Therefore the understanding and the 
1 We distinctly conceive the understanding or intellect as a faculty 
of knowing whose object is being, the will as a faculty of conation 
whose object is the good, substance as a nature or essence whose specific 
property is self-subsistence. The three concepts fall wholly outside 
each other. 
2 We are able to reason in this way because we are dealing with 
things proportionate to our intellect, which apprehends them by 
a proper and distinct concept (things which are, we say, known by 
their essence). In such cases, if two concepts are wholly external to 
each other, it is because the things they piesent to the mind really 
differ one from the other. Otherwise our intellect would deceive us. 
It is in this way that we prove that quantity or extension is an accident 
really distinct from corporeal substance, and that in every created 
thing the essence is really distinct from the existence. (On the latter 
point see John of St. Thomas, Cursus Phil., Phil. Nat., q. 7, a. 4.) When, 
however, the distinction is due to the standpoint from which the mind 
229
AN INTRODUCTION TO PHILOSOPHY 
will are real things in us distinct from our substance, 
consequently accidents (belonging as we shall see 
later to the category of quality) . There are, therefore, 
necessary accidents (without which the subject cannot 
exist) real and really distinct from the substance. 
The different schools which profess conflicting 
doctrines on the problem of substance may be repre-sented 
roughly by the following scheme : 
Philosophy of Aristotle and St. Thomas 
There are as many substances as there are individuals. 
In virtue of its substance each of these possesses primary 
being, but there are in each real accidents really distinct 
from the substance. 
Substantialists Phenomenalists 
There are no real acci- There is no substance ; 
dents really distinct from the accidents which are 
the substance, which is the apparent to the senses or 
sole reality. the consciousness (pheno- 
Descartes, Leibniz, and mena) are the sole reality, 
especially Spinoza. The English sensationalists 
— 
German pantheists of the neo-Kantian school. Philo-nineteenth 
century. sophy of pure becoming. 
Descartes denied the existence of real accidents 
really distinct from the substance. He identified 
corporeal substance with extension and the substance 
of the soul with the act of thought. He thus set 
views the same object we have indeed two distinct concepts, but not 
wholly external one to the other. For example, I distinguish in Peter 
his being a man and his being an animal, though in reality they are one 
and the same being. But the concept man, far from being external to 
the concept animal, imphes it on the contrary. 
230
ontology: substance and accident 
philosophy on a path which could only lead to 
pantheism (for if there are no accidents distinct from 
substance, every substance is its action—a perfection 
which belongs to God alone—and the concept of, 
substance becomes identical with that of absolute 
Being or God, with whom everything is thus confused), 
or if pantheism is to be avoided, to the denial of 
substance, which such a philosophy will do its 
utmost to disprove and to banish from the human 
intellect. 
Spinoza erected on the Cartesian foundation a 
monism or absolute pantheism from which Leibniz 
attempted in vain to escape by substituting for the 
single substance of Spinoza an infinite multitude of 
individual substances (monads), thus in effect replacing 
Spinoza's God by a boundless host of deities. Though 
they rejected the notion of substance, for which they 
substituted that of becoming or evolution, and regarded 
the thing-in-itself not as an object which imposes itself 
on the mind but as a background of the mind which 
produced the object, the German metaphysicians in 
the succession of Kant (Fichte, Schelling, Hegel) may 
be ranked among the exponents of pantheistic sub-stantialism 
since they posited a single principle which 
constitutes by its development the stuff and reality of 
all things. 
In the phenomenalist camp the English sensationalists 
and associationalists maintained that states of consciousness 
(sensations, emotions, ideas, etc.) are the sole reality 
accessible to us, and attempted to reduce the whole 
of psychology to the mutual association of these states 
of consciousness. The philosophers of pure becoming 
(Bergson, in particular, who thus joins hands with 
231
AN INTRODUCTION TO PHILOSOPHY 
" 
Heraclitus over a gulf of twenty-five centuries) deny 
the existence of anything permanent in things, and 
maintain that change without any subject of change 
is the sole reality. (In psychology these philosophers 
are opposed to the formxr group inasmuch as they 
substitute " a continuous stream of consciousness 
(W. James) for a mosaic or " host " of states of con-sciousness. 
But they are agreed in rejecting the 
notion of substance.) 
Kant (eighteenth century) substituted for the dis-tinction 
between substance and accidents in things 
(substance and accidents being equally objects of 
knowledge, the former by means of the second) ^ an 
opposition of two separate worlds, the world of things 
as they are in themselves {things-in-thernselves, noumena) 
and the world of phenomena manufactured by our 
mind. He regarded the thing-in-itself as wholly 
unknowable, though he affirmed its existence. This 
thing-in-itself, sought now in the subject of knowledge, 
became all in all to the German pantheists of the 
nineteenth century. Renouvier and the French neo- 
Kantians, on the contrary, taught that the thing-in-itself, 
the substance, is not merely unknowable but 
absolutely non-existent, and the concept of it 
chimerical. 
The various phenomenaHst philosophers just men-tioned 
failed to perceive that what they really deny 
is the accident, not the substance. What they under-stood 
hy phenomena is a mere pseudo-substance expressed 
by a concept ashamed of itself and self-contradictory, 
a substance pulverised, melted down, emptied of real 
subsistence, something which is not accident, but 
1 See above, p. 226. 
232
ontology: substance and accident 
being of a beings the pure complement of being, and 
inconceivable except as correlated with substance. 
Since they have never really understood what is 
meant by substance, and posit, under the appellation 
of phenomena, pseudo-substances, they quite naturally 
refuse to admit another substance behind these pseudo-substances 
of their imagination. 
THE INDIVIDUALITY OF SUBSTANCE 
The being primarily apprehended by the intellect 
from the standpoint of existence (the substance) is 
something individual. The intellect indeed appre-hends 
it as individual, for it apprehends the being of 
things only by turning to the sensations and images 
which reveal things to us under the conditions of 
their existence and in their individuality. Moreover, 
that alone can really exist which is by nature a com-pletely 
circumscribed and self-contained unit, or an 
individual, 
[a) Our intellect, it is true, can have no direct 
knowledge of this substance in its individuaUty ; it 
simply knows, by turning to the images from which 
it derives its ideas, that this substance is individual, 
it does not know in what its individuahty consists ; 
Peter's substance is directly revealed to it only by a 
universal idea. Peter's substance thus perceived, in 
abstraction from his individuality, is simply Peter's 
nature apprehended in the characters which constitute 
his essence, strictly speaking. And since we say of man 
that he moves, laughs, possesses understanding and will, 
etc., as we say it (primarily and in the first place) 
of Peter or Paul—since in consequence the property of 
existing beneath the accidents, which strictly belongs 
233
AN INTRODUCTION TO PHILOSOPHY 
to the subject and his individual nature, is thence 
transferred to the nature of the subject stripped of its 
individuaHty by abstraction—we shall also give the 
name of substance, though in a secondary sense 
(substantia secunda), to Peter's nature, as apprehended 
in abstraction from his individuality, that is to say, 
the universal essence man or humanity. On the other 
hand, by substance in the primary sense {substantia 
prima), we shall understand the individual substance.^ 
It is now clear that when we consider the being 
primarily apprehended by the intellect in material 
things we stress either the individual being or the 
universal being according as we consider this being 
primarily apprehended in relation to existence or in 
relation to intelligibiUty. 
In relation to intelligibility the being primarily 
apprehended in things by the intellect is the essence, 
strictly speaking, which in itself is not individual and 
exists in the mind in a condition of universaHty ; and 
it is only in an improper sense that the term essence 
is used of the essence individuated by the individual 
matter (that is, of the individual nature). 
In relation to existence, on the contrary, the being 
primarily apprehended in things by the intellect is 
1 In the Aristotelian and Scholastic vocabulary the term substantia 
prima, ovaia irpLbrrj, denotes (see above note) the individual nature of the 
subject of action without determining whether or no it is terminated 
by subsistence. Usually indeed it does in fact denote the terminated 
nature, or subject of action, the hoc aliquid. It does not, however, 
denote formally the subject of action taken as such and contrasted 
with the (non-terminated) nature. That function belongs to the terms 
suppositwn and persona {vw6(rra.<n$). 
Remark that the distinction between the subject of action and 
the nature (non-terminated by the subsistence) is in the main due to 
the Schoolmen. Aristotle himself did not make it explicitly. 
234
ontology: substance and accident 
the individual substance/ and it is in a secondary 
sense that the term substance is used of the nature 
stripped of its individuality by abstraction (that is, 
of the essence in the strict sense) * 
. 
Here we may call to mind what was said above ' 
about the individual nature. We see at once how 
we should classify the different concepts with which 
we have made acquaintance hitherto.* 
what { primarily as intelli- i 
the gible {essence in the 
J 
sub- strict sense) and in 
| 
ject j virtue of which it  
that which is : is exists 
essence in the-; - primarily as complete- n 
wide sense ly circumscribed {in-dividual 
nature) and 
in virtue of which 
it exists as such 
^ <Aa<u)AicA primarily is: the pri-marysubject 
ofaction {sup-positum, 
person) J 
existence 
in a particular respect : accident 
absolutely : substantia ^ 
secunda 
' substarue 
substantia prima 
act of being 
: 
1 Taken precisely as that in virtue of which the subject possesses its 
first being, the substance, substantia prima, is the subject's individual 
nature. We said above (p. 214) that that in virtue of which a thing is 
susceptible of existence is the universal essence, the reason being that 
we were then considering precisely what is the ground on which a thing 
is susceptible of existence, in contrast with that which is merely a 
condition or state in which it must be in order to exist. Here, however, 
we are dealing with that in which the existence of the thing considered 
precisely as in the state necessary in order to exist is grounded, and 
this is not the universal essence, but the individual nature of the subject. 
2 Hence in the order of pure spirits and there alone (where no distinc-tion 
exists between the individual nature and the essence, see above 
p. 2 1 1 , n. 3 ; p. 2 1 3, n. 2), the substance in the primary sense of the word 
is also the essence, strictly speaking. In the material order, on the other 
hand, substance in the primary sense of the word is the subject's 
individual nature, and it is onJy in a secondary sense that the essence, 
strictly speaking, is called substarue, ' See page 209. 
* The quod and the quo. We have already observed (see above, 
p. 200, note 1) that the essence taken in the concrete or as attributed 
235
AN INTRODUCTION TO PHILOSOPHY 
(b) Per se, a se, in se. We have defined substance 
as a thing whose property it is to exist by itself (per se) or 
in itself (in se). We must determine carefully the 
exact sense of these expressions. 
A thing is said to exist in itself {in se) when it does 
not exist as part of a whole previously existing, but 
to the thing {what a thing primarily is as intelligible) is not presented 
to the mind in its purity ; it is in fact presented to the mind together 
with the thing or subject which it determines. To possess it in its 
purity, it must be conceived separately, without the thing or subject 
it determines, as for instance when we speak of humanity or, to force 
language, of the being man, the entity man. In that case it must be defined 
as that in virtue of which a thing primarily is what it is as intelligible, or, in 
other words, that in virtue of which a thing is constituted in a particular 
degfee of being primarily intelligible. For this reason it will he 
better to substitute in our synopses for the term what the phrase that in 
virtue of which. So we finally get the following table : 
INTELLIGIBILITY 
BEING^ 
. that in 
virtueof 
which 
{id quo) 
Peter exists and is / in a particular respect : 
what he primar-J whiteness accident 
ily is as intelli-j 
gible {esserue in |^ absolutely : 
the strict sense) humanity . . substantia^ 
Peter exists and is^ secunda 
that which is : what he primar 
essence in thej ily is as complete-wide 
sense ly circumscribed 
(essence indivi-duated 
by mat 
ter, the individ-ual 
nature) 
that which primarily exists : 
suppositum, person {id quod) 
act of being : that in virtue of which 
{id quo) a thing is outside nothingness 
or its causes : existence 
Peter's indivi-^ 
dual nature 
substance 
ysubsiantia 
prima 
Peter EXISTENCE 
The distinction between the quod and the ^mo plays a part of the first 
importance in the metaphysical analysis of things. 
236
ontology: substance and accident 
itself constitutes the whole which exists. In this sense 
Peter exists in himself. 
A thing is said to exist by itselfor in virtue ofitself (^^r 
se), when it is brought into existence in virtue of itself, 
or of its own nature (by the causes on which it depends, 
if it is a created nature) . In this sense Peter exists j&^r se.^ 
Philosophy makes frequent use of this expression, 
per se. It always means in virtue of itself, in virtue of its 
own essence [per suam essentiam)—whether the quality 
under consideration forms part of the essence of the 
thing or necessarily results from it as its principle (in 
which case per se is opposed to per accidens) ^ or whether 
we merely wish to state that the attribute under con-sideration 
immediately pertains to the thing which does 
not receive it through the intermediary of anything 
other than its own essence (in which case per se is opposed 
to per aliud). It is in this sense that the subject of 
action exists per se, whereas the accident exists per aliud. 
But the expression per se does not mean in virtue of 
itself or of its own nature as the absolutely first principle or 
as the complete and ultimate explanation. This is something 
totally different, which is expressed by the phrase 
a se, of orfrom itself (as opposed to ab alio). That which 
is a se is evidently per se, but that which is per se is not 
^ Existence per se or in se can, we have already seen (see pp. 2 1 8, 223) 
, 
be ascribed, as it is understood in a more or less strict sense, either to 
substance in general (that which exists per se and in se contains in itself 
whatever is necessary in order to receive existence and is not part of 
a previously existent whole) or exclusively to the subject of action 
{suppositum or person, which contains in itself everything necessary in 
order to receive existence, and exists in no respect as part of a whole). 
* For example, Peter is per se alive, endowed with intellect, and the 
faculty of laughter, the artist is per se one who fashions objects. But 
Peter is per accidens a sufferer from influenza or the heir to a large 
fortune, the artist per accideru celibate or married, etc.
AN INTRODUCTION TO PHILOSOPHY 
by any means for that reason a se. That which exists 
a se or from itself, possessing in itself the entire explana-tion 
of its existence, is uncaused ; God alone is from 
himself, a se. Created substances on the contrary 
[created subjects of action) are caused ; they exist per se, 
in virtue of their essence ; they do not exist a se. 
In their own nature they possess everything necessary 
to receive existence, but not to possess an existence 
not received from without. They are sufficient by 
themselves to exist, in the strictly qualified sense that 
they do not exist as something which belongs to some-thing 
else, but, absolutely speaking, they are by no 
means a sufficient ground of their own existence. 
That which is a se cannot cease to exist ; that which is 
per se without being a se can lose its existence. 
The distinction between that which exists a se and 
that which exists per se is perfectly clear. Nevertheless 
certain philosophers have lost sight of it, notably 
Spinoza, who ascribes aseity [a-se-itas) to every 
substance. (From which it follows immediately that 
there is only one substance, and that everything 
is God—monism and pantheism). When indeed 
Spinoza defines substance as that which is in itself and 
is conceived by itself he really means, as the context 
shows,^ that which to be and to be conceived needs absolutely 
nothing except itself Descartes had already defined 
substance ambiguously as res quae ita exsistit ut nulla 
alia re indigeat ad exsistendum, a thing which exists in such 
a fashion that it has need of nothing else in order to exist. ' 
1 Cf. Ethics, i, 7. " Principia Philosophiae, i, 51. 
238
VII 
ONTOLOGY : ACT AND POTENTIALITY 
When we studied being first from the standpoint 
of intelligibility, then fi-om that of existence, we saw 
that the object primarily apprehended by the 
intellect, being in the primary sense of the term, was 
in the former case what we call essence, in the latter 
what we call substance. 
We must now consider the being of things (under-standing 
the term being in its most general and 
indeterminate sense) from the standpoint of action, in 
reference to the manner in which things behave in 
reality, or, if you prefer, in reference to what they do. 
This new standpoint acquaints us with a third primary 
sense of the term being. 
(a) What is the first truth which the intellect grasps 
as soon as it has formed the notion of being ? It is 
sufficient to consider the notion to see at once that 
what is, is (principle of identity), or again that what is, 
cannot not be at the same time and in the same relation 
(principle of non-contradiction). That is to say, that 
everything is what it is, that it is not what it is not, 
and that it is everything that it is. 
We will now consider what things do, what is their 
natural behaviour, what is the primary fact of experi-ence 
grasped by the senses and consciousness. Things 
change. The arrow flies, the animal runs, what was 
239
. 
AN INTRODUCTION TO PHILOSOPHY 
cold becomes hot under the action of fire, food becomes 
flesh, what was Uving dies, and every spring that which 
had no being comes into existence. 
Although, like all our primary notions, it is very 
difficult to explain scientifically, everybody knows by 
experience in what this great fact of change or motion 
consists. ¥/e may say that wherever there is change 
there is a transition (from one being to another, or 
from one mode of being to another). And for 
transition to exist, there must be something which 
undergoes it, something which is the subject of change, 
for example a subject which ceases to be in a particular 
place or a particular thing {terminus a quo, the arrow 
pressed to the bow, thefood, the seed), to be in another place 
or another thing {terminus ad quern, the arrow in the 
target, flesh, the mature plant) 
There is no change without a subject whi:h is 
changed, and which must be some particular thing 
before the change is effected ^—in other words being 
is prior to change. 
Those, indeed, who maintain that change is prior 
to being, and that there is change without a subject 
which is changed and which is some particular thing 
before the change is effected, deny the principle of 
identity and fall into absurdity. For when they take 
up this position, they must either continue to accept 
the notion of being, in which case to affirm that there 
is change without a subject of change, or that change 
is prior to being, is to affirm that what has no being 
1 For example, in a substantial change the first matter, which is not 
a being, but potential being, is the subject which is changed and consti-tutes 
a particular body, or particular being, by its union with a particular 
substantial form, before it constitutes some other body by its union with 
another substantial form. 
240
ontology: act and potentiality 
changes, which is manifestly absurd ; or they must 
reject the notion of being as illusory and argue that 
instead of conceiving being we must conceive change, 
in which case they must reject as false, together with 
the notion of being, the principle of identity which is 
bound up with it, and maintain that thought is 
essentially deceptive, which is equally absurd. 
We are, therefore, absolutely obliged to hold that 
being is prior to change, and that there is no change 
without a subject which is changed, and which is some 
particular thing before it changes ; or, in the language 
of philosophy, that there is no motion without a subject 
which is moved. 
(b) We will now turn away from experience and 
every sensible representation and attempt to consider 
change with our intellect, that is to say, in terms of 
being, the formal object of the latter. We shall 
inquire how or in what respect the starting-point of 
change can thus become the goal. You will answer, 
perhaps, that it is according as it is this or that, in 
respect of what it is, that the starting-point becomes 
the goal. But the starting-point is nothing but what 
it is, and is already everything that it is, and therefore 
in this respect is incapable of becoming, for it already 
is. You may then say that it is according as it is not 
this or that, in respect of what it is not, that the starting-point 
becomes the goal. But in respect of what it is 
not, the thing is nothing whatever, is pure nothing, 
and therefore cannot be the source of the product of 
change. It is incapable of becoming, for it simply 
is not. 
Hence the starting-point of change cannot become 
its goal—either in respect of what it is or in respect 
241
AN INTRODUCTION TO PHILOSOPHY 
of what it is not. In other words, the new being which 
is the product of change can be derived neither from 
the being which akeady exists, nor from a nothing, 
which has no existence whatever. Is change, therefore, 
impossible, as Parmenides maintained ? And are we 
obliged with him to deny the evidence of our senses, 
which witnesses to the fact of change ? 
{c) No. But we are obliged to develop and explore 
our idea of being. Evidently in the analysis we have 
just made something has been left out. The starting-point 
of change is no doubt already everything which 
it is, but it is not yet all which it can be ; it is not yet 
that particular thing it is destined to become, but it 
possesses the means to be it, it can be it. Therefore 
between being and not being there is the power of being. 
It is neither in respect of what it is, nor in respect of 
what it is not, but in respect of what it can be that 
the starting-point of change becomes its goal. 
The arrow is here {on the bow, for instance) and from 
the standpoint of being pure and simple, it is nowhere 
else ; but it can be there {at the goal, for example), 
and possesses the means to be there. Bread is bread 
and nothing but bread, and not flesh, so far as it is, 
in the sense of being pure and simple, that is to say, 
of being completely realised ; but it can cease to be 
bread and become flesh. There is in it that which 
enables it to undergo the change under the action of 
certain determinate causes. 
POTENCY OR POTENTIALITY 
Things therefore are not confined and held fast by 
what they are and what they are not. Even while 
242
ontology: act and potentiality 
they are here and not there, this and not that, they 
possess the power to be there and no longer here, 
that and no longer this. But so long as they are here 
or are this, that power which they possess remains 
mere power and is not manifested. 
This power in them is as such something real. 
Consider a man asleep. He neither sees nor speaks 
nor walks. But he is not therefore bUnd, paralysed, 
or dumb. He is really capable of seeing, speaking, 
and walking. While he does not speak he retains 
the power to speak, he has it in him ; whereas he 
cannot without a violation of nature become a tree 
or a bird. Or again take a billiard ball at rest. It 
is immobile (not moving). But it is not therefore 
immovable. It is really capable of motion. While it 
does not move, it retains the power of being moved, 
it has it in itself; whereas it has no natural power 
of passing through a wall. The power of being is not 
being in the full and primary sense of the term ; but 
power of being without as yet being is not sheer non-entity. 
Power of being taken precisely as such is irre-ducible 
either to nonentity or to being pure and simple. 
It is something different from either, something sui 
generis for which philosophy must find a place. Pre-cisely 
so far as things can be something they are not, 
they, after an inferior fashion, are. 
We have thus found something which does not 
deserve to be called being, on which that title can 
be bestowed only in a secondary and improper sense, 
as an alms, so to speak, but which nevertheless is real. 
It is what philosophers term potency or potentiality. 
In using the term potency we must be on our guard 
against ambiguity. This potency is not that of which 
243
AN INTRODUCTION TO PHILOSOPHY 
we think when we say that a being is potent. This 
potency is not an active power ; power to effect 
something or to work, at least when understood as 
active, is the absolute contrary of the power or potency 
with which we are now concerned, being not potency 
but act. The potency of which we are speaking is 
entirely passive, nothing more than a real power of 
being or becoming. Wax is in potency to receive the 
impress of the seal, water in potency to become ice or 
vapour. The active powers (for instance, the faculties 
of the soul) are also rightly termed potencies or potenti-alities^ 
but only in so far as they are not, or are capable 
of not being, actually operative, or so far as they are 
simple capacities of action or operation. 
ACT 
Since, however, power of being, though not nothing, 
is not being in the full and primary sense of the term, 
we must find a name for being in the full and primary 
sense of the term as distinguished from potentiality. 
Philosophers call it act. 
Here also we must beware of ambiguity. We are 
not concerned, at least not primarily and chiefly, with 
an act in the ordinary sense of the word, with doing 
or action. Action or operation is indeed an act, being 
in act, but it is what is termed the secondary act [actus 
operationis) . Action presupposes being. And the 
primary act is the act of being [actiis exsistentiae) , more-over 
of being a particular thing {actus essentiae) . For 
example, a body is luminous in act, even when it is 
not illuminating anything else. Clay, once modelled, 
is a statue in act, water at 32° Fahrenheit is ice in act, 
244
ontology: act and potentiality 
and the moment anything effectively is one thing or 
another and especially the moment anything exists, 
it is in act. 
Act may therefore be defined as being in the strict 
sense of the term, taken in the fullness thus signified, 
or again the finished, the determinate, or the perfect as 
such. Potentiality, on the other hand, is the determin-able, 
the perfectible, that which is capable of being 
finished, as such ; not a being but a real power of 
being. 
We must take care not to attempt to think with 
our imagination these concepts of act and potentiaUty. 
They can be thought by the intellect alone. Least of 
all must we conceive of potentiaUty as some sort of 
being in act which we imagine as more or less hazy, 
indefinite, inactive, and hidden in the object. Poten-tiaUty 
in itself is absolutely incapable of being repre-sented. 
It is not a spring or an organ hidden in the 
thing, nor a character prefigured in it after the fashion 
of an imaginary statue outUned beforehand by the 
veins of marble within the block, nor yet an act thwarted 
or rendered abortive, like an effort or pressure over-come 
by the resistance of an obstacle. It is absolutely 
nothing done or in process of doing, absolutely nothing 
in act. In itself it cannot be conceived (for in that 
case it would necessarily be conceived as something 
determinate). It can be conceived only by means of 
the act (the particular thing) with which it is corre-lated, 
as the simple power of being that particular 
thing. 
Conclusion XVII.—Being, considered in relation 
to the fullness and perfection which the term 
245
AN INTRODUCTION TO PHILOSOPHY 
signifies, is divided into being in the strict sense or 
act, and power of being or potentiality. 
We are now in a position to understand change. 
The product of change arises neither out of being in 
act nor out of nothing, but from potential being. In 
other words, the action of the efficient cause draws, 
educes, from the potentiality of the subject the deter-mination, 
the form, which was wanting in the starting-point 
of the change and characterises its goal, as when 
the action of fire educes from the potentiality of 
water (the water is cold, but can be hot) the determina-tion 
(a specific intensity of heat) which characterises 
it as the result of the change. The change is the 
transition from potentiality to act, or, more accur-ately, 
according to a definition to which we must 
return later, it is the act of a thing in potentiality taken 
precisely in respect of its potentiality : actus exsistentis in 
potentia prout in potentia. 
ACT AND POTENTIALITY IN THINGS 
From what has been said it follows that all change-able 
things, in whatever respect they are subject to 
change, are compounded of potentiality and act. 
God alone, since he is absolutely unchangeable, is 
devoid of any potentiality. Since he is subsistent 
Being itself or the Fullness of being, he is incapable 
of becoming ; there is no perfection which he does 
not possess or rather is not already ; he is pure act. 
The being of all other things, on the contrary, is 
too poor and too weak to reaUse simultaneously 
everything they are capable of being. For every one 
of them there is really open a vast range of possibilities, 
246
ontology: act and potentiality 
of which they can never rezilise more than a few, and 
that by changing. 
Here we may observe that the obscure and mysterious 
concept of first matter whose acquaintance we made 
when we studied the notion of individual nature is 
that of a pure potentiality in the order of substance, 
which can be any and every body and by itself is none. 
It is the purely potential principle which in union 
with an actual principle (a substantial form) constitutes 
a particular corporeal substance, and is the subject 
of substantial changes. 
Potentiality and act divide between them the 
totality of created being, both in the order of substance 
and in the order of accidents. In other words, they 
are, like being itself, transcendental objects of thought 
which exceed or transcend every limitation of class 
or category, and include all created things. The 
substance of bodies is compounded of potentiality 
{first matter) and act [substantial form) . The substance 
of incorporeal things [pure spirits) is not composite ; 
it is in respect of that which constitutes its nature or 
essence wholly act. But it is not therefore pure act 
(in the case of pure created spirits), for this substance 
{substantial essence) is itself potentiality in relation to 
that which is the ultimate act of everything real 
{actualitas omnis formae) or existence : pure spirits do 
not derive their existence from themselves, a se ; they 
can not be. 
On the other hand, we may remark that every 
accident {whiteness, strength, virtue, etc.) is an act 
{accidental form) which determines the subject and is 
itself sometimes in potentiality in respect of further 
determinations. The intellect, for example, is an 
247
AN INTRODUCTION TO PHILOSOPHY 
accident (an accidental form) whose subject is the soul, 
and it is in potentiality in respect of a particular act 
of thought. 
It is clear that all the notions with which we have 
become acquainted hitherto can be classified in the 
following fashion : 
' uncreated being : God, ens a se 
primarily as in- 
.pure ac 
creat 
ed 
being 
that 
which 
is 
essence < 
in the 
wide 
sense 
what telligible {ess 
the ence in the strict 
sub- ^ sense) and in vir- 
I 
tnai ject t^e of which it 
exists 
V primarily as com-pletely 
circum-scribed 
{individ-ual 
nature) and 
in virtue of 
which it exists 
as such 
that which primarily is : 
the primary subject 
of action {supposit-um, 
person) 
j 
 act of being : existence 
in a particu-lar 
respect 
: 
absolutely : 
substantia 
secunda 
substantia 
prima 
accident {ac-  
cidentalform) 
substance 
composed (in 
the corporeal 
order) of po-ten 
tiali ty 
and act, and 
itself actual-ised 
by the 
ultimate 
act of every-thing 
real 
beii^ 
com' 
pound 
oi pote 
tialit 
and c 
Aft-er what has been said, it is sufficient to consider 
the notions of potentiality and act to see immediately 
the truth of the following Axioms : 
(i) Potentiality cannot exist in the pure state, apart 
from any act. This is evident. For, since existence 
is an act, potentiahty can only exist in beings which 
are in some other respect in act.^ 
1 Hence first matter cannot exist separately unactualised by some 
particular substantial form. Similarly essence is in relation to the 
act of existence a potentiality really distinct from existence but actual 
in virtue of existence. 
248
ontology: act and potentiality 
(ii) Nothing is educed from potentiality to act except 
by some being in act. It is plainly impossible that 
that which is in potentiality, that which is capable 
of having a determination or a perfection but does 
not have it, should give to itself what it lacks, so 
far as it does not possess it, that is to say, so far as 
it is in potentiality. 
(iii) Act is prior to potentiality. A consequence of 
the preceding axiom. 
^ 
(iv) Potentiality is essentially relative to act and is 
for the sake of the act [potentia dicitur ad actum). 
It is indeed only in relation to the act that the 
potentiality can be conceived (only in relation to 
being white that we can conceive the power of being 
white) ; and it is also only for the determination or 
perfection that the determinable and the perfectible 
as such are. 
(v) Act and potentiality belong to the same order ; that is 
to say, both must be in the order of substance, or 
both in the order of accident. For it is evident that 
every act which at the same time completes and 
specifically determines a potentiaHty must belong 
to the same order as that potentiality. The activity 
of thought, for example, belongs to the order of 
accident like the faculty itself from which it proceeds 
and which is in potentiality to that activity. 
(vi) Everything acts according to its nature in act. 
1 This axiom is the metaphysical explanation of the truth previously 
affirmed : being is prior to becoming or change. Absolutely speaking 
this is true. In the order of material causality on the other hand 
potentiality is prior to act, becoming prior to being, the seed prior to 
the tree. But the seed itself presupposes the tree which produced it 
and at the beginning of the entire process the actuality of the First 
Cause.
AN INTRODUCTION TO PHILOSOPHY 
Since activity is an act {actus operationis) which is 
brought into being by the subject from which it 
proceeds it presupposes (as laid down by Axiom ii) 
that the latter is in act to the extent to which it 
produces that activity. The same truth differently 
enunciated is expressed by the dictum action or 
operation manifests being {operatic sequitur esse). 
(vii) The combination of two beings in act cannot 
produce something which is one of itself We call one 
of itself {unum per se), as opposed to one by accident^ 
a thing which constitutes a single being and not a 
conjunction of beings, in other words a thing which 
is one in virtue of the nature by which it exists. 
For example, a living organism is a unit of itself, 
whereas a machine or a house is an accidental unit.^ 
This distinction once understood, it is plain that 
two beings in act, and as such constituting two 
beings, can never by their combination constitute 
anything except a conjunction of beings, that is to 
say, an accidental unit.* 
Once more on this question of act and potentiality 
we find philosophers divided into three great schools. 
The school of Aristotle and St. Thomas teaches the 
distinction between potentiality and act, the priority 
1 No doubt, if you destroy the unity of the machine or the house, 
you destroy the machine or the house, but you do not destroy the 
natures (or substances) of which it is composed {iron, steel, bricks, etc.). 
Destroy, on the other hand, the unity of an organism, you destroy its 
very nature {substance). 
2 This axiom plays an important part in natural philosophy and 
particularly in psychology. For instance, the Cartesian conception 
which regards the soul and the body as independent of each other, 
two complete substances, is unable to explain the substantial unity of 
the human being, because two complete substances are two beings 
in act. 
250
. 
ontology: act and potentiality 
of act to potentiality, the reality of motion and 
becoming, but the priority of being to motion. It 
also shows that between God (the pure act) and all 
things besides (compounds of potentiality and act) 
there is an absolute and infinite difference. 
Exaggerated intellectualism (Parmenides, Spinoza, 
Hegel) refuses to admit the notion of potentiality, 
because by itself it is obscure. If, however, everything 
which is, is wholly act or pure act, either motion must 
be unreal (Parmenides) or contraries identical (Hegel) 
and creatures must possess the same nature as God 
(pantheism) 
Anti-intellectualism (Herachtus, Bergson) equally 
rejects the distinction between potentiality and act, but 
because the notion of being is in the opinion of these 
philosophers illusory. If, however, being is denied in 
favour of becoming or pure change, pure act can no longer 
exist ; and however the exponents of this school may 
struggle to escape the logic of their position, God must 
possess the same nature as things (pantheism). 
Moderate Intellectualism 
[School of Aristotle and St. Thomas) 
Potentiality and act in things. God or the pure act 
is absolutely distinct from created things. 
Exaggerated Intellectualism Anti-intellectualism 
No potentiality in things. Neither act nor being. 
Everything is absorbed Everything is absorbed in 
either in pure being, or in change or pure becoming, 
the contradiction which and God is identical or 
constitutes becoming, and continuous with things, 
created things are identified 
with God. 
Material and formal, virtual and formal (or actual), 
implicit and explicit, in accomplished act, in express act. 
251
AN INTRODUCTION TO PHILOSOPHY 
In connection with the notions of potentiality and 
act philosophers use certain expressions whose meaning 
must now be explained. 
We have just learned the meaning of the two 
correlatives in potentiality and in act. The marble 
before it has been carved is a statue in potentiality ; 
as soon as the sculptor has given it the form he intends, 
it is a statue in act. 
{a) Closely related to these expressions in potentiality 
and in act are the expressions so frequently met in 
philosophy material and formal. They have been 
borrowed from natural philosophy (cosmology) 
which proves that every corporeal substance is 
compounded of two principles, Jirst matter (pure 
potentiality in the order of substance) and substantial 
form or first act.^ 
The terms material and formal have passed from 
natural philosophy into all branches of philosophy, to 
designate, by analogy, on the one hand whatever, in 
itself indeterminate and potential, plays the part of a 
subject which receives a determination, on the other 
hand whatever possesses of itself a determining, actuali-sing, 
and specificatory function, or again whatever 
is taken as possessing a particular character, in a 
particular aspect. It is in this sense that, as we have 
already seen,^ we distinguish between the material 
and formal objects of a virtue, science, or faculty. 
Hence arises in particular the distinction between 
the material and formal statement. We speak 
materially when we do not take the things of which we 
speak precisely as possessing the characters denoted by 
the words we use ; we speak formally when in the 
1 See above, pp. 166-168. ^ See above, p. 106. 
252
ontology: substance and accident 
. 
things of which we speak we consider not so much the 
subject which possesses these characters as the 
characters themselves, with the sharp contour and 
clear-cut line they describe in it. This distinction is 
extremely important. Formal statement should, in-deed, 
be the constant aim of philosophy ; and, on the 
other hand, many propositions are true formaliter 
loquendo which are false materialiter loquendo, and vice 
versa. For examples the following propositions are 
true understood formally, but false if understood 
materially : 
Everything which is, is good (so far as it is). 
The common good is always preferable to the good of 
the individual (provided the common good is under-stood 
formally, in which case the union of the soul 
with God, that is with the transcendent common good 
of all creatures, is to be preferred to everything else) 
Superiors ought always to be obeyed. (So far as they 
are superiors, and command nothing opposed to the 
orders of a superior of higher rank.) 
There are men who are natural slaves. (If slave is 
understood formally, as meaning destined for manual 
or servile work.) 
Every virtue is stable. (If we consider solely its 
quality of virtue.) 
Knowledge is infallible (so far as it is knowledge). 
Conversely, the following propositions are true only 
in a certain context, and if understood materially.^ 
* It would be of interest to show how philosophy, since it has given up 
the technical language of Scholasticism, has increasingly tended to use 
terms in a material rather than a formal sense. Hence a number of 
badly stated problems, and a host of misunderstandings both between
AN INTRODUCTION TO PHILOSOPHY 
This picture is the Adoration of the Magi. 
This book is the philosophy of Pythagoras. 
Speech has been given to man to conceal his thoughts. 
Philosophy is proud. 
The British Constitution is good because it is illogical. 
{b) We must be careful to distinguish in potentiality 
from virtual, for they signify entirely different things.* 
A thing is said to be virtual, or to exist virtually, when it 
is contained in another thing of superior rank, not in 
its being or proper determination (its formality) but 
under another being or determination (another 
formahty), so that it is truly there according to the 
virtue or degree of perfection which belongs to it, but 
not formally or actually. In this case, the being in 
which it is found is not in potentiality in respect of it, 
but on the contrary in act after a higher fashion. 
Its superiority is, so to speak, an obstacle which 
prevents the thing which it virtually contains from 
being present with its proper and inferior determina-tion 
(formality). 
Thus the perfections of all corporeal objects exist 
modern philosophers themselves, and still more between the modern 
philosophers and the ancients, with their formal terminology. 
It may also be observed that certain philosophic terms understood in 
a material sense, have acquired a meaning totally different from their 
original significance. Take, for example, the term object. For the 
ancients the object meant what is placed be/ore the mind or presented to it, 
considered formally as such. Hence imaginary beings, the chimaera 
for example, were said to exist objectively or as objects present to the mind, 
but not really or as things existing outside the mind. The modems, on 
the contrary, understand by an object the thing itself or the subject 
which is presented to the mind, and to exist objectively is therefore the 
same as to exist really or outside the mind. 
1 Observe, however, that the expression potentially, in potentia, is 
sometimes used improperly in the sense of virtually.
ontology: act and potentiality 
. 
virtually in God, conclusions are virtually contained 
in the premisses, partial lives exist virtually in the life 
of an organism. 
{c) We must now call attention to the fact that the 
contrast between implicit and explicit is not to be 
confused with that between virtual and formal {actual) 
A thing contained implicitly in another may be there 
formally or actually, not virtually : but it is present in 
a confused fashion, wrapped up and hidden as a flower 
lies hidden and folded in the bud. For example, in 
the truth Peter is a man there is implicitly contained 
the further truth Peter is a rational animal. 
{d) Finally a thing given formally and explicitly 
may be in act in two different ways. Take, for instance, 
a man who is running as fast as he can to escape his 
enemies. If I say he is fleeing I mention what he is 
doing in express act {in actu signato) (what he is doing as 
stressed or expressed, so to speak, by his intention). 
If, however, I say that he is accelerating the rate of his 
breathing I mention what he is doing only in accomplished 
act {in actu exercito). 
Similarly, a man who reads Ronsard, Lamartine, or 
Victor Hugo, to count the number of times they use 
the words aimer or cherir^ reads the poets, to be sure, 
and reads them formaDy and expUcitly, but that is 
not what he does in intention. On the contrary, 
when we consider his object in reading the poets, we 
must say that he is preparing an essay of stylometric 
literary criticism. We may say that he reads the poets 
in effect or in accomplished act, but that expressly 
or in express act he is preparing the essay in question. 
Again, when we repeat lilia agri non laborant neque 
nent, thinking solely of the meaning of the sentence, 
855
AN INTRODUCTION TO PHILOSOPHY 
what we then know in express act is the lilies of the 
field, which are present to our mind as neither toiling 
nor spinning. But at the same time we know in 
accomplished act the nominative plural, lilia^ which in 
turn we shall know in express act, if we return to 
the sentence in question and submit it to grammatical 
analysis. 
That is to say, the phrase in express act {in actu signato) 
is used of things to which intellect or will are directed, 
when they are the object of a concept of the intellect 
or an intention of the will specially formed for them, 
and are thus presented to the mind or brought into 
being under the actual heading or on the actual ground 
expressed by their name. When on the contrary 
they are presented to the mind or brought into being 
on occasion of something else and without being 
intended in themselves, we say that they are present 
only in accomplished act {in actu exercito). 
256
VIII 
THEODICY (natural THEOLOGy) 
. 
Metaphysics studies being as such, but for that 
very reason is obHged to study the cause of being. 
That is why its highest branch, the coping-stone so 
to speak of the entire metaphysical edifice, is con-cerned 
wdth him who is subsistent Being itself. This 
branch of metaphysics is called natural theology (the 
science of God in so far as he is accessible to natural 
reason, or, from another point of view, so far as he 
is the cause of things and author of the natural order) 
Since Leibniz it has also been known by the less 
appropriate name theodicy. 
Leibniz in his Theodicy (1710) undertook to defend 
Divine Providence against the attacks of the sceptics 
(particularly Bayle). The term theodicy (etymologi-cally 
: justification of God) has been used since that 
time to denote the branch of philosophy whose object 
is God. But the name is objectionable on two grounds : 
first because the providence of God has no need to 
be justified by philosophers ; secondly because provi-dence 
and the problem of evil are neither the sole nor 
the most important questions of which natural theology 
has to treat. 
The primary questions discussed by natural theo-logy 
are obviously those which concern God's existence 
itself. 
For the existence of God is not in fact, as Male- 
257
AN INTRODUCTION TO PHILOSOPHY 
branche and the ontologists believed, evident to us 
immediately and prior to any discursive activity of 
the mind ; it is in virtue of that intellectual operation 
which is the activity most profoundly distinctive of 
man, namely ratiocination, that it becomes evident to 
us, and to attain that certainty reasoning must proceed 
not from the mere idea or notion of perfect being (the 
ontological argument of St. Anselm and Descartes) 
but from facts whose existence is established beyond 
dispute. St. Thomas, resuming the entire tradition 
of the past, shows by five different arguments how 
the conclusion God exists is imposed with absolute 
necessity on the human reason. There exist in the 
world motion or change ; beings and events newly 
brought into being ; things which are and are capable 
of not being ; things graded in degrees of perfec-tion, 
whose perfection, which consists in being, is more 
or less limited, obscured, mingled with imperfection ; 
irrational natures disposed towards an object or end, 
as is proved, not only by the complex system of the 
universe or the structure of Uving organisms, but even 
by the simple aptitude of every agent to produce its 
specific operation. To account for these various facts 
we are compelled—for under pain of absurdity we are 
obliged to stop at an ultimate explanation of existence 
—to admit a cause which moves without being 
moved, causes without being caused, cannot lack 
existence, contains in its purity the perfection of which 
things partake in greater or less degree, possesses an 
intellect which is the final ground of all natures and 
the first principle of all things. Such a cause we 
term God ; it is pure act, deriving its existence from 
itself {a se) . In other words, being itself is its nature 
258
THEODICY (natural THEOLOGY) 
or essence, it is subsistent Being itself, he who is. This 
conclusion, which for the philosopher involves the 
most sublime truths of metaphysics, is reached very 
simply by common sense, for it is in truth the most 
fundamental natural operation of the human under-standing, 
so that it can be denied only by denying 
reason itself and its first principles (the laws of identity 
or non-contradiction, sufficient reason, causality) ; and 
as the history of philosophy shows only too plainly, 
the mind has no other choice than between the alter-natives 
: " the true God or radical irrationality." ^ 
It is also the province of theodicy to show with 
what manner of knowledge we know God and in due 
course to study his nature and perfections, in particular 
the unity, simplicity, and immutability which are immedi-ately 
deducible from the perfection of underived 
existence {aseity), the fundamental character of pure 
act, and which prove most clearly that he is abso-lutely 
and essentially distinct from the world ; his 
relations with the world, his knowledge, his activity 
as Creator and Mover ; and finally the problems 
involved in the divine foreknowledge of contingent 
events, particularly man's free acts, and those arising 
out of the existence of evil in the universe. 
The school of Aristotle and St. Thomas teaches that 
God is known by the natural reason analogically, so 
that we perceive the divine perfections {being, unity, 
goodness, wisdom, love, etc.) in the mirror of creatures, 
without asserting any unity of nature, common 
measure, or proportion, or mixture or confusion of 
any kind between God and created things. This 
^ Garrigou-Lagrange, Dieu. son existence, sa nature, Paris, 3rd ed., 
1920.
. 
AN INTRODUCTION TO PHILOSOPHY 
teaching is opposed to two contrary errors, the error 
of the agnostics, who maintain that the Divine Being 
is beyond the ken of our intellect and God unknowable 
by the reason (sceptics ; phenomenalists ; positivists 
like Comte and Spencer, the Kantian school as a 
whole), and the error of the pantheists, who confuse 
the Divine Being with the being of created things 
(Parmenides, Heraclitus, the Stoics, Spinoza, the 
German metaphysicians after Lessing and Kant ; the 
modernists and immanentists) 
Philosophy of Aristotle and St. Thomas 
God is known by analogy and is absolutely distinct 
from creatures. 
Pantheism Agnosticism 
God confused with crea- God unknowable, 
tures. 
260
IX 
. 
THE PHILOSOPHY OF ART 
ETHICS 
The aim of the practical sciences is to know, not for 
the sake of knowing, but to procure by some action 
the good of man (other than the pure act of knowing 
truth). But the good of man can be understood in 
two different senses ; either of this or that particular 
good or of the good which is in itself alone the good 
of man and which, as we say, determines the signifi-cance 
of human life. 
the philosophy of art 
Of the different practical sciences which are con-cerned 
with the good of man from the first standpoint 
(that of particular goods, and not of the absolute 
good of human hfe), none, as we have pointed out 
already,^ is a philosophy. For none of these proposes 
to regulate human action in reference to the supreme 
cause in the practical order, that is (for in the practical 
order the aim or end pursued fulfils the function of 
cause or principle) in reference to the last end (the 
absolute good of man) 
Nor are these practical sciences sciences in the 
strict sense, for they do not proceed by way of demon-stration, 
drawing conclusions from their premisses. 
1 See p. 149. 
261
AN INTRODUCTION TO PHILOSOPHY 
They are rather arts than sciences, and belong im-mediately 
to the wide category of art, not to that of 
science. 
The essential character of art taken in its complete 
extension is to instruct us how to make something, 
so that it is constructed, formed, or arranged, as it 
ought to be, and thus to secure the perfection or 
goodness, not of the maker, but of the object itself 
which he makes. Art therefore belongs to the 
practical order in the sense that it instructs us how to 
make something, considering not the use we should 
make of our free will but the manner in which the 
work as such and in itself should be executed. We 
may thus say that art is concerned with what is to be 
ina.dc, factibile, 7Tot.Y]T6v. 
This formal character of making is fulfilled primarily 
in the material objects produced or fashioned by 
man {the factibile in the strict sense). But in a wider 
sense it is to be found also in works of a purely spiritual 
nature. In this case it goes beyond the sphere of 
practice as such, so far as practice is opposed to 
speculation and signifies an activity other than that 
of pure knowing. It is in this sense that there is 
making in the purely speculative order (a form of 
argument, a proposition, are works, but works of the 
speculative reason), and there are arts, logic, for 
example, which are speculative arts. 
In order, however, to establish a general theory of 
art and making we must have recourse to the highest 
and most universal concepts and principles of human 
knowledge. Such a theory therefore belongs to the 
domain of philosophy. 
The province of philosophy thus defined is indeed 
262
THE PHILOSOPHY OF ART. ETHICS 
practical, since it is concerned with making, and its 
object is to order from above the branches of practical 
instruction. Nevertheless, since it is in the strict 
sense a science, it cannot be essentially practical, 
but remains essentially speculative in virtue of its 
object and method of procedure ; moreover, it is 
extremely remote from actual practice. Indeed, not 
only has it no concern with the application of the 
rules of art to a particular work to be accomphshed, 
but further it formulates rules which are far too 
general to be capable of such immediate apphcation 
and to be correctly termed rules of art in the strict 
sense ; it is therefore practical only in an improper 
sense and very imperfectly. 
The individual arts alone (branches of study 
essentially practical) possess rules sufficiently detailed 
to be immediately applicable to a particular work, 
and their apphcation belongs solely to them. Further, 
with the exception of the fine arts, whose object, 
beauty, being itself universal and immaterial, enables 
philosophy to perform effectively though from a 
remote height her office of supreme arbiter, the arts 
since they are devoid of any um'versal character, except 
the fact of being arts, of which philosophers can take 
cognizance, escape her jurisdiction almost completely. 
If we would accurately describe this branch of 
philosophy, we should term it the philosophy of making, 
but we shall call it simply the philosophy of art. ^ Here 
^ The term aesthetics, which has now become current, would be 
doubly incorrect here. Modern writers understand by the word the 
theory of beauty and art, as though the philosophy of art were the place 
in which to treat questions concerning beauty considered in itself (such 
questions belong to ontology) , and as though art were confined to the 
fine arts (a mistake which vitiates the entire theory of art). Moreover, 
263
AN INTRODUCTION TO PHILOSOPHY 
we must first inquire what is the nature of art, if it is 
indeed, as St. Thomas teaches, a virtue of the practical 
intellect, and how it is distinguished on the one hand 
from the speculative virtues {understanding offirst principles^ 
knowledge, wisdom) , on the other from the moral virtues, 
prudence in particular ; how art is to be subdivided 
and the different arts classified ; and finally what are 
the first principles and distinctive conditions—though 
solely of the highest and most general order—of those 
arts which have beauty for their object (the fine arts) 
and which thus occupy a superior rank among the 
arts. 
ETHICS 
The practical science which aims at procuring man's 
unqualified good, his absolute good, is that of morals 
or ethics. Since its distinctive object is not the 
the word aesthetics is derived etymologically from sensibility {aiaddvofxai. 
= feel), whereas art, and beauty also, are matters of the intellect, quite 
as much as of feeling. 
Scholastic textbooks do not usually devote a separate treatise to the 
philosophy of art, and either study its problems in psychology alone, 
or, the better to explain the concept of prudence, in ethics. It would 
be necessary to classify the philosophy of art, like ethics itself, under 
natural philosophy, if we kept to the single standpoint of the 
specification of the sciences by their formal object. But if we adopt the 
wider standpoint of the end to which the sciences are ordered, we must 
distinguish practical from speculative philosophy, and it is equally 
necessary to distinguish, in practical philosophy itself, the philosophy 
of making and the philosophy of doing {cf. the author's Art and 
Scholasticism). This treatment presents the double advantage of 
corresponding to a very marked trend of modem thought, which tends 
to devote a special treatise (aesthetics) to the theory of art, and of 
returning to one of Aristotle's fundamental classifications : Tracra bidvoia 
fj irpaKTLKTf) i) iroirjTLKr) 7J OeuprjTiKTj {Metaph., ii, I, 1025 b 25). Cf. Top., 
vi, 6, 145 a 15, and viii, i, 157 a 10 ; Metaph., vi, i ; Nic. Eth., vi, 2, 
1 139 a 27. Hamelin {op. cit., pp. 81 sgq.) makes a convincing defence 
against Zeller of Aristotle's real opinion on this matter. 
264
. 
THE PHILOSOPHY OF ART. ETHICS 
perfection of the works produced and fashioned by 
man but the good and perfection of the agent himself, 
or the use he freely makes of his faculties, it is in the 
strict sense the science of action, the science of human 
acts (in technical language, of the agibile or Trpaxxov, 
that is to say, of the free use, so far as it is free, of our 
faculties) 
Ethics is as practical as any true science in the strict 
sense can be, for it teaches not only the most general 
rules ofremote application, but also the particular rules 
applicable to the particular action to be performed. 
But at the same time this science has in view, not 
some particular secondary end, but the last end (the 
absolute good) of man, the supreme cause in the 
practical order. It is therefore a philosophy. It is 
without qualification the practical philosophy. 
Note.—Though ethics is as practical as any science 
in the strict sense can be, we must not therefore suppose 
that it is essentially practical (no science vere et proprie 
dicta is essentially practical), or that it is sufficient to 
make men behave rightly. It supplies, it is true, rules 
immediately applicable to particular cases, but it 
has no power to make us constantly apply them as we 
should in particular cases,^ in spite of the difficulties 
1 On the contrary, the essentially practical sciences, that is to say the 
arts, themselves apply their rules to particular cases. These sciences 
are, strictly speaking, practical, but are not sciences in the strict sense 
but only improperly. 
There are thus many degrees of practicality. The philosophy of art 
(whose end is practical, and whose object an operabile, but to be known) 
provides no rules immediately applicable to particular cases. It is 
only improperly and very imperfectly practical. 
Ethics (whose end is practical, and whose object is also an operabile, 
but also to be known) does not apply, but provides rules immediately 
applicable to particular cases. It is as practical as a science in the 
265
— 
AN INTRODUCTION TO PHILOSOPHY 
arising from our passions and the complexity of 
material circumstances. It remains, therefore, 
essentially speculative in its final object {knowledge 
of human acts) and in its procedure (the deduction of 
truths from their premisses, not incitement to action) 
and is thus practical only in an improper sense. ^ If 
man is to do the right in the order of action moral 
science must be supplemented by the virtue of 
prudence^ which, if" we make use of it, makes us always 
judge correctly of the act we should perform, and will 
always that which we have thus judged to be right. 
On the other hand, ethics only supplies rules of 
human conduct in the natural order and in relation 
to man's last end as it would be, if it were a natural 
happiness. But since, in fact, man's last end is a 
supernatural good [God possessed not by the imperfect 
knowledge of human reason as such, but by the 
strict sense can be practical, but it is not strictly speaking or perfectly 
practical. 
The arts 
medicine, for example, or engineering—whose object is 
something to be done (not merely an operabile, but envisaged opera-biliter), 
provide rules immediately applicable and actually apply 
them to particular cases, but only by enabling us to judge of what is 
to be done, not by making us wiU to do it (for the artist can make a 
mistake and still be an artist (because he wills to make it). They are 
in the strict sense practical—but do not attain the highest degree of 
practicability. 
Finally prudence (whose object is also something to be done) applies 
to particular cases the rules of moral science and reason, not only 
teaching us to judge of the act to be performed, but making us employ 
as we ought our free activity (for the prudent man, as such, always wills 
the right). It is, strictly speaking, practical, and attains the highest 
degree of practicability. 
1 Cf. St. Thomas, Sup. Boet. de Trin., q. 5, a. i, arf 3 : Scientia moralis, 
quamvis sit propter operationem, tamen ilia operaiio non est actus scientiae, sed 
actus virtutis, ut patet V Ethic. Unde non potest diet ars, sed rnagis in illis 
operationibus se habet virtus loco artis et idea veteres dijfinierunt virtutem esse 
artem bene recteque vivendi, ut dicit August. X, de Civ. Dei. 
266
THE PHILOSOPHY OF ART, ETHICS 
beatific and deifying vision of the Divine Essence), and since 
his actions must be regulated in reference to this 
supernatural end and so as to enable him to attain it, 
ethics or philosophic morality is evidently inadequate 
to teach him everything he needs to know in order to 
act rightly. It must be completed and elevated by 
the teachings of revelation. 
The epithet practical, applied to ethics, does not 
merely mean having as its aim an activity other than that 
of simply knowing (in this sense practical, whether used of 
art or morals, is opposed to speculative) ; but more 
strictly concerned with action and behaviour (the 
TrpaxTov, the distinctive sphere of moral science and the 
moral virtues as contrasted with the TtoiTjTov, the 
distinctive sphere of art). 
The fundamental question which practical philosophy 
must answer before any other is in what consists 
(from the standpoint of the natural order) the last 
end or absolute good of man. It must then study 
the actions by which man approaches or departs 
from his last end, examining first their nature and 
inner machinery, then what constitutes their moral 
character, that is to say, renders them morally good 
or bad. Ethics must, therefore, study in themselves 
the supreme rule of such actions (questions which 
treat of the eternal law and the natural law) and their 
immediate rule (questions relating to the conscience); 
it must also study the intrinsic principles from which those 
acts proceed, that is to say, the moral virtues and vices. 
But since ethics is a practical science it must not 
be content with these universal considerations ; it 
must proceed to the more particular determination of 
267
AN INTRODUCTION TO PHILOSOPHY 
^ 
human acts and their rules. It is therefore obliged to 
study in great detail the rules which regulate man's 
conduct first so far as they concern the good of the agent 
himself and secondly so far as they concern the good 
of others (consequently the virtue o^justice). 
The latter inquiry introduces a number of most 
important questions pertaining to what is called 
natural right, and treating in the first place of man's 
obUgations to God (a question of natural religion), 
secondly of his obligations to his fellow-men. Here 
are discussed the questions which concern men as 
individuals [the rights of the individual, for example, 
the right of property), and those which concern them as 
members of a natural whole whose common good 
individuals must serve—the family and the poUtical 
society (the rights of society). 
Aristotle divided the science of morality, of human 
conduct [ethics in the wide sense) into three parts : 
the science of man's actions as an individual, ethics 
(in the stricter sense) ; the science of his actions as 
a member of the domestic society, economics ; the 
science of his actions as a member of the city (the civil 
society), politics.^ 
On the fundamental question of ethics—the question 
of man's last end—we find for the last time the schools 
of philosophy divided roughly into three groups. 
The school of Aristotle and St. Thomas teaches that 
the entire moral life depends on man's tendency to 
his sovereign good or happiness and that the object 
1 I.e. of religion as it would be apart from the supernatural order 
to which man has in fact been raised. 
2 See on this point Nic Eth., vi, 9, 11 42 a 9 ; Eud. Eth., i, 8, 12 18 
b 1 3, and the two first chapters of the Politics. Cf. Hamelin, op. cit., p. 85. 
268
. 
THE PHILOSOPHY OF ART. ETHICS 
in which this happiness consists is God—whom, more-over, 
we ought to love, not for our own sake, but for 
himself (precisely because he is our last end, that 
is to say, that which is willed and loved for itself, not 
for the sake of anything beyond) 
The schools which find the end and rule of human 
conduct in pleasure {hedonism, Aristippus, Epicurus), 
utility {utilitarianism, Bentham, John Stuart Mill), the 
state (Hegel and sociologists), humanity (Auguste Comte), 
progress (Herbert Spencer), sympathy (the Scottish 
school), pity (Schopenhauer) or the production of the 
superman (Nietzsche), assign as man's last end some-thing 
created, and thereby degrade him below himself. 
The schools which claim that virtue (the Stoics, 
Spinoza) or duty (Kant) is self-sufficient, either because 
virtue is itself happiness, or because the pursuit of 
happiness is immoral, assign as man's last end man 
himself, and thereby, while seeming to deify man, 
really, Hke the schools last mentioned, degrade him 
below himself, for the greatness of man consists in 
the fact that his sole end is the uncreated Good. 
Thomist Philosophy 
[Ethics of Happiness or the Sovereign Good) 
Man is ordered to a last end other than himself, and 
this last end is God. 
Moral Systems which degrade Moral Systems which deify 
Man Man 
Man is ordered to a last Man is not ordered to 
end other than himself, any last end other than 
and this end is something himself, his own virtue 
created [hedonism, Epicure- being his last end (iS'toia'^m), 
anism, utilitarianism, etc.). or his goodness not de-pending 
on any good for 
which he has been made 
[Kantism). 
269
AN INTRODUCTION TO PHILOSOPHY 
Thus on every one of the great problems of 
philosophy the doctrine of Aristotle and St. Thomas, 
when compared with the doctrines of other 
philosophers, appears as an eminence between two 
contrary errors. This is an additional argument for 
its truth to be added to those enumerated earUer.^ 
The truth, indeed, is not to be found in a philosophy 
which keeps the mean between contrary errors by its 
mediocrity and by falling below them, being built up 
by borrowing from both, balancing one against another 
and mingUng them by arbitrary choices made without 
the Hght of a guiding principle {eclecticism) ; it must be 
sought in a philosophy which keeps the mean between 
contrary errors by its superiority, dominating both, so 
that they appear as fragments fallen and severed from 
its unity. For it is clear that, if this philosophy be 
true, it must reveal in full what error sees only in part 
and distorted by a bias, and thus must judge and 
secure, by its own principles, and in the hght of its 
own truth, v/hatever truth error contains though it 
cannot distinguish. 
1 See pp. 99-101. 
270
CONCLUSION 
; 
Philosophy is thus divided into three principal 
sections : logic, speculative philosophy, practical philosophy 
or, if we take account of the subdivisions of these three 
sections, into seven main sections : minor logic, major 
logic ; the philosophy of mathematics, the philosophy of 
nature, metaphysics ; the philosophy of art and ethics. 
This order is represented by the following table : 
I J I. Minor or formal logic : the rules of reasoning 
LOGIC 1^2. Major OT material logic : the matter of resisoning 
3. The philosophy of mathematics : quantity  . —„,.,,. i the material world J 
'"^^^ &/ 
4. 1 he phuosobhy of nature : { , , , '^ 
 man psychology 
(truth (epistemolog^') criticism 
being in general ontology 
being a se (natural theology) theodicy 
II 
SPECULATIVE 
PHILOSOPHY 
III r6. The philosophy of art : making 
PRACTICAL 
PHILOSOPHY [7. Ethics OT moral philosophy : action, conduct 
The division of philosophy into speculative and 
practical depends, not on the specific character of the 
various philosophic sciences, but on the end which they 
pursue. If that end is knowledge alone, the philosophy 
is speculative ; if the good of man, it is practical.^ 
If the philosophic sciences are classified from the 
standpoint of their specific character, ^ ethics, which 
1 See St. Thomas, Sup. Boet. de Trin., q. 5, ad 4. 
' This specification depends essentially on the degree of abstraction 
or degree of immateriality of the object studied. 
271
AN INTRODUCTION TO PHILOSOPHY 
treats of the moral virtues ^ and whose formal object 
is human action, and the philosophy of art, which treats 
of the practical intellectual virtues and whose formal 
object is human making, are divisions of the science of 
man, which itself belongs to natural philosophy (though 
it enters also into metaphysics). From this point 
of view we can recognise as specifically distinct 
philosophic sciences only logic, metaphysics, and the 
philosophy of nature, also the philosophy of mathematics, if 
this is not regarded as a subdivision of metaphysics or 
of the philosophy of nature. 
1 Sic pertinet ad philosophiam [naturalem) , et est pars illius, quia agit dt 
anima ut est actus corporis, et consequenter de moralihus ejus, (John of 
St. Thomas, Cursus. phil., i, p. 732 ; Log., ii* q. 27, a. i.) 
272
Introduction of philosophy
Introduction of philosophy
Introduction of philosophy
Date Due 
Due Returned Due Returne- 
M' ^ 
'^' ' '" ^ 
^^n n - - 
[jm 1 3 2^^ 
; 
i 
j 
1 
I 
i 
V 
~
U'y.'yERSirY OF FLORIDA 
3 12b5 OMMflM^fli 2
Introduction of philosophy

More Related Content

PDF
DETAILED ARCHITECTURAL STUDY OF BRIHADESHWARA TEMPLE,THANJAVUR.pdf
PPTX
Judaism
PPT
History of Western modern Philosophy
PDF
An introduction-to-philosophy
PDF
The Study Of Philosophy A Text With Readings 7th Andrew Pessin
DOC
An introduction-to-philosophy
PPTX
5. method of-philosophy-raymund report
PPTX
L1 philosophy-130628222719-phpapp02
DETAILED ARCHITECTURAL STUDY OF BRIHADESHWARA TEMPLE,THANJAVUR.pdf
Judaism
History of Western modern Philosophy
An introduction-to-philosophy
The Study Of Philosophy A Text With Readings 7th Andrew Pessin
An introduction-to-philosophy
5. method of-philosophy-raymund report
L1 philosophy-130628222719-phpapp02

Similar to Introduction of philosophy (20)

PDF
Indian and Western Philosophies.pdf
PDF
Classical Philosophy A Contemporary Introduction 1st Edition Christopher Shields
PDF
The Philosophy Of Knowledge Kenneth T Gallagher
PPTX
Philosophical foundations and The Philosophy of Education.pptx
PDF
Unit 1
PPTX
Introduction to Philosophy - Definition & Nature.pptx
PPTX
Meaning and nature of philosophy -.pptx
PPTX
political philosophyfadgghtyhuu67u 1.pptx
PDF
Invading the invisibles-ja_edgarton-1931-359pgs-phi
PPTX
382604941-Definition-Nature-And-Branches-of-Philosophy.pptx
PPTX
Introduction to Philosophy of a human pe
PPTX
Philosophy Lecture 1.pptx
PPTX
Nature of philosophy
PPTX
Logic-and-Critical-Thinking preliminary ppt
PPT
Introduction to philosophy
DOCX
Philosophy
PPTX
Lecture 1 Introduction to Philosophy
PPTX
CHAPTER ONE & TWO LOGIC AND PHILOSOPHY.pptx
DOC
What is philosophy1
PDF
Basic Concepts of Ancient Philosophy Martin Heidegger
Indian and Western Philosophies.pdf
Classical Philosophy A Contemporary Introduction 1st Edition Christopher Shields
The Philosophy Of Knowledge Kenneth T Gallagher
Philosophical foundations and The Philosophy of Education.pptx
Unit 1
Introduction to Philosophy - Definition & Nature.pptx
Meaning and nature of philosophy -.pptx
political philosophyfadgghtyhuu67u 1.pptx
Invading the invisibles-ja_edgarton-1931-359pgs-phi
382604941-Definition-Nature-And-Branches-of-Philosophy.pptx
Introduction to Philosophy of a human pe
Philosophy Lecture 1.pptx
Nature of philosophy
Logic-and-Critical-Thinking preliminary ppt
Introduction to philosophy
Philosophy
Lecture 1 Introduction to Philosophy
CHAPTER ONE & TWO LOGIC AND PHILOSOPHY.pptx
What is philosophy1
Basic Concepts of Ancient Philosophy Martin Heidegger
Ad

More from Heritagecivil Kasetsart (20)

PPT
ศิลปะตะวันออก lol
PPT
มรดกอารยธรรมโลก
PPT
The heritages of world civilization
PDF
01999031 western music romantic
PDF
01999031 western music baroque.
PDF
01999031 western music classical.
PDF
Man lg handout s
PDF
Man lg eng hndout
PDF
PPT
World major religion
PPT
Philosophy history
DOCX
Philosophy and religion
PDF
Introduction of philosophy
DOCX
Philosophy and religion
PPT
มรดกอารยธรรมโลก(อ.ไพรินทร์)
PPT
มรดกอารยธรรมโลก
PPT
มรดกอารยธรรมโลกเพิ่มเติม ปี 2557
PPT
ยุคปัจจุบัน(อัพเดท2557)
ศิลปะตะวันออก lol
มรดกอารยธรรมโลก
The heritages of world civilization
01999031 western music romantic
01999031 western music baroque.
01999031 western music classical.
Man lg handout s
Man lg eng hndout
World major religion
Philosophy history
Philosophy and religion
Introduction of philosophy
Philosophy and religion
มรดกอารยธรรมโลก(อ.ไพรินทร์)
มรดกอารยธรรมโลก
มรดกอารยธรรมโลกเพิ่มเติม ปี 2557
ยุคปัจจุบัน(อัพเดท2557)
Ad

Recently uploaded (20)

PPTX
Share_Module_2_Power_conflict_and_negotiation.pptx
PDF
LEARNERS WITH ADDITIONAL NEEDS ProfEd Topic
PDF
LIFE & LIVING TRILOGY - PART (3) REALITY & MYSTERY.pdf
PDF
Mucosal Drug Delivery system_NDDS_BPHARMACY__SEM VII_PCI.pdf
PDF
MBA _Common_ 2nd year Syllabus _2021-22_.pdf
DOCX
Cambridge-Practice-Tests-for-IELTS-12.docx
PPTX
Unit 4 Computer Architecture Multicore Processor.pptx
PDF
Uderstanding digital marketing and marketing stratergie for engaging the digi...
PDF
Paper A Mock Exam 9_ Attempt review.pdf.
PPTX
B.Sc. DS Unit 2 Software Engineering.pptx
PDF
BP 505 T. PHARMACEUTICAL JURISPRUDENCE (UNIT 2).pdf
PPTX
Module on health assessment of CHN. pptx
PDF
ChatGPT for Dummies - Pam Baker Ccesa007.pdf
PPTX
Virtual and Augmented Reality in Current Scenario
PDF
Empowerment Technology for Senior High School Guide
PDF
David L Page_DCI Research Study Journey_how Methodology can inform one's prac...
PDF
What if we spent less time fighting change, and more time building what’s rig...
PDF
1.3 FINAL REVISED K-10 PE and Health CG 2023 Grades 4-10 (1).pdf
PDF
BP 704 T. NOVEL DRUG DELIVERY SYSTEMS (UNIT 2).pdf
PDF
CISA (Certified Information Systems Auditor) Domain-Wise Summary.pdf
Share_Module_2_Power_conflict_and_negotiation.pptx
LEARNERS WITH ADDITIONAL NEEDS ProfEd Topic
LIFE & LIVING TRILOGY - PART (3) REALITY & MYSTERY.pdf
Mucosal Drug Delivery system_NDDS_BPHARMACY__SEM VII_PCI.pdf
MBA _Common_ 2nd year Syllabus _2021-22_.pdf
Cambridge-Practice-Tests-for-IELTS-12.docx
Unit 4 Computer Architecture Multicore Processor.pptx
Uderstanding digital marketing and marketing stratergie for engaging the digi...
Paper A Mock Exam 9_ Attempt review.pdf.
B.Sc. DS Unit 2 Software Engineering.pptx
BP 505 T. PHARMACEUTICAL JURISPRUDENCE (UNIT 2).pdf
Module on health assessment of CHN. pptx
ChatGPT for Dummies - Pam Baker Ccesa007.pdf
Virtual and Augmented Reality in Current Scenario
Empowerment Technology for Senior High School Guide
David L Page_DCI Research Study Journey_how Methodology can inform one's prac...
What if we spent less time fighting change, and more time building what’s rig...
1.3 FINAL REVISED K-10 PE and Health CG 2023 Grades 4-10 (1).pdf
BP 704 T. NOVEL DRUG DELIVERY SYSTEMS (UNIT 2).pdf
CISA (Certified Information Systems Auditor) Domain-Wise Summary.pdf

Introduction of philosophy

  • 3. Digitized by the Internet Archive in 2010 with funding from Lyrasis IVIembers and Sloan Foundation http://www.archive.org/details/introductiontophOOmari
  • 7. AN INTRODUCTION TO PHILOSOPHY
  • 8. BT THE SAME AUTHOR THREE REFORMERS LUTHER—DESCARTES—ROUSSEAU PRAYER AND INTELLIGENCE TRANSLATED BY ALOAR THOROLD ART AND SCHOLASTICISM TRANSLATED BY J. F. SCANLAN THE THINGS THAT ARE NOT CAESAR'S TRANSLATED BY J. F. SCANLAN ST. THOMAS AQUINAS—ANGEL OF THE SCHOOLS TRANSLATED BY J. F. SCANLAN RELIGION AND CULTURE TRANSLATED BY J. F. SCANLAN FREEDOM OF THE INTELLECT TRANSLATED BY F. J. SHEED FREEDOM IN THE MODERN WORLD TRANSLATED BY R. o'sULLIVAN
  • 9. AN INTRODUCTION TO PHILOSOPHY BY JACQUES MARITAIN TRANSLATED BY E. I. WATKIN NEW YORK SHEED & WARD, INC.
  • 10. Published by Sheed & Ward, 31 Paternoster Row, London, E.C.4 and by Sheed & Ward, Inc., 63 Fifth Avenue, New York First printing, March, igjy Twelfth printing, April, ig4'j Printed in the United Stales ofAmerica
  • 11. PUBLISHER'S NOTE The French edition of this work, under the title Introduction G^ndrale a la Philosophic, appears as the first volume of seven, which deal with Formal Logic, Theories of Knowledge, Cosmology, Psychology, Metaphysics, Ethics, ^Esthetics, and the History of Philosophy. But, since six out of the seven volumes remain to be written, it has been thought better to issue the present volume quite independently. The series as a whole is intended to provide text-books for a regular university course as it is found in France, and with that particular end in view prints in larger type those paragraphs which the student should read first, and in smaller type those paragraphs which are merely explanatory or expansive of them. This schematising has been abandoned to make the volume serve for general readers, no university course being envisaged in England or America. The translation has been madefrom the eleventh French edition.
  • 13. PREFACE My chief aim in composing an Elements of Philosophy series, to which this book may serve as an introduction, is to give a faithful presentation of the system of Aristotle and St. Thomas, and in its Ught to judge the important systems which have followed each other during the last three centuries and the principal problems discussed by modern philosophy. I have tried to adapt the method of exposition to contemporary conditions, and in parti-cular have followed of set purpose a progressive order of exposition—as far as possible the order of intellectual discovery—never appealing to any truth not already known and understood, and never introducing a new notion or proposition for which the way has not been prepared by those which have gone before it and led up to it. The method has obliged me to depart on several points from the procedure of the traditional text-books—above all, considerably to magnify the importance and extend the scope of this Introduction. Yet, thus, I have but returned to the method followed by Aristotle himself. The first three books of his Metaphysics are, in fact, nothing but an extensive introduction. A work of this kind, if it is to be thorough, demands the detailed discussion of certain points, without which the study it seeks to promote would lose all its value as a mental discipline. I should be untrue to tradi-
  • 14. PREFACE tional philosophy if I reduced it to a few main theses which have lost their freshness, and a few common-places of a spiritualist metaphysic, and neglected to bring out its fine intellectual contours and display its power of penetrating analysis. The present work is intended for beginners. It can therefore make no attempt to reproduce the depth or the wealth of subtle dialectic to be found in treatises written for specialists, and remains strictly elementary. It must, however, preserve the scientific character proper to a philosophical exposition. Some readers may take alarm at scholastic termino-logy. Yet no science, no discipline, no form of sport, even, or industry, can dispense with a special termino-logy— often far more arid and artificial than the vocabulary of philosophy. To require that philo-sophers should use everyday language implies that their science is just an enterprising topic of conversa-tion, idle arm-chair speculation for after dinner. On the other hand, it may legitimately be demanded that no technical term be used until it has been clearly defined. Finally, I would say that, if the philosophy of Aristotle, as revived and enriched by St. Thomas and his school, may rightly be called the Christian philosophy, both because the Church is never weary of putting it forward as the only true philosophy and because it harmonises perfectly with the truths of faith, nevertheless it is not proposed here for the reader's acceptance because it is Christian, but because it is demonstrably true. This agreement between a 8
  • 15. PREFACE philosophic system founded by a pagan and the dogmas of revelation is no doubt an external sign, an extra-philosophic guarantee of its truth ; but it is not from its agreement with the Faith, but from its own rational evidence, that it derives its authority as a philosophy. Nevertheless, reason and faith, while distinct, are not separate, and, since I am writing principally for Christian readers, I have not denied myself an occa-sional reference to knowledge familiar to every Catholic, or to certain theological applications of philosophic principles, the better to put philosophy in its proper place in Christian minds, or to help them to maintain the unity of their thought. The fact remzdns that in our arguments and in the very structure of our exposition of philosophy, it is not faith, but reason, and reason alone, which occupies the entire ground and holds undivided sway.
  • 17. CONTENTS INTRODUCTORY PACK 17 PART ONE THE NATURE OF PHILOSOPHY I. PHILOSOPraC THOUGHT BEFORE PHILOSOPHY IN THE STRIC3T SENSE , . . . .23 Introduction . . . . -23 Primitive tradition .... 24 The Semites and the Egyptians . . 25 The Indo-Europeans : ... 26 (a) The Persians .... 26 (b) The Indians : . . . -27 (i) Brahmanism ... 28 (ii) Buddhism • • • 33 (iii) Other schools . . 36 (c) The Chinese .... 38 Limitations of human wisdom . . 42 The Greeks the chosen people of reason 44 THE PRE-SOCRATIC PHILOSOPHERS ... 46 The Sages ..... 46 The lonians : . . . . -47 (a) Thales and his successors . . 48 (b) The great physicists : . . 50 (i) Heraclitus ... 50 (ii) Democritus ... 52 (iii) Anaxagoras • . . 54 The Italians : Pythagoras • . • 55 The Eleatics : Parmenides ... 60 u. II
  • 18. CONTENTS m. THE SOPHISTS AND SOCRATES . Introduction The sophists Socrates .... (a) Ethics and knowledge (b) Irony, maieutic, dialectic (c) Moderate intellectualism 64 65 68 69 70 72 IV. PLATO AND ARISTOTLE . The minor Socratics . Plato .... (a) His theory of ideas . (b) His system of philosophy (c) Its limitations . Aristotle .... (a) Corrections of Plato . (b) The Aristotelian system (c) Aristotle's works Aristotle and St. Thomas . Philosophia perennis 74 75 75 78 81 82 83 87 92 97 99 V, DEFINITION OF PHILOSOPHY . Scientific knowledge . Its material object Its formal object Conclusion I Further considerations 102 103 107 108 loB VI, PHILOSOPHY AND THE SPECIAL SCIENCES Philosophy judges the special sciences It governs them It defends them It is pre-eminendy free Further observations . Conclusion II 12 III 113 117 118 118 123
  • 19. CONTENTS VII. PHILOSOPHY AND THEOLOGY . . . . I24 Nature of theology . . . . 1 24 Theology judges philosophy . .126 Philosophy submits to theology its conclusions, not its premises . 126 Philosophia ancilla theologiae . . .129 Further considerations . . .129 Conclusion III . -132 VUI. PHILOSOPHY AND COMMON SENSE . . '133 Unscientific knowledge . . -133 Philosophy is derived from common sense, understood as the natural apprehension of first principles . 136 Common sense may accidentally judge philosophy . . . . .138 Conclusion IV . . 141 The method of philosophy . . 141 PART TWO THE CLASSIFICATION OF PHILOSOPHY I. THE MAIN DIVISIONS OF PHILOSOPHY . Logic. Theoretical philosophy. Prac tical philosophy .... . 147 Their objects 151 Conclusion V . 153 II. LOGIC 154 Correct reasoning Ideas and images 154 154 Conclusion VI . 156 Individual and universal . 157 Conclusion VII ^59 147 13
  • 20. CONTENTS The problem of universals (a) Nominalism (b) ReaUsm . (c) Moderate Realism 159 159 160 160 III. THE PHILOSOPHY OF MATHEMATICS AND THE PHILOSOPHY OF NATURE The term body . The philosophy of mathematics The philosophy of nature (a) Mechanism (b) Dynamism (c) Hylomorphism Psychology Problem of the origin of ideas Conclusion VIII . 163 • 163 164 . 166 . 166 169 . 170 Abstraction : Problem ofhuman nature Conflicting schools .... 172 172 174 163 IV. CRITICISM (ePISTEMOLOOY) Being qua being . Criticism .... Problem of truth Conclusion IX Conflicting schools : . (a) Scepticism (b) Rationalism (c) Moderate intellectualism Problem of the object of the intellect Conclusion X . Being and intelligibility Conclusion XI . 14 178 178 179 181 181 182 182 183 185 187 187 187 178
  • 21. CONTENTS V. ONTOLOGY : ESSENCE Problems of ontology . Essence (a) In the wide sense {b) In the strict sense Characteristics of this essence Conclusion XII Further observations . Our intellect can apprehend essence Conclusion XIII Further observations . Essence is universal in the mind . Conclusion XIV Individual nature and matter : (a) Individual nature (b) First matter (c) Archetypal being (d) Nature, essence, and quiddity 189 189 191 191 194 197 201 201 203 203 203 205 207 207 208 209 210 213 VI. ONTOLOGY ! SUBSTANCE AND ACQDENT Origin of these notions Substance .... Conclusion XV Further observations Accident ..... Conclusion XVI Further observations . Conflicting schools The individuality of substance . (a) Substantia prima, substantia secunda (b) Per se, a s«, in se . . . 217 217 222 224 224 227 227 228 230 233 233 236
  • 22. CONTENTS VII. ONTOLOGY : ACT AND POTENTIALITY Origin of these notions (a) Identity and change . (b) Their apparent incompatibility (c) Solved by the concept potentiality Potency or potentiality Act Conclusion XVII The nature of change Act and potentiaHty in things Axioms i-vii .... Conflicting schools Terminology : . (a) Material and formal (b) Virtual and formal (actual) (c) Implicit and explicit (d) In express act, in accomplished act . . . . . 239 239 239 241 242 242 244 245 246 246 248 250 251 252 254 255 255 VIII. THEODICY (natural THEOLOGY) Subsistent being itself 257 257 IX. THE PHILOSOPHY OF ART. ETHICS Introduction The philosophy of art Ethics Divisions of ethics Conflicting schools 261 261 261 264 267 268 CONCLUSION .... Classification of philosophy 271 271 16
  • 23. INTRODUCTORY ' Philosophers were once called wise men. It was Pythagoras who first invented the term philosophy (91X1* TT^i^ (To^iat;', love of wisdom),^ observing that wisdom belongs in the strict sense to God alone, and for that reason not wishing to be called a wise man, but simply a friend or lover of wisdom. His modesty was itself a mark of great wisdom, for the sublimity and difficulty of the highest truths, and the weakness of our nature "in so many respects enslaved," forbid man to acquire " a property right in wisdom " such that he can employ it in entire freedom. As a result of the many necessities to which he is subject, he holds it only by an insecure title, so that he may be termed not wise, but far more truly a beggar at wisdom's door. Nevertheless philosophy is nothing other than wisdom itself so far as it is accessible to human nature. It is not a wisdom supernaturally infused into our souls which man possesses in virtue of a superhuman illumination. Neither is it a wisdom wholly sponta-neous and unconscious (such as within its limits is the prudence of animals, and even the wisdom of simple souls), which he possesses in virtue of a natural instinct. It is the wisdom of man as man, which he ' Cicero, Tusc, v. 8 ; cf. Diogenes Laertius, i, 12. » Aristotle, Metaph. i, 2, 982 b. St. Thomas, In I Metaph., 1. 3. Cf. De Veritate, q., 7, a. 7. 17
  • 24. AN INTRODUCTION TO PHILOSOPHY acquires by the labour of his intellect, and it is for that very reason that his wisdom is gained with such difficulty and held so insecurely, and that those who seek it should be called philosophers rather than wise men. Such is the nature of philosophy derived from the etymology of the term and its employment in ordinary speech. A philosopher is a man humanly wise. And the man who devotes himself to philosophy, by so doing undertakes to show his fellows the sublimest views at which man's understanding can arrive of the great problems which solicit the mind of the race. The definition of philosophy as " human wisdom " is still a superficial definition, and a nominal definition, which simply renders agreement possible as to the sense of the term. To attain a more profound defini-tion, a real definition which reveals the nature of the object, we shall study in the sequence of concrete history the formation or genesis of what men have agreed to call philosophy. In so doing we shall follow, so far as it is possible in an explanatory text-book, the actual method of Aristotle, too often neglected by books which teach his conclusions, but apparently ignore his spirit. That great reahst advanced nothing a priori and always studied the historical development of a problem before he proposed his own solution, which thus appeared as the natural goal of a process of discovery. Such a method will no doubt compel us to undertake a considerable digression into the field of history, but it is, nevertheless, in our opinion indispensable. On the one hand, from the practical and educational standpoint, an account of the historical origins of i8
  • 25. INTRODUCTORY philosophic thought is the best method of acquainting beginners with the problems of philosophy, introducing them into the world, entirely new to them, of rational speculation, and furnishing them, incidentally, with much extremely useful knowledge. Their first requisite is to know what they are studying, and to possess a sufficiently live and accurate notion of the problems of philosophy presented in their simplest form. On the other hand, in justice to our subject itself, to state straight away, with no previous examination or concrete justification, conclusions relating to the nature of philosophy, its object, dignity, and so forth, would be to present the traditional conception of philosophy under an arbitrary and a priori aspect wholly alien to it, and to risk enslaving our pupils to empty formulae. By beginning, on the contrguy, with a brief outHne of the history of ancient philosophy up to Aristotle, that is to say until the conclusion of its formative period, we display philosophy in its origin and construction, and thereby show how the transition was effected between the teaching of common sense and the scientific knowledge of philosophers, how the great philosophic problems arose of themselves, and how a particular conception of philosophy, which will be put later to the test of discussion, results inevitably from this historical inquiry, and naturally forces itself upon the mind. We need not fear to insist upon these preliminary questions, which we shall have to consider again from another angle in criticism. They concern the very existence, the nature, and the value of philosophy. 19
  • 27. PART ONE THE NATURE OF PHILOSOPHY
  • 29. PHILOSOPHIC THOUGHT BEFORE PHILOSOPHY IN THE STRICT SENSE Philosophic speculation, precisely because it is the supreme achievement of reason, is unknown to all the so-called primitive races. Indeed, even of the civilisations of antiquity the greater part either have possessed no philosophy or have failed to discover its true nature and distinctive character. In any case, philosophy only began to exist at a very late period about the eighth and especially the sixth century B.C., and then found the right path to truth by a success which must be regarded as extra-ordinary when we consider the multitude of wrong roads taken by so many philosophers and philo-sophic schools. Nevertheless, some of the most elementary truths with which philosophy deals were known long before philosophy itself had come to birth, and the more important of these are to be found in a more or less rudimentaiy form and more or less seriously cor-rupted among all the peoples of antiquity, even at the most remote epochs. But it was not from the philosophers that these peoples had learned them ; their knowledge was derived in part from that wholly spontaneous and instinctive exercise of reason which we call common sense, but above all from primitive tradition. 23
  • 30. AN INTRODUCTION TO PHILOSOPHY The most reliable inductions of history combine with the conclusions of theology ^ to prove the existence of a primitive tradition, common to the different branches of the human race and going back to the origin of mankind. And even in default of any positive sources of information, it is a very reasonable conjecture that the first man received from God knowledge together with existence, that by education he might complete the work of procreation. But was it possible that this knowledge, together with the primitive religion in which it was incorporated, could be transmitted in its integrity by the human race ? We have, on the one hand, truths of the loftiest sublimity to be handed down from one generation to another, yet, on the other, an intelligence dominated by the senses and imagination. A dis-proportion so extreme inevitably deteriorated the tradition received at the outset, as little by little the rust of obHvion gathered upon it, error defiled it, and it fell a prey to the corruptions of polytheism and the more degraded forms of religion (animism, totemism, idolatry, magic, etc.). Nevertheless, in spite of the changes which it underwent, the primitive tradition has preserved for mankind throughout the ages a deposit, progressively diminishing no doubt, of fundamental truths. In this deposit were included many philosophic conceptions—that is to say, concep-tions which concerned the most sublime problems within the scope of reason. But since they were taught only by a religious tradition which corroborated the instinctive teachings of common sense, they were 1 p. Lemonnyer, O.P, (following Schmidt), La Rivilaiion primitive et les doTtnSes actuelles de la science. Paris, 191 4. 24
  • 31. PRE-PHILOSOPHIG THOUGHT ; known in a pre-philosophic fashion and existed in a pre-philosophic state. It is not surprising that all peoples in the primitive^ stage of history were ignorant of philosophic speculation. But it is more astonishing that even certain civilisations were devoid of philosophy—for example, the Semitic, and the Egyptian, which is, in this respect, in the same category as the Semitic. Despite the high level of scientific culture reached by the intellectual aristocracy of these races, the sole philosophic conceptions, it would seem, which the Egyptians and Chaldeans possessed were a few very general ideas, implicit in their religion, concerning the Deity, the human soul and its state after death, and the precepts of morality. These truths, which, moreover (as in the case of every race), are purer the further back we follow their history, were never made the subject of rational study and specula-tion, but were simply accepted, as also were their scientific beliefs, as part of a sacred tradition. Religion took the place of philosophy, and from religion these races received certain philosophic truths philosophy they had none. In this matter the Jews did not differ from their fellow Semites. Scornful of human wisdom and the achievements of pure reason, and, indeed, without aptitude for such investigations, they produced no philosophers (at least not before Philo, who was a contemporary of * Primitive in respect of a particular brainch of the great human tree and so far as our knowledge of the past extends, but not primitive in the absolute sense. Far behind what we term the primitive state of the peoples known to us lies a long stretch of human history of which we know nothing. 25
  • 32. AN INTRODUCTION TO PHILOSOPHY Jesus Christ), but they possessed the prophets and the Law. All the great Indo-European civilisations, on the other hand, manifest an impulse, which no doubt took widely different forms, towards rational and, in the strict sense, philosophic speculation. But, except in Greece (and to a very partial extent in India), this impulse nowhere succeeded in achieving an indepen-dent scientific discipline distinct from religion. A traditional religion did not in this case take the place of philosophy, but philosophy, or, we should rather say, human wisdom, penetrated religion and was confused with it. The wise man fulfilled a sacred function. He was not the head of a philosophic school, but the founder of a religious sect, if not of a new religion. {a) Among the Persians,'^ whose original religion, so far as we know it from inscriptions, was a fairly pure monotheism, Zoroaster or Zarathustra founded Mazdeism or Zoroastrianism (about the eighth or sixth century B.C. ?), a powerful achievement of speculation which systematised (and incidentally distorted) certain funda-mental truths derived from the primitive tradition, * In this summary review of the great Aryan religions, we have been obliged not only to isolate by a process of abstraction the intellectual aspects of those religions with which the philosopher is concerned, but, moreover, to simplify considerably and reduce to an artificial classification doctrines whose vast and fluctuating complexity (this is especially true of Brahmanism and Buddhism), and occasional inconsistency, dismay historians. It should be added that the explana-tions of Oriental thought given by scholars are still largely conjectural and, in all probability, especially as far as philosophy is concerned, in many cases extremely inadequate. 26
  • 33. PRE-PHILOSOPHIG THOUGHT in the attempt to give a rational explanation of the vast problem which has faced human thought from the outset, the problem of evil. By his failure to perceive that God is the sole supreme principle and the source of everything which exists, so far as it partakes of being, and that evil is mere privation of being without positive existence, and therefore that no creature is evil by nature, Zoroaster ended in dualism and taught the existence of two principles imcreated and co-eternal, the principle of Good {Ormuzd) and the principle of Evil (Ahriman), who share the dominion of the universe and whose un-relenting struggle constitutes its history. So far as Ahriman is to be identified with the rebel angel of primitive tradition, Zoroastrianism tended to make the devil a god striving against God. (b) Among the peoples of India, whose intellectual and religious history is far more complex (since in this field no certainty has yet been reached, we present the interpretation of their beliefs which seems to us most probable), we witness a remarkable pheno-menon. When the original religion—the primitive reUgion of the Vedas ^—no longer proved sufficient to satisfy the intellectual demands or social needs of a more advanced civilisation, philosophic notions, which seem to have originated as interpretations of sacrifice and other sacred ritual, but developed in a spirit hostile to the ancient traditions and the cult of the ^ The most ancient among the religious books of the Hindus ( Veda means knowledge), the Rig-Veda, is apparently not older than the twelfth century b.c. Vedic religion seems to have been an incoherent polytheism coloured by a vague pantheism. 27
  • 34. AN INTRODUCTION TO PHILOSOPHY gods, found a home among the sacerdotal caste and took possession of the priesthood. To reconcile the perpetuation of their office with their new opinions, the priests, while continuing to perform the traditional ceremonies, directed their worship no longer to the old gods, but to the undefined and secret forces of the universe. (i) This resulted, after a period of confusion, in the formation of a new system, Brahmanism (or Hindu-ism), which is essentially a philosophy, a metaphysic, a work of human speculation, but being, so to speak, clothed in the ornaments of the sanctuary, was invested from the outset with the sanctions and attributes of a religion. A divine origin was ascribed to the books in which it was taught (the Brahmanas and Upamshads)^ and they could be obtained only from the priests. Hence Brahmanism may be called a sacred, hieratic or theological metaphysic, and already in the eighth century B.C. the supremacy of the priestly caste among the Hindus seems to have realised in its fashion that social and spiritual sovereignty of the philosopher-priest and the religion of science which was the dream of certain nineteenth-century thinkers. It is true that the science which those thinkers wished to invest with a sacred character was the science of phenomena, or, as it was termed, positive science, which is not wisdom, even human, and, as Auguste Gomte justly observed, is incapable of producing order in any department. The human science to which Brahmanism gave a divine character was, on the contrary, the science of ultimate realities, metaphysics, human wisdom in the strict sense : a powerful effort of metaphysical thought (so far as we can judge of it 28
  • 35. PRE-PHILOSOPHIG THOUGHT from documents whose interpretation is still far from certain), but the product of reason still untrained, incapable of maidng the necessary distinctions and of avoiding internal contradictions, seduced by the dream of an intuitive knowledge of the All, angelic rather than human, and doomed by its very ambition. This system, at least when we consider its pre-dominant tendencies, taught that the First Principle of the world, named Brahma ^ or Atman,^ constitutes in himself the intimate reality of everything which truly exists, whence logically follows pantheism, or the identification of God with his creation.^ Nevertheless, 1 From the name of the occult and sacred force which gave ritual its efficacy and pervaded all things. Originally regarded as the first emanation of the supreme God, it became for the Brahmans the unique source of being. The masculine noun, Brahma, designates the First Principle as God and Lord, the neuter, Brahman, as the one imp>er5onal substance. 2 From the name of the principle of life (the " self " transcending the phenomenal individual), which was regarded as animating man and the imiverse. ' The term pantheism is relatively recent, having been introduced into the vocabulary of philosophy by Toland in the eighteenth centiiry. But the doctrine it designates is as ancient as the earUest philosophical errors. For a system to be pantheistic, it need not explicitly identify God and creatures (very few pantheists fulfil this condition). It is sufficient that its teachings are logically ineconcilable with an absolute distinction between God and creatures. This observation is particularly important for the study of Oriental philosophies, of which pantheism is the original sin. Indeed, it arises in their case from the very method of thought they employ, which appears to consist primarily in the treatment of analogous concepts (realised differently in different objects) as though they existed as such outside the mind, which led them to conclude that things which re-main the same become on different planes of reality essentially different. For example, Atman is both the supreme principle of the universe, transcending all multipUcity, and the principle which distingiiishes and constitutes every personality. Like the Schoolmen, but for 29
  • 36. AN INTRODUCTION TO PHILOSOPHY an attempt was made to avoid this conclusion. The Supreme Principle, which possesses neither personality nor knowledge, to which no attribute can be applied, which is absolutely unknowable by any concept, however universal, not even by the concept of being, so that it must be called Nothing or Non-Being, is the sole true reality. Therefore the existence of everything multiple or limited, everything we can know by our senses or even by our concepts, is as such illusion, mere appearance. This is idealism, the denial of the reality of the world and of things. But the bare existence of this appearance or illusion is an evil, indeed evil pure and simple. The existence of individual objects and of this cosmic delusion which is called Nature (Maya), and which keeps us captives of the manifold and the transitory, is essentially evil and the source of all suffering. The problem of evil, therefore, seems to dominate the entire speculation of the Indian metaphysicians, as also of the Persian sages. But the Persians, whose bent was practical, always considered evil under the aspect ofsin, and, obsessed with the differences between moral good and evil, which they attempted to use as a criterion to divide beings into two metaphysical categories, ended in dualism. The Hindus, on the different reasons, the Indians distinguish between the personality (which is for us the spiritual subsistence of the soul) and the material individuality (which arises from the dispositions of the body). This mode of thought, which we meet again more or less emphasized in every doctrine of theosophic orientation, makes it possible to avoid the appearance of pantheism, because its inherent self-contradiction permits the affirmation of essential differences between terms which should logically be identified. But, precisely because these affirmations are only possible in virtue of a fundamental self-contradiction, it inevitably involves a real pantheism. 30
  • 37. PRE -PHILOSOPHIC THOUGHT contrary, exclusively occupied with contemplation, regarded evil pre-eminently under the aspect of suffering, or rather privation, in the sense in which metaphysicians understand the term.^ Led astray by a profound realisation of a great truth which they were unable to apprehend clearly (for while it is very true that it were better for us not to exist than to exist without being imited to God, they believed it were better for all things not to exist than to exist without being God), they ended in a pessimism which, though undoubtedly very different from the romantic pessimism of a Schopenhauer, was primarily the barren renunciation of a proud intellect, and attempted to be self-sufl5cient. What, then, in their conception did wisdom teach man ? It taught him to free himself from suffering and illusion, and with that object to rid himself of all individual existence. The Brahmans held the doctrine of the transmigration of souls, or metempsychosis ; they believed that souls, on the death of the organism which 1 From this point of view Indian speculation may be said to afford a prominent example of pure metaphysical intellectualism. Regarding things solely from the standpoint of intellectual speculation and the universal order, and not from the standpoint of the rectitude of the human will and that particular order by which man is ordered to his last end, it quickly came to lose sight almost entirely of the notion of moral good and evil, and its ethics consists primarily in a metaphysical purification, directed exclusively to a particular ideal of intellectual knowledge. An analogous tendency is observable in every system which confuses by an exaggerated intellectualism the moral with the metaphysical order (a confusion which is glaring in Spinoza's Ethics, for example) and, failing to recognise that God is not only the provisor universalis of creation, but also the provisor particularis of the moral life (cf. St. Thomas, Sum. TheoL i, q. 103, a. 8, with Cajetan's Commentaty) , ends by claiming to transcend the distinction of good and evil and denying the existence of moral evil. 31
  • 38. AN INTRODUCTION TO PHILOSOPHY ; they had animated, passed into another organism, thus living successively in different bodies of men, animals, or plants.^ The punishment of the wicked and foolish consisted accordingly in continuing to undergo in a series of reincarnations the pain of individual existence. The soul of the wise man, on the contrary, was delivered from the yoke of transmi-gration ; absorbed or reabsorbed in Atman, it escaped the sufferings of the world by losing all distinctive individuality. The ethics of Brahmanism teaches the means whereby this deliverance can be achieved ; the wise man progresses towards that goal in this life by means of contemplation. Brahmanism understands that contemplation is a beginning of beatitude in this life ; but, as it mistakes the nature of beatitude, so it mistakes the nature of contemplation. The con-templation which it claims to teach is, in fact, only a metaphysical contemplation, or rather a species of supra-rational vision, which it expects to achieve by the merely natural powers of the created intellect unlike Christian contemplation, it is the product of the intellect alone, not of supernatural charity and the 1 S05 at least, metempsychosis is currently understood. It is not unlikely that this interpretation of the doctrine is the popular translation of a doctrine less crude, according to which every being passes through an indefinite series of states or cycles of existence, each of which is only lived once, and our earthly existence is simply one particular state among many others. If this be the case, the doctrine of successive reincarnations originated in an unintelligent distortion of this theory, still further corrupted when it was introduced into the West. (The possibility, however, remains that originally the Pythagoreans and Orphics understood the transmigration of souls in a symbolic sense.) It is also possible, on the contrary, that the theory in question was a learned interpretation, elaborated by the Indian metaphysicians, of a popular belief in transmigration. 32
  • 39. PRE - P HI LO S O P HIC THOUGHT infused wisdom which accompanies it. Its aim is union with God by knowledge, not by love. Instead of admitting an activity overflowing from its own superabundance, it withdraws from activity of any kind, which it abandons wholly to the inferior powers. By this metaphysical contemplation, Brahmanism proposes to put us gradually in possession of our last end and initiate us into the blessed state of the delivered. Since it thus strives to reach by man's unaided powers heights which grace alone can attain, it results in a pseudo-mysticism of a purely intellectual character (in contrast to other, purely emotional, forms of false mysticism) in which the wise man, hoping not only to be united with God, but to blend with him, intoxicates himself not with God, but with his own self-annihilation. Hence (apart from those instances of genuine spirituality which grace is always free to produce) a host of counterfeits of supernatural mysticism, also of ascetic exercises and methods, including among their baser forms (with the fakirs) those tours de force of exaggerated asceticism which prove that the mortification of the flesh, when not regulated by reason and dictated by love, can be as fallacious as pleasure. Naturalism is thus the final characteristic and the capital vice of Brahmanism,^ as indeed of philosophic mysticism in general, whether it be the product of Brahmanism, Buddhism, neo- Platonism, or Islam. (ii) From the sixth century onwards new schools ^ We do not mean that Brahmanism descends to the adoration of sensible nature, above which, on the contrary it claims to rise completely. By the term " naturalism " we here mean the claim to arrive at union with God and perfection without the supernatural assistance of grace. 33
  • 40. AN INTRODUCTION TO PHILOSOPHY — . arose in India, some orthodox, others heterodox. Of these the principal was that founded by Cakya-Muni, surnamed the Buddha ^ (the enlightened, the sage) Buddhism, a doctrine essentially negative and solvent, directed, moreover, to practice rather than to specu-lation, may be regarded as the corruption and dis-solution of the Brahman philosophy. Substituting for that which is that which passes away, refusing to say that anything does or does not exist, and admitting only a succession of impermanent forms without fixed foundation or absolute principle in other words subordinating being to what is known as becoming or feri—^it showed, at the very time at which in Greece Heraclitus formulated the philosophy of flux, all the characteristics of a perfect evolutionary system, and, if it declared the existence of God, as of a substantial self and an immortal soul, unknow-able {agnosticism), its real tendency was to deny the existence of God (atheism), and to substitute for substance of any kind a stream or flux, regarded indeed ^ as itself real, of forms or phenomena (phenomenalism).^ Hence for Buddhism metempsy-chosis consists in a continuous chain of thoughts and feeHngs (a stream of consciousness, as we should term it to-day) passing from one mode of existence to another in virtue of a sort of urge towards Ufe, due itself to the desire to live : it is desire which is the " His actual name was Gautama. The name Cakya-Muni means the ascetic or hermit {muni) of the race or clan of the Cakya. Buddha lived during the second half of the sixth century B.C. He would seem to have died about the year 477. 2 At least by Buddha's original disciples. ' " Ever>'thing is empty, everything vmsubstantial " was a saying of Buddha's. 34
  • 41. PRE-PHILOSOPHIC THOUGHT cause of existence and " we are what we have thought." At the same time, the teaching of deliverance from suffering, which in Buddhism, even more than in Brahmanism, dominates the entire system, assumes a different and even more radical form. Evil is no longer merely the possession of individual or personal existence ; it is existence itself : it is evil to be, and the desire of existence is the root of all suffering. The wise man must therefore destroy in himself man's natural longing for existence and for beatitude, the fullness of being ; he must abandon all hope and extinguish every desire. He will thus attain the state of emptiness or toial indetermination called nirvana (literally nakedness, metaphorically immortality^ refresh-ment, thefarther bank—the term, in itself indefinite, was never defined by Buddha), which will deliver him from tlie evil of existence and the yoke of trans-migration, and which, in the logical consequence of Buddhist principles, must be regarded as the annihila-tion of the soul itself. For since the soul is only the chain or current of thoughts and feelings which derive their existence from the desire to be, to extinguish that desire is to extinguish the soul. This nirvana is the goal for whose attainment Buddhism made use of the ascetic practices which it took over with considerable mitigation from Brahman-ism, also of its moral code ^ which is thus directed, not to God, but to a species of mystical nothingness * We here tmderstand moral code in a very wide sense as meaning a code of behaviour. If the expression be taken as implying moral obligation, whose ultimate basis is the Christian doctrine of God the transcendent Creator, we must conclude that Buddhism, as indeed all the Oriental religions, Indian or Chinese, has no moral code. 35
  • 42. AN INTRODUCTION TO PHILOSOPHY as its last end. Moreover, the source and ultimate measure of Buddhist ethics is man, not God. If it rejected the system of castes which exaggerated the demands of social order and divided man almost into distinct species, it was only to dissolve social order of any kind in an absolute equality and individualism. And though it prescribed a universal benevolence (which extended even to prohibiting the slaughter of animals and to a compulsory vegetarianism), almsgiving, pardon of injuries, and non-resistance to the wicked, its motive was not love of one's neighbour as such, whose positive good and (by imphcation) existence we are bound to will, but to escape suffering to oneself by extinguishing all action and energy in a kind of humanitarian ecstasy. Buddhism is, therefore, a proof that gentleness and pity, when they are not regulated by reason and dictated by love, can deform human nature as much as violence, since they are then manifestations of cowardice, not of charity. This doctrine of despair is not only a heresy from the point of view of Brahmanism ; it is an intellectual plague to humanity, because it proceeds from the negation of reason. It is not, therefore, surprising that we find in it the majority of the fundamental errors by which contemporary attacks on reason are inspired. If at the present day it has found a warm welcome among certain circles in Europe, it is because all those who hope to derive from humanitarianism a moral code of human kindness for the acceptance of an atheistic society are already implicitly Buddhists. (iii) Buddhism is a philosophy, agnostic and atheistic, which nevertheless usurps the social and 36
  • 43. PRE-PHILOSOPHIG THOUGHT ritual functions of a religion. It is as a religion that it has won the allegiance of so many millions.^ In certain other schools to which Brahmanism gave birth—schools recognised as " orthodox "—we find, on the other hand, a tendency towards the normal distinction between philosophy and religion. These darshanas, it is true, would seem to be not so much distinct systems as complementary aspects of one and the same doctrine, the Brahmanist meta-physics. Here we may pass over the Vedanta, the most complete statement of that metaphysics and its doctrine of deliverance ; the Mimamsa, a species of commentary on the ritual and an explanation of the unseen forces set in motion by every act ; the Sankhya, founded, it is said, by Kapila (fifth or sixth century B,c. ?), which treats of the emanation of all things from their source, and seems to have taught, like Plato, a psychological dualism which explziins suffering by the union souls contract with matter ; and also Toga^ which teaches the practical methods which lead to contemplation, that is to say, the total loss of conscious-ness and identification with the universal Being (Iskvara) by a supra-rational knowledge. But the darshana Vaisesika, ascribed to Kanada (about the fourth century B.C. ?), which includes a rough outline of cosmology, and divides everything which exists into a number of fundamental classes or categories, sub-stance, quality, movement, association, difference, and ^ However, in proportion as it has secured wide acceptance, Buddhism has ceased to be atheistic, only to fall into the most degraded concep-tions of deity. Popular Buddhism as practised to-day in many parts of Asia, where, to adapt itself to existing beliefs, it has assumed the most varied shapes, is nothing more than a form of idolatry, totally different from philosophic Buddhism. 37
  • 44. AN INTRODUCTION TO PHILOSOPHY — inherence, and explains the four elements of ponder-able matter, earth, water, air, and fire, by the union of indivisible and indestructible particles, " atoms " in the language of philosophy,^ and the darshana Nyaya^ founded by Gotama, which attempts to construct a theory of reasoning and proof—that is to say a logic, though a logic extremely confused and incomplete are clearly the rough sketches of a work strictly and solely philosophical. But these crude attempts did not lead to a completed system, and Indian thought never achieved a rational and autonomous philosophy. {c) When we turn to the Far East and consider the very ancient civilisation of China^^ we find that when the primitive religion of the Chinese, which seems to have been fairly pure,' had from the twelfth century b.c. undergone gross corruption and materialisation, substituting the sky for God,* worshipping the sun and moon, paying divine worship to the souls of ancestors and to spirits, and allowing itself to become tainted by magic and sorcery, wise men were compelled here 1 Kanada, however, to explain this union, attributed real qualities to his atoms. Observe that Brahmani^m, which rejects atomism, admits five elements (ether being the fifth) ; Buddhism on the contrary, which has welcomed atomism, only four. 2 Whatever be the racial appurtenance of the Chinese, their history undoubtedly shows closer connections with the Aryans than with the Semites. It is for this reason that Chinese philosophy is discussed in the present section. * It taught the existence of one sole God — Shang-ti—personal, intelligent, distinct from the world, Sovereign Ruler of the races of mankind ; also the immateriality and immortality of the human soul, and even offered to the spirits of ancestors the same sacrifices and marks of reverence as to the good spirits who are guardians of men. * In all probability Heaven {Tien) was in origin simply a meta-phorical synonym of the Sovereign Ruler {Skang-H). 38
  • 45. PRE-PHILOSOPHIC THOUGHT also to seek a remedy for a decadence which about the sixth century B.C. threatened their civiHsation with utter ruin. It has long been beUeved that the Chinese sages were simply moralists wholly occupied with laying down rules of conduct and completely indifferent to metaphysical speculation. This is a true account only of Confucius and his followers ; it does not seem to be appHcable to Lao-Tse, though we can only accept with considerable reserve the interpretations of his teaching offered by certain modern Taoists. According to their account, Lao-Tse (born 604 b.c) was himself the disciple of a tradition whose oldest monument is the Ti-King, a book which consists essentially of sixty-four graphic symbols (hexagrams or double trigrams) arranged in a series of mechanical groups/ formed by combining simpler signs and susceptible of very many interpretations (metaphysical, logical, mathematical, moral, political, astronomic) , each number corresponding analogically with the others. The metaphysical speculation of the Ti-King appears to have been primarily concerned with the question, How can the Absolute, being wholly self-sufficient, act and manifest itself? It distinguishes in the supreme and sole First Principle or Perfection two different aspects, Chien^ the unmovirig and unknowable source of all activity, and Chuen, knowable activity, which eternally manifests perfection in a process of spiral evolution and an endless flux of forms. But these two aspects merge in one single and self-identical being, and all things, after passing through all the forms 1 Raymond Lull, in his attempts to create an ideographic algebra, employed an analogous procedure. 39
  • 46. AN INTRODUCTION TO PHILOSOPHY of evolution (of which the human cycle is but one curve), must return to Chien. This metaphysic may therefore be described as a species of evolutionary pantheism. It constitutes the foundation of Lao-Tse's system (Taoism), his chief contribution being an element of occultism and asceticism,^ Tao (the Way), the eternal goal and process of evolution, is the road by which all things must pass to arrive finally at the complete cessation of activity [Nibban, the Chinese nirvana), in which they are reabsorbed in nothingness and become one with the first principle of ail activity. The wise man will imitate the Tao by cutting himself off from all things, for the Way, though it has pro-duced beings, does not partake of their movements. " Having built this house, it dwelleth not therein." Detached from wealth, passions, and sensible experi-ence, and knowing that evil is mere appearance, he trains himself in solitude, secrecy, and humility (a humility which has nothing in common with the Christian virtue of that name, being nothing more than prudence and contempt for one's fellow men), until he reaches a state of perfect knowledge in which he no longer acts except by the pure intelligence. The wisdom, the illusory wisdom, to which Taoist asceticism leads its disciples, an asceticism which makes use of opium, as Buddhist asceticism of hypnosis, is for man a principle of revolt, therefore the adept 1 It may be added that in the twelfth century a.d. Chu-Hi, who has been regarded, mistakenly it would seem, as a materialist, formulated, in the tradition of Lao-Tse, a system which in the Chinese system of education has become, practically speaking, the official philosophy. It explains the constitution of things by a dual principle {li and ki) which is not without resemblance to the duality of form and matter in Aristotle and the Alexandrians. 40
  • 47. PRE -PHILOSOPHIC THOUGHT must keep it a secret for himself and a narrow circle of initiates.* No doubt Confucius (Kung-fu-tse, 551-479 B.C.), who, unlike Lao-Tse, represents for the Chinese a moderate and practical wisdom (which, from its place in their system of education and its own active character, is generally accessible), preserved many truths of the primitive wisdom. He avoided, however, every ultimate question, and confined himself to an ethic purely human, social, earthly, and even commonplace. Opportunism, he observes, is the distinctive mark of the wise man. Every predeter-mined line of action, every preconception is a mistake. In all matters one should pursue a middle course, live unfettered by one fixed purpose, embrace no opinion with enthusiasm, reject nothing because it is antipathetic, do whatever seems best in the circum-stances of the moment and as the situation demands. Confucianism, a system intended for the multitude, ended in pure materialism. Taoism, which claimed to address a small circle, and which, if the interpreta-tion given above be correct, constitutes, together with Brahmanism, one of the most singular attempts ever made by man to attain, in that ignorance of love which seems an aboriginal characteristic of Oriental thought, a wisdom exclusively of the intellect, by which he could deify himself in metaphysics, has experienced in China alternate periods of triumph and persecution, and has organised, apparently ever since the opening centuries of our era, secret societies * " Empty their heads, and fill their bellies," was Lao-Tse's advice to a statesman ; " weaken their minds and strengthen their sinews. To teach the people is to ruin the State." 41
  • 48. AN INTRODUCTION TO PHILOSOPHY in which it has definitively taken refiige since the seventeenth century and in which it has degenerated into a philosophic and political occultism of the most pernicious type. This brief historical sketch has shown the important part in the hfe of humanity played by sages and their wisdom. All these nations, situated on the frontiers of darkness, and lacking a divine revelation of truth, were obliged, when their religions proved incapable of satisfying the needs of the individual soul or of society, to have recourse to the wisdom supplied by human reason. This wisdom, in the civilisations of which we have spoken hitherto, was never differentiated from religion, but, on the contrary, encroached on the domain of the latter and claimed to conduct men to their last end, until in India we actually find that Brahmanism successfully achieved that canonisation of metaphysics which threatened the Greco-Roman world in the reign of the neo-Platonic emperor, Julian the Apostate. It achieved also that transfusion into religion of a human philosophy attempted by Kantian metaphysics in the nineteenth century {Modernism) . It has also shown how this human wisdom has everywhere proved bankrupt, and how, even before philosophy took shape as an independent discipline, most of the great philosophic errors had been already formulated. From the very first, the most arduous problems tower hke mountains before the intellect of man ; the problem of evil, the problem of being, the problem of the becoming and flux of things. It is not, therefore, surprising that a reason liable to 42
  • 49. PRE-PHILOSOPHIG THOUGHT error the moment it transcended the elementary truths within the range of common sense, a reason still unstable and undisciplined, and therefore all the more ambitious, went astray from the outset and opened the history of metaphysics with the dualism of Zoroaster and the pessimism of the Hindu, the pantheism and idealism of the Brahmans, the atheistic evolutionism of Buddha, and the illusory wisdom of Lao-Tse. When it became more modest, it was only to fall into the ethical positivism of Confucius, re-nouncing all sublimity and even denying its own raison d'etre. Nor should it surprise us to find the same errors reappearing at a later stage, v/hen philo-sophy had been fully elaborated. Error, at whatever period ofhuman history it may arise, is due to a failure of man's reasoning power—is, so to speak, a return of its primitive weakness, and therefore of its very nature retrograde. A further fact, however, calls for remark here, a fact only too well established by this prehistory, so to term it, of philosophy : namely, that these funda-mental errors are not unsubstantial and insignificant dangers ; they may succeed, to the bane of those diseased cultures which they condem.n to sterility. Truth (in all matters which transcend the data of common sense) is not, as those are apt to beheve who have had the good fortune to be bom into a culture formed by it, given to man ready made, like a natural endowment. It is difficult to attain, and hard to keep, and only by a fortunate exception is it possessed uncontaminated by error and in the totahty of its various complementary aspects. We have therefore the most urgent cause to be grateful for the possession 43
  • 50. AN INTRODUCTION TO PHILOSOPHY of a revelation, by which God has given us from on high, besides a knowledge of supernatural truth inaccessible to reason, a sure and easy access to the essential elements of the same truth which, so far as it falls within the natural order, is indeed accessible to our speculation, but can be so easily missed by it. Those also have the strongest claim on our gratitude who from below, by the strenuous exercise of their reason and unaided by revelation, succeeded in bringing to light the principles and laying the perma-nent foundation of this natural truth, and in con-structing a true and progressive human wisdom, in other words a philosophy, which, when met later and raised by the truth revealed from heaven, would be incorporated into the fabric of a higher wisdom, theology, the wisdom of man deified by grace, wisdom in the highest sense of the term. How highly there-fore ought we to prize the sacred heritage of Greek thought ! In Greece^ alone in the ancient world, the wisdom of man found the right path, and as the result of a fortunate harmony of the soul's powers and of a long effort to achieve mental order and discipline human reason attained its full vigour and maturity. In con-sequence, the small Hellenic race appears among the great empires of the East like a man amidst gigantic children, and may be truly termed the organ of the reason and word of man as the Jewish people was the organ of the revelation and word of God. It was in Greece alone that philosophy achieved her autonomy and was explicitly distinguished from religion. At least during the purest and most glorious 44
  • 51. PR E-PHILOSOPHIG THOUGHT age of the Heilenic mind, it recognised its own boundaries and was content to claim a strictly limited territory—the scientific study of purely rational truths —whereas Greek religion, already very much degraded in the time of Homer, became increasingly incapable of satisfying the needs of the intelligence, and grew more corrupt every day. True, the time would come when the Greeks, arrogantly abusing philosophy and reason, would attempt to embrace the things of God within the limits of their wisdom, " would become vain in their thoughts " and deserve the condemnation pronounced by St. Paul on the wisdom of this world, " which is foolishness in the sight of God." But their philosophy, though born of their mind, is undefiled by their corruptions, and its sole object is the truth. 45
  • 52. II THE PRE-SOGRATIG PHILOSOPHERS The earliest thinkers of Helias were the poets, the interpreters of traditional religion. Myth-makers, like Hesiod or Homer, sometimes prophets, such as that Epimenides of Cnossos who purified Athens from pestilence by erecting altars to unnamed divinities, they have no place in the history of philosophy in the strict sense, Greek philosophy, as Aristotle shows, only began with Thales of Miletus, one of the Sages or Gnomics, who lived in the seventh or sixth cen-turies B.G.^ The primary aim of these Sages, traditionally seven in number (their names are variously handed down by ancient writers), was to improve the conduct of their fellow citizens. Their aphorisms, some of which Plato quotes in the Protagoras, do no more than embody the practical lessons they had learned from their experience of life. They were men of acdon, legisla-tors, or moralists, men of prudence, but not yet philosophers. Alone among them Thales embarked on scientific speculadon. Geometrician and astrono-mer, he demonstrated that ail the angles inscribed in a semicircle are right angles, and appears to have predicted—no doubt owing to his acquaintance with ^ For the iragitnents from the early philosophers quoted in this chapter Professor Burnet's translation (Early Greek Philosophy) has been used I for Aristotle's Metaphysics Professor W. D. Ross's translation. 46
  • 53. . THE PRE-SOCRATIC PHILOSOPHERS Babylonian science—the total eclipse of the sun which occurred on May 28th, 585. The philosophers who succeeded him were still for the most part men who played an active part in public affairs, ardent politicians of the city state ; but, in spite of this practical activity, they were more or less clearly conscious from the beginning of the true nature of their wisdom. Moreover, save in the case of a few exceptional individuals (for instance, Empedocles, the miracle worker, and Pythagoras, who founded a religious sect) , Greek philosophy was from the very first distinct from religion—indeed it took shape as a critic and foe of the popular mythology and was manifestly the product of pure reasoning. In this work we are concerned only with the progressive development of Greek philosophy from Thales to Aristotle, for it was during this period that philosophy, with its absolute validity for mankind as a whole, took definite shape. The process occupied some three centuries and is divisible into three great epochs—the period of formation (the pre-Socratic philosophers), the period of crisis (the Sophists and Socrates), the period of fruitful maturity (Plato and Aristotle) THE lONIANS (a) Human reason now set out with its unaided powers in search of the principles and causes of things. What first strikes man's intelHgence is what he sees and touches, what he knows by his senses, and when he attempts to understand anything, he begins by asking what it is made of. Therefore the first thinkers of Hellas only considered in things the material of which 47
  • 54. AN INTRODUCTION TO PHILOSOPHY " they are made, their matter (what we shall learn to call the material cause), which they naively took to be a complete explanation of the object. Moreover, since the most universal and most important phenomenon of nature is change, especially the change by which one body becomes another {e.g. bread becomes flesh, wood fire), they concluded that the original matter of which all things are fashioned must be identical in all, the common subject of all corporeal changes. But since they were still unable to conceive any impalpable or invisible principle, they thought they had discovered this matter in some one of the , elements perceived by the senses. Thales, for example (624-546), influenced by traditional myths which derived all things from the primordial waters, and arguing from the fact that plants and animals " are nourished by moisture and that the germ of animal life is moist, concluded that water is the sole substance, preserving its identity through all the transformations of bodies. For Anaximenes (588-524) this substance was air, for Herachtus (540-475 ?) fire, for Anaximander (610- 547) the boundless (by which he understood the indeterminate, arceipov), a fusion of all the contraries. Moreover, water, air, fire, and the boundless were regarded as something active, Uving, and animate, endowed by an internal force with a manifold and unlimited fecundity. This was the meaning of Thales's dictum, all things " are full of gods," Travxa nkripy] 6eoiv.^ From the history of this extremely primitive Ionian school, whose philosophy is termed hylozoist because it ascribed life (Ccoy)) to matter (uXv)), we learn 1 Aristotle, De Anima, i, 5, 41 1 a 7. 48
  • 55. THE PRE-SOCRATIC PHILOSOPHERS to regard as the most elementary and crude of philosophic doctrines the materialistic monism which teaches the existence of a one single substance of a material nature, and evolutionism^ which attempts to explain everything by an historic process of unfolding, development, or evolution of something pre-existent. Evolutionism, which, owing, on the one hand, to German metaphysics, on the other to Darwin and Spencer, became so popular in the nineteenth century, was already taught in Greece by the physicists of the sixth and fifth centuries b.c.^ Anaximander in particular taught the eternal evolution of worlds " which rise and set at long intervals," and held that animals sprang from the mud of the sea floor, clothed at first, as with a species of shell, in a prickly bark which they shed on dry land, ^ and that man arose from animals of another species,' having been originally formed within the bodies of fishes, where he developed, being ejected as soon as he had become sufficiently large to provide for himself.* Later Empedocles of Agrigentum (493-433 ?), whose speculation in other respects marks an advance on that of the lonians,^ explained the origin of Uving beings 1 In India about the same date Buddhism was formulating, as we have seen, the religion of evolutionism. * Plac. Philos., V, 19, i. Dox. 430, 15. ' Pseudo-Plut., Strom., frag. 2, Dox, 579, 17. * Plut., Sjmp., q, viii. 579, 17. " For a single corporeal substance Empedocles substituted four elements specifically different, the four which became later the four classical elements of ancient chemistry—earth, water, air, fire. His dominant interest was to discover the efficient cause of the evolution of things, which he believed to consist in the two great motive forces love and hate. Empedocles was not only a philosopher : he was also a magician, doctor, poet, orator, and statesman. Aristotle ascribes to him the invention of rhetoric. 49
  • 56. . AN INTRODUCTION TO PHILOSOPHY by the separate production of the individual organs and members, e.g. the head, eyes, arms, which were subsequently joined by chance in every possible combination, of which only those have survived which were fitted to Uve {cf. the Darwinian principle of the survival of the fittest) It is also worth remark that, before Democritus, Anaximander and Empedocles also sought, Uke the pseudo-scientific evolutionism of modem times, to explain all things mechanically, that is to say—as the result of a simple aggregation of material elements effected by local motion. (b) Among these physicists, as Aristotle termed them, or philosophers of sensible nature, must be reckoned three great thinkers, Heraclitus, Democritus, and Anaxagoras. (i) Heraclitus of Ephesus,^ a lonely and proud genius who despised the multitude and popular reHgion, drew heroically fi*om the thought of the Ionian philosophers its ultimate metaphysical presuppositions, and thereby fixed for aU succeeding ages one of the possible extremes of speculation and error. A particular reality perceived in things had taken hold of his intellect with such force that he became its hopeless slave. That reality was change or becoming. His vision was so fixed on the change which all things undergo that he declared that change alone is real. IlavTa pel, aU things are in flux ; and men are fools to trust in the stability of their false happiness, " when they are born, they wish to live and to meet their ^ The dates of Heraclitiis's birth and death are uncertain. He was in his a-Kix-fif the prime of his age, about 500 B.C. 50
  • 57. THE PRE-SOGRATIC PHILOSOPHERS doom—or rather to rest—and they leave children behind them to meet their doom in turn." We do not touch the same thing twice nor bathe twice in the same river. The very moment we touch an object, it has already ceased to be what it wa5 before. Whatever exists changes from the very fact of its existence. That is to say change has no abiding and permanent subject identical with itself, Uke an ivory billiard ball which remains an ivory billiard ball while it is in motion. We are therefore compelled to pronounce boldly that that which is (the thing which changes) at the same time is not (because there is nothing which persists throughout the change). " We step and do not step into the same river ; wc are and are not." Moreover, contraries must be pronounced identical. The sea is the purest and the impurest water. Good and ill are one. *' No one," writes Aristotle in a famous passage, " can possibly conceive that the same thing does and does not exist. According to some, Heraclitus was of a different opinion, but we are not obliged to believe that a man really thinks whatever he says. The reason of the opinion held by these philosophers was that the only realities they admitted were sensible objects, and, since they perceived that sensible nature is in perpetual motion, some have held with Cratylus ^ that no statement can be made about it ; he was content to wag his finger." * This scepticism was the inevitable consequence of Heraclitus's philosophy of pure flux, despite his personal conviction, * One of Heraclitus's most famous disciples. He was Plato's first teacher. (Aristotle, Metaph., i, 6.) * Metaph., iv, 5, loio a 13. 51
  • 58. AN INTRODUCTION TO PHILOSOPHY passionately held, of the reality and value of truih. " If you do not expect the unexpected," he said, " you will not find truth, for it is hard to be sought out and difficult." HeracUtus is thus the philosopher of evolution and becoming. In his view, all things are diflbrentiations produced by discord or strife {7z6'kz[io<^ Tcar^p Tcavrcov) of a single mobile principle which he conceives in the form of fire, a fire ethereal, living, and divine. So from the outset stands out in the clearest light that fatal necessity which chains every philosophy of pure becoming to monism ' or to pantheism^ " If," wrote Aristotle,' '' you maintain that all beings are one, you simply return to Heraclitus's opinion. All things are then confused, good and evil become identical, man and the horse are one and the same thing. But this is really to maintain not that beings are one, but that they are nothing." (ii) Born within a few years of Heraclitus's death, Democritus of Abdera (470-361 ?), who had a more superficial intellect and a predilection for ideas easily comprehensible, attempted to discover in the flux of sensible phenomena a permanent and unchanging element ; but in his search for this unchanging element he made use of his imagination rather than his understanding. Therefore the sole reality he would recognise was something which, though it is inaccessible to the senses, can nevertheless be appre-hended by the imagination—namely, pure geometrical quantity as such, stripped of all qualities (colourless, ^ The doctrine that ail things are one single being. ' The doctrine which identifies the world with God. * Phys., i, 2, 185 b 19. 52
  • 59. THE PRE-SOCRATIC PHILOSOPHERS scentless, tasteless, etc.), and possessed solely ofextension in the three dimensions of space. Democritus found the explanation of everything in the plenum, which he identified with being, and the void, identified with nonentity. The plenum was divided into indivisible parts of extension (" atoms "), which were separated one from another by the void and in a state of ever-lasting motion, and differed only in shape,^ order,' and position.' The order of the universe and the structure of individual beings he attributed to the blind necessity of chance. Thus Democritus * intro-duced into Greek philosophy during the lifetime of Socrates the doctrine of atomism and more generally the philosophy termed mechanical, which raises geometry to the position of metaphysics, reduces everything to extension and motion, and professes to explain the organisation of the universe by a host of fortuitous coincidences. In this fashion the Parthenon could be " explained " as the result of throwing stones one on another during an indefinite term of ^ As, for example, A differs from N. » As AN differs from NA. ' As N differs from the same letter placed differently : Z. * As also his master Leucippus. Had Leucippus and Democritus come in anyway under the influence of the Indian philosopher Kanada? The more likely hypothesis is a coincidence due to similarity of intellec-tual oudook, particularly if Kanada, whose date is very uncertain, was contemporary with or even posterior to Democritus. Speaking generally, there seems no reason to believe that Oriental speculation so influenced Greek thought as to teach it in the strict sense or transmit any particular system. That, on the other hand, it influenced the Greeks by arousing a spirit of speculative inquiry and providing in-tellectual material (which they alone were able to treat scientifically) is the obvious conclusion from the simple fact that Greek philosophy originated in those provinces of the Hellenic world which were in contact with the Ea^t. 53
  • 60. AN INTRODUCTION TO PHILOSOPHY years, the tragedies of Racine as due to the in-discriminate shuffling of type for a sufficient length of time. (iii) Finally Anaxagoras of Glazomenae (500-428), who was a friend of Pericles, and in the maturity of his powers when Democritus was born and HeracUtus had just died, turned Greek philosophy towards a higher source of illumination, and cor-rected rather than continued the speculation of the lonians by the aid of ideas which, it must be con-fessed, he either worked out badly or was unable to use. On the one hand, he perceived that the material principle ofwhich all bodies are formed, and which the lonians had identified with one particular element, must already somehow contain in itself the entire diversity to which it will give birth : unless everything were in everything, nothing could come from nothing.^ He therefore concluded that the principle in question consisted of an endless mixture of all natures and quahties in such fashion that each corporeal particle contained within itself the elements {homoeomeries) of all the rest ; for example, each particle of the bread we eat contains invisible elements of the bone, blood, flesh, etc., which will be discovered later, changed only in their relative proportions, in each particle of bone, blood, flesh, etc. It was a bizarre conception, and, as taught by Anaxagorsis, not worth serious discussion, but nevertheless a crude adumbration of Aristotle's great conception of first matter {materia prima) which is nothing in act, but aU bodies in potentiality. 1 Cf. Aristotle, Phys., i, 4, 187 a 26. Simpliciiu, Phys,, 155, 23. 54
  • 61. — THE PRE-SOCRATIG PHILOSOPHERS On the other hand—and it is his chief claim to distinction—he realised that the material principle, that of which all things are made, is insufficient to explain them. We must also discover the agent that produces them (the efficient or motor cause) and the end for which the agent acts (the final cause). Is it, as Plato was to ask later, a sufficient explana-tion of the fact that Socrates is sitting in prison to say that he has bone, joints, and muscles arranged in a particular fashion? We must also know who brought about that disposition of these bones and muscles—namely, Socrates himself by his will—and why he willed it. Because Anaxagoras arrived at the recognition that there must necessarily exist, besides the material elements of the world, a separate Intelligence (voui;') to which the ordering of the universe is due, he alone, as Aristotle remarks " kept sober " when all the other philosophers of his period, drunk with the wine of sensible appearances, " spoke at random." ^ THE ITALIANS Besides the school of Ionian philosophy, there existed in the Hellenic world of the sixth and fifth centuries B.C. two other great philosophic schools the Pythagorean and the Eleatic. Pythagoras of Samos (572-500 ; according to other authorities, 582-497), the founder of a philosophic society of a religious and political character, which held the reins of government in several cities of Magna Graecia (Southern Italy), and was later dissolved by 1 Metaph., i, 3, 984 b 18. 55
  • 62. AN INTRODUCTION TO PHILOSOPHY violence,^ understood that there existed realities of a higher order than those perceptible by the senses. But it was by the study of numbers that he had arrived at the knowledge of these invisible reaUties, whose immutable order dominated and determined the process of becoming ; and henceforward he had understanding only for numbers. Not content with teaching that there is present in all objects and in the universe as a whole a hidden principle of measure and harmony, he taught that numbers—by which this harmony is revealed to our senses—are the sole true reality, and regarded them as the very essence of things. Pythagoras was not only conversant with the important observations of Oriental astronomy, but, by his fundamental discovery of the relationship between the pitch of sounds and the length of vibrating strings, had reduced to the rigidity of numerical law so fugitive a phenomenon as sound. Imagine the awed astonishment with which he must have dis-covered behind the flux of sensible phenomena the intelligible constant and immaterial proportions which explain to the mathematician the regularities we observe. Consider, moreover, the mysterious symbolic value of numbers attested aUke by the sacred traditions of mankind and the most positive of philosophers (from Aristotle, who paid homage to the holiness of the number 3, to Auguste Comte, who will construct an entire mythology of the prime numbers) , and it is easy to understand how naturally the thought of Pythagoras * In this society absolute obedience prevailed even in the speculative sphere. It was in the Pythagorean brotherhood, not in the schools of the Christian Middle Ages, that everything yielded to the Magister dixit, airrbs ^<pa 56
  • 63. THE PRE-SOCRATIC PHILOSOPHERS and his disciples passed from the sign to the cause and made the symbol a principle of reality. Consequently, numerical principles were regarded as the principles of everything that exists ; from the opposition between the determinate and the indeter-minate (infinite) are derived all the fundamental pairs of opposites—odd and even, the elements of number, the one and the many, right and left, male and female, rest and motion, straight and crooked, light and darkness, good and evil, the square and the quadrilateral with unequal sides—which determine the nature and activity of things. Every essence has its number and every essence is a number. The number 4, for example, is not simply a figure ofjustice, it constitutes the essence of justice ; similarly the number 3 constitutes holiness, the number 7 time, the number 8 harmony, the number 5 the union of the sexes, the number 10 perfection. When numbers which in themselves are not localised receive a position in space, bodies come into existence. Thus all speculation on the origin or nature of things resolves itself into speculation on the genesis and properties of numbers. Pythagoras, therefore, and his school, to whom mathematics, music and astronomy owe so much, never arrived at the true conception of the first philosophy or metaphysics. They achieved, it is true, a degree of abstraction superior to that at which the lonians had halted, and did not, like them, confuse metaphysics with physics. But they confused it with the science of number, into which, moreover, they imported qualitative interpretations ; and, con-sequently, in spite of their effort to reach the object 57
  • 64. AN INTRODUCTION TO PHILOSOPHY of pure intelligence, they were held fast in the bonds of imagination. And if, on the other hand, they perceived that the nature of things is intrinsically determined by immaterial principles more real and truer than that which is tangible and visible, they were not yet able to attciin the notion of the formal cause, whose full elucidation was reserved to Aristotle alone. It is to Pythagoras, as we have already remarked, that we owe the term, philosophy. A passage ofDiogenes Laertius (viii, 8) shows that for him the dignity of science consisted in its purely speculative and dis-interested character, a point on which Aristotle, at the beginning of his Metaphysics, was to insist strongly. " Human life," he said, " may be compared to the pubhc games, which attract diverse sorts of men. Some come to compete for honours and the crowns of victory, others to trade, others, the more noble sort, solely for the enjoyment of the spectacle. Similarly in Ufe some work for honour, others for profit, a few for truth alone; they are the philosophers. ." . . Pythagoras appears to have taught the unity of God, whom he regarded as one omnipresent Spirit from whom our spirits emanated. He was the first to give the universe the name y.6a[ioc^, which, like the Latin mundus, conveys the idea of beauty and harmony. The most famous and the most derided of his tenets was the doctrine of the transmigration of souls, or metempsychosis, which he probably derived not from Egypt, as Herodotus suggests, but from Hinduism (by way of Persia) ^ a doctrine which very early ; ^ As Gomperz observes, " the Asiatic Greeks and a portion of the population of India were already subject, when Pythagoras left his 58
  • 65. THE PRE-SOCRATIC PHILOSOPHERS obtained in Greece the adhesion of the Orphics and the Pythagoreans. " Coming one day upon a puppy which was being cruelly beaten," the aged Xenophanes wrote of Pythagoras in mordant verse, " he lamented its fate and cried out in pity ' Stop Don't beat him. ! That is the soul of one of my friends him ; I recognise by " his voice.* The Pythagoreans also believed that the revolution of cosmic cycles must produce the everlasting recur-rence at enormous intervals of all things, reproduced identically even in the most insignificant details. " According to the Pythagoreans," Eudemus told his disciples, " a day will come when you will be all gathered again, sitting in the very same places to listen, and I shall be telUng you the same story once more." ^ Astronomy was among the sciences which the Pythagorean school cultivated most successfully. Philo-laus, who taught that the earth, the sun, and all the stars revolved around a mysterious centre of the universe filled with fire, may be regarded as a distant precursor of Copernicus. But even in this sphere the Pythagoreans betrayed in the most flagrant fashion the vices of the exclusively mathematical mind. native Ionia, to the same ruler, Cyrus, the founder of the Persian Empire " {Thinkers of Greece, i, 3). Speaking more generally, it was, it would apf>ear, by way ol the Pythagorean school that certain distinctively Oriental conceptions and modes of thought first entered Greece, to pass from Pythagoreanism to Platonism and neo-Platonism, and thence, swollen by further additions, into Gnosticism and the more or less undergroimd stream of heterodox speculation. 1 Simplicius, Phys., 732, 30 D. Nietzsche, who was obsessed and driven to despair by the thought " of the everlasting recurrence of things," derived this singular conception from Greek philosophy. 59
  • 66. AN INTRODUCTION TO PHILOSOPHY " The Pythagoreans," wrote Aristotle/ " having been brought up in the study of mathematics . . . suppose the whole heaven to be a musical scale and a number. All the properties of numbers and scales which they could show to agree with the attributes and parts, and the whole arrangement of the heavens, they col-lected and fitted into their scheme ; if there was a gap anywhere, they readily made additions so as to make their whole theory coherent. For example, as the number i o is thought to be perfect and to comprise the whole nature of numbers, they say that the bodies which move through the heavens are ten ; but as the visible bodies are only nine, to meet this they invent a tenth—the counter-earth—not studying the phenomena to discover their causes and test their hypotheses, but imposing upon the phenomena their hypotheses and preconceived beliefs, thereby claiming to assist God to fashion the universe." THE ELEATICS Though it cannot, strictly speaking, be said that the school of Elea founded metaphysics, since it failed to keep a firm grasp of the truth, it must receive the credit of having raised Greek thought to the meta-physical level and attained the necessary degree of abstraction. The oldest of the Eleatics was Xeno-phanes, a wandering rhapsodist, born about the year 570 at Colophon, whence he migrated to Elea in southern Italy—banished, no doubt, by the Persian invasions. Xenophanes poured scorn upon the mytho-logy of the poets and the opinions of the common 1 Melaph., , 5, 986 a. De Caelo, ii, 13, 293 a. 60
  • 67. THE PRE-SOGRATIG PHILOSOPHERS people. " Far better," he said, in slighting reference to the honours paid to athletes, " is our art than the strength of men and horses." He taught the absolute unity of God, but confounded him with the universe, declaring in a pantheistic sense that God is one and all, ev xal Tcav. But the most profound thinker, indeed the true founder of this school, was his disciple Parmenides of Elea (born 540), the Great Parmenides, as Plato called him. Transcending the w^orld of sensible phenomena and even that of mathematical forms or essences and numbers, he attained to that in things which is purely and strictly the object of the intellect. For it can scarcely be denied that the first truth about things which the intellect perceives is that they exist, their being. The notion of being, thus abstracted, impressed Parmenides so powerfully that it fascinated him. As his contemporary Heraciitus was the slave of change, Parmenides was the slave of being. He had eyes for one thing alone : what is is, and cannot not be ; being is, non-being is not. Parmenides was thus the first philosopher who abstracted and formu-lated the principle of identity or non-contradiction, the first principle of all thought. And as he contemplated pure being, he perceived that this being is completely one, absolute, immutable, eternal, without becoming, incorruptible, indivisible, whole and entire in its unity, in everything equal to itself, infinite ^ and containing in itself every perfec-tion.- But while he thus discovered the attributes of him who is, he refused to admit that any other being ^ Simpiicius, Phys., 144, 25-145, 23. (Diels, frag. 8, 22.) ' Aristotle, Phys., i, 3. 61
  • 68. AN INTRODUCTION TO PHILOSOPHY could exist, and rejected as a scandal to the reason the being mingled with non-entity, because produced from nothingness, of every creature. He was thus led so far astray that he ascribed to the being of the world that which belongs only to uncreated being. And rather than be false to what he believed were the exigencies of being and reason, he preferred to refuse heroically the witness of the senses and deny the existence in the universe of change or multiplicity. Change, motion, becoming, as also the diversity of things, are but an illusory appearance. There exists only being, the one. Does not change imply that an object both was and was not (what it becomes), and at the same time continues and ceases to be (what it was) ? Does not multiplicity imply that what is (this) is not (that) ? Do not, therefore, multiphcity and change contradict the fundamental principle that what is possesses in itself being and not non-being ? It was in defence of Parmenides's doctrine of the impossibility of change that his disciple Zeno of Elea ^ (born 487) composed his famous arguments, by which he claimed to prove that the very concept ofmovement is self-contradictory : arguments fallacious, no doubt, but of singular force and refutable only by the doctrine of Aristotle. Thus Parmenides, reaching the opposite pole to Heraclitus, fixed, as he did once for all one of the extreme hmits of speculation and error, and proved that every philosophy of pure being, for the very reason that it denies that kind of non-being which ^ Not to be confused with Zeno the Stoic, who lived much later (350-264) and was born at Cittium in Cyprus. 62
  • 69. THE PRE-SOGRATIC PHILOSOPHERS Aristotle termed potentiality and which necessarily belongs to everything created, is obliged to absorb all being in absolute being, and leads therefore to monism or pantheism no less inevitably than the philosophy of pure becoming. 63
  • 70. Ill THE SOPHISTS AND SOCRATES The long effort of these speculative pioneers which we have briefly recapitulated had equipped human thought with a number of fundamental truths. But if, looking backwards with a knowledge of the mighty synthesis in which all those truths, then partially per-ceived, have been harmonised and balanced, we can contemplate with admiration the gradual formation of the vital centres and arteries of philosophy, at the time, in fifth-century Hellas, these good results were concealed not only by the medley of con-tradictory theories, but by the number and gravity of prevalent errors, and it seemed as though the entire movement had achieved nothing but disorder and chaos. The Greek thinkers had set out with high hopes of knowing everything, and climbing the sky of wisdom in a single step. As a result of this immoderate ambition, and because they lacked discipline and restraint in handling ideas, their concepts were embroiled in a confused strife, an interminable battle of opposing probabilities. The immediate and obvious result of these attempts at philosophising seemed the bankruptcy of speculative thought. It is not, therefore, surprising that this period of elaboration produced a crisis in the history of thought, at which an intellectual disease imperilled the very existence of philosophic 64
  • 71. THE SOPHISTS AND SOCRATES speculation. This intellectual disease was sophistry, that is to say, the corruption of philosophy. THE SOPHISTS Sophistry is not a system of ideas, but a vicious attitude of the mind. Superficially the sophists were the successors and disciples of the thinkers of an earlier generation—even the word sophist originally bore no derogatory significance—in reaHty they differed fi-om them completely. For the aim and rule of their knowledge was no longer that which is, that is to say, the object of knowledge, but the interest of the knowing subject. At once wandering professors accumulating honours and wealth, lecturers, teachers of every branch of learning, journalists, if one may so call them, supermen, or dilettanti, the sophists were anything in the world but wise men. Hippias, who achieved equal eminence in astronomy, geometry, arithmetic, phonetics, pro-sody, music, painting, ethnology, mnemotechnics, epic poetry, tragedy, epigram, dithyramb, and moral exhortation, ambassador of EUs, and jack-of-all-trades (he attended the Olympic Games in clothes made entirely by himself), reminds us of some hero of the ItaUan Renaissance. Others resemble the philosophes of the eighteenth or the " scientists " of the nineteenth century. But the most characteristic feature of all aUke was that they sought the advantages conferred by knowledge without seeking truth. They sought the advantages conferred by knowledge so far as knowledge brings its possessor power, pre-eminence, or intellectual pleasure. With this in view, 65
  • 72. AN INTRODUCTION TO PHILOSOPHY they put themselves forward as rationalists and walking encyclopaedias ; to every question they had an answer ready, deceptively convincing ; ^ and they claimed to reform everything, even the rules of grammar and the gender of nouns. ^ But their favourite study was man, of all the most complex and uncertain, but one in which knowledge is easiest coined into power and reputation ; and they culti-vated most assiduously law, history, casuistry,' politics, and rhetoric. They professed to be teachers of virtue. They did not seek truth. Since the sole aim of their intellectual activity was to convince themselves and others of their own superiority, they inevitably came to consider as the most desirable form of know-ledge the art of refuting and disproving by skiliul arguments, for with men and children alike destruction is the easiest method of displaying their strength, and the art of arguing with equal probability the pros and cons of every question—another proof of acumen and skill. That is to say, in their hands knowledge altogether lost sight of its true purpose, and what with their predecessors was simply a lack of intellectual discipline became with them the deliberate intention to employ concepts without the least regard for that delicate precision which they demand, but for the pure ^ Critias, for example, considered belief in the gods as the invention of an astute statesman who sought to keep the citizens obedient by clothing the truth in a garment of fiction. 2 It was Protagoras who attempted to rationalise the genders of nouns : desiring, for example, that nijvis (anger) should be given the masculine gender, also ttt^Xt?^ (helmet), etc. * Recall, for example, the celebrated discussion between Protagoras and Pericles after an accidental homicide in the course of an athletic contest, on the question who ought to be punished : the man who arranged the contest, the unskilful player, or the javelin itself. 66
  • 73. THE SOPHISTS AND SOCRATES pleasure of playing them off one against the other — an intellectual game of conceptual counters devoid of solid significance. Hence their sophisms or quibbles. Their ethics were of a piece. Every law imposed upon man they declared to be an arbitrary convention, and the virtue they taught was in the last resort either the art of success, or what our modern Nietzscheans call the will to power. Thus, of the spirit which had inspired the lofty intellectual ambitions of the preceding age, the sophists retained only the pride of knowledge ; the love of truth they had lost. More ardently than their pre-decessors they desired to achieve greatness through knowledge, but they no longer sought reality. If we may use the expression, they believed in knowledge without believing in truth. A similar phenomenon has recurred since in the history of thought and on a far larger scale. Under these conditions the sole conclusion which sophism could reach was what is termed relativism or scepticism. Protagoras of Abdera (480-410), for ex-ample, maintained that " man is the measure of all things—of what is, that it is, and of what is not, that it is not," by which he meant that everything is relative to the dispositions of the subject and the truth is what appears true to the individual. And his contemporary, Gorgias of Leontini (died 375), a famous orator, in his book entitled Of Mature or the Non- Existent, taught (i) that being is not, in other words, that nothing exists : non-existence is non-existence, and therefore it is—a quibble on the word is which one day would be Hegel's grand metaphysical game hence being, its contrary, is not ; (ii) that if anything 67
  • 74. AN INTRODUCTION TO PHILOSOPHY existed we could not know it ; (iii) that if anyone could know anything, he could not communicate his knowledge to another. SOCRATES It was Socrates (469-399) who saved Greek thought from the mortal danger into which the sophists had brought it. Except for the fact that he took no fees for his instruction, his manner of Ufe was externally the same as theirs. Like them, he spent his time in discussions with young men, and it is not surprising that a superficial observer, such as Aristophanes, confused him with the sophists. In reality, he waged against them an unrelenting war and opposed them at every point. The sophists claimed to know everything and did not believe in truth ; Socrates pro-fessed ignorance and taught his hearers to seek nothing but the truth. Thus his entire work was a work of conversion. He reformed philosophic reasoning and directed it to the truth, which is its proper goal. This work was of such importance for the fiiture of the human intellect that it is not strange that Socrates accomplished it as a mission divinely imposed. He possessed not only an extraordinary power of philo-sophic contemplation (we are informed by Aulus Gellius and Plato that he sometimes passed entire days and nights motionless, absorbed in meditation) but also something which he himself regarded as daimonic or inspired, a winged fervour, a free but measured force, even perhaps at times an interior instinct of a higher order, gifts suggestive of that extraordinary assistance in regard to which Aristotle 68
  • 75. THE SOPHISTS AND SOCRATES said that those who are moved by a divine impulse need no guidance of human reason, since they possess in themselves a better principle.^ He compared him-self to a gadfly sent to sting the Athenians awake and force upon their reason a constant examination of conscience, a service which they repaid with hem-lock, thus affording the aged Socrates, already on the verge of the grave, opportunity for the most sublime death to which merely human wisdom can lead. (a) Socrates was not a metaphysician, but a prac-titioner, a physician of souls. His business was not to construct a system, but to make men think. This was the method by which he could best conquer a sophistry whose radical vice was not so much an error of doctrine as a deformity of the soul. For the chief topic of Socrates's discussions was the problem of the conduct of human Hfe, the moral problem. His ethics, as far as we can judge it by the reports of Plato and Xenophon, seems at first sight to have been dictated by narrowly utiUtarian motives. What I ought to do is what is good for me, and what is good for me is what is useful to me—really useful. But at once the need becomes evident of discovering what is really useful to man ; and at this point Socrates compelled his hearers to acknowledge that man's true utility can only be determined by reference to a good, absolute, and incorruptible. By thus constantly raising the question of man's last end,^ and directing men towards their sovereign good, he went beyond utilitarianism of every description, and, ^ Magn. Moral., vii, 8. Cf. Eth., vii, i. ' A question which he himself seems to have answered some-what ambiguously. 69
  • 76. AN INTRODUCTION TO PHILOSOPHY with the full force of a sane common sense, vindicated the supremacy of moral good and our great eternal interests ; his ethics thus passed over into the meta-physical sphere. In the second place Socrates proved by every method of argument that in order to behave rightly man's first requisite is knowledge ; he even went so far as to maintain that virtue is identical with knowledge and therefore that the sinner is simply an ignorant man. Whatever we think of this mistake, the fact remains that for Socrates ethics was nothing if not a collection of truths established by demonstra-tion, a real and a genuine science. In this two-fold character, metaphysical and scientific, of his moral teaching, he stood in radical opposition to the sophists, and may be regarded as the founder of ethics, {b) But it was impossible to found scientific ethics without defining at the same time the laws which determine scientific knowledge of every description. Here we reach the essence of the Socratic reform. By returning to reason itself to study the conditions and value of its progress towards truth, that is to say, by the use of logical and critical reflection, Socrates disciplined the philosophic intelligence, showed it the attitude to adopt and the methods to employ in order to attain truth. The first requisite was to cleanse the mind of the false knowledge which pretends to get to the bottom of things with a few facile ideas. That is why he always began by leading those whom he entangled in the net of his questions to confess their ignorance of that which they had been certain they knew best (Socratic irony). But this was merely the preliminary stage of his method. Soon the questioning began 70
  • 77. THE SOPHISTS AND SOCRATES afresh, but its object now was to lead Socrates's inter-locutor, whose thought it guided in the direction desired, to discover for himself the truth of which he had admitted his ignorance. This was the essence of the Socratic method, his maieutic, the art of intel-lectual midwifery. Moreover, Socrates reaUsed so thoroughly that the attainment of truth is a vital and personal activity, in which the teacher can only assist his pupil's intelUgence, as a doctor assists nature, but the latter is the principal agent,^ that he compared the acquisition of knowledge to the awakening of a memory dormant in the soul, a comparison from which Plato was to derive his famous theory of reminiscence (dva[xvif)(Ti.(;'). How, then, did this maieutic form the philosophic intellect ? By determining its proper object, teaching it to seek the essences and definitions of things. ^ Socrates was never weary of recalling reason to this one object : that which the subject of discussion is, what is courage, piety, virtue, the art of ship-building or cobbling, etc. All these have a being peculiar to themselves, an essence or nature which the human understanding can discover and express by a definition which dis-tinguishes it from everything else. Because Socrates thus required that the essential should in all cases be distinguished from the accidental, and because he persistently employed his intellect in the search for essences, his philosophy may be termed the philosophy of essences. It was no longer a question of reducing everything to water, fire, numbers, or even absolute being, nor yet of finding some indeterminate concept 1 Cf. St. Thomas, Sum. Theol., i, q. 117, a. i. 2 Cf. Aristotle, Metaph., xi, 4, 1078 b 17-3^'. 71
  • 78. Am INTRODUCTION TO PHILOSOPHY sufficiently elastic to enfold everything like a shapeless mantle. On the contrary, Socrates's aim was to attain the proper intellectual expression of everything—to define and determine its essence by a concept applicable only to itself. At the same time Socrates taught the reason, if not by a finished theory of ratiocination, or by construct-ing, as Aristotle would construct later, a logic of syllogism and demonstration, at least practically and of set purpose, to employ concepts, not, as in the barbarous word-play of the sophists, as weapons to deliver strokes at haphazard, but in such a fashion that they fitted exactly the outhne and structure of reahty. He thus created dialectic, an instrument of knowledge, as yet no doubt defective, but which paved the way for the correct notion of scientific knowledge, and was compared by Plato to the art of the expert chef who cuts up a fowl by distinguishing and following carefully the smallest joints of its anatomy. (c) Thus this unwearied disputer, for all his superficial scepticism, possessed an invincible con-fidence in the intellect and in science—but of an intellect disciplined, humble in its attitude towards reality, and a science aware of its limitations, advancing successfiilly and securely in the apprehension of truth only so far as it respected the sovereignty of the real and was conscious of its ignorance in every direction. In this we recognise Socrates as the teacher of the scientific spirit, as also of the philosophy which we shall learn to know as moderate intellectualism. By his logical and critical work he forged the instrument indispensable for the progress of the mind and turned 72
  • 79. THE SOPHISTS AND SOCRATES the crisis created by sophistry to the profit and salva-tion of reason. By his work as a teacher of morality, he not only founded the science of ethics, but Uberated thought from the fascination of the sensible, and unintentionally, perhaps, set philosophic speculation on the road to metaphysics, wisdom in the strict sense. This he did simply by raising philosophy (this was the true significance of the Socratic demand for self-knowledge) from exclusive occupation with the physical universe ^ to the study of human nature and human activities, which contain a spiritual element of a higher order altogether than Uie stars or the entire universe of matter. But Socrates was no more than a pioneer of genius. He gave the impulse, but never reached the goal. When he died everything was still in the air. For method is not enough, a systematic body of doctrine is necessary ; and Socrates, though his teaching was fertile in fruitful hints, possessed, apart from the elements of ethics, no doctrine in the strict sense. The doctrinal completion of his work and the construction of the true philosophy were reserved for Plato and Aristotle. 1 Parmenides himself arrived at the metaphysical conception of being by an exclusive consideration of the corporeal imiverse. 73
  • 80. IV PLATO AND ARISTOTLE So undoctrinal was Socrates's teaching that his disciples developed it along very divergent lines. The philosophers known as the minor Socratics, who seized upon some partial aspect of the Master's thought, which they distorted more or less, were either moralists pure and simple (like the Cyrenaics,^ who placed man's last end in the pleasure of the moment, and the Cynics,^ who, going to the opposite extreme, deified force of character or virtue), or logicians infatuated by the love of argument (eristics) , like the neo-sophists of Elis and especially of the school of Megara,' who tended to deny the possibility of knowledge and, by compelling philosophers to find answers to their argu-ments, indirectly contributed to the progress of logic. The Megarians denied that in any judgment one thing could be predicated of another. According to them, this amounted to affirming that the predicate was the subject and thus everything became identical with 1 The leading philosophers of this school were Aristippus of Gyrene, Theodore the Atheist, Hegesias, and Anniceris. 2 The name was derived from the gymnasium in Athens (Kwdaapye^) where Antisthenes taught. The chief Cynics were Antisthenes (born 445 B.C.), Diogenes of Sinope (400-323), and Crates of Thebes. * The principal representatives of the school of Elis were Phaedo and Menedemus, of the Megarian, Euclid of Megara (not to be confused with Euclid the geometrician), Eubulides of Miletus, Diodorus Cronos, and Stilpo. 74
  • 81. PLATO AND ARISTOTLE everything else. Hence being exists is the sole legitimate proposition, and the Eleatic metaphysics the only true philosophy. PLATO The appellation major Socratics belongs only to Plato, Socrates's intellectual heir, and his disciples. Plato (427-347),^ whose father was of royal descent and whose mother traced her pedigree to Solon, ambitious to reign as king in the intellectual domain, endeavoured to combine in the powerful unity of an original system the entire host of speculations which he found scattered and fragmentary in the conflicting systems of his philosophic predecessors. With him philosophy attained her majority. But the work which he attempted and which the Socratic reform had made possible remained incomplete and defective. Under the impulse of his lofty and daring genius, the intellect soared too fast and too high, and failed to achieve by a final victory the conquest of reaHty. (fl) Like Parmenides, Plato understood that the subject of metaphysics is being itself But he refused to absorb everything which exists in the unity of immutable and absolute being. And he was thus led to the discovery of important metaphysical truths. He perceived that, since things are more or less perfect, more or less beautiful and good, more or less deserving of love, and since there are things whose goodness is mixed with evil—which in philosophic terminology participate in goodness—there must neces- 1 After extensive travels, Plato settled in Athens, v^'here he purchased the estate of a certain Academus to be the home of his school, known therefore as the Academy. 75
  • 82. AN INTRODUCTION TO PHILOSOPHY sarily exist a being in which goodness, beauty, and perfection are full and entire, unmingled with their contraries, a being which is the ground of the beauty and goodness of everything else. His thought thus ascended to the true God transcending the world and distinct from it, whom he saw as goodness itself, the absolute good, the good, so to speak, in person. But this was not the most prominent aspect of Platon-ism. As we pointed out above, Socrates's philosophy —a philosophy rather suggested by his practice than formulated theoretically—was the philosophy of essences ; Plato's philosophy was, before everything else, the philosophy of ideas. Socrates had shown that what we must seek and attain at all cost are the essences of things which the mind apprehends and expresses in a definition. What, then, is it wliich the intellect perceives when it appre-hends the essence of a man, a triangle, white, or virtue ? Clearly man, abstracting from Peter, Paul, John, etc., triangle, abstracting from any particular triangle, isosceles or equilateral, and similarly white and virtue. Moreover, the concept or idea of man or triangle persists the same when applied to a host of men or triangles individually different. In other words, these ideas are universals. Further, they are immutable and eternal in this sense, that even if, for example, no actual triangle existed, the idea of triangle, with all the geometrical truths it involves, would remain eternally the same. Moreover, these ideas enable us to contemplate, pure and unalloyed, the humanity, triangularity, etc., in which the different beings we know as men, triangles, etc., participate. Failing to analyse with sufficient accuracy the nature 76
  • 83. — PLATO AND ARISTOTLE of our ideas and the process of abstraction, and applying too hastily his guiding principle, that what-ever exists in things by participation must somewhere exist in the pure state, Plato arrived at the conclusion that there exists in a supra-sensible world a host of models or archetypes, immaterial, immutable, eternal, man in general or man in himself, triangle in itself, virtue in itself, etc. These he termed ideas, which are the object apprehended by the intellect, the faculty which attains truth—that is to say, they are reality. But what, then, is the status of the sensible world ? What are we to think of the individual changing and perishable things we see and handle ? Since they are not the ideas, they are not reality. They are, as Heraclitus taught, mere becoming. Plato did not deny their existence, but regarded them as feeble and deceptive images of reality, the object of opinion (So^a), not of certain or true knowledge, fleeting as shadows cast upon a wall. Man, therefore, captive of the body and the senses, is like a prisoner chained in a cave, on the wall of which he sees pass before him the shadows of the living beings who move behind his back in the sunlight—fugitive shadows, evading his grasp, of the idea-substances lit by the Sun of the intelligible world, God or the idea of the good. But, after all, is a metaphor an explanation ? The Platonic ideas are that in virtue of which things possess their specific natures man in himself or humanity is that which makes Socrates a man, the beautiful in itself or beauty is that which makes Alci-biades or Gallias handsome. In other words, the Platonic ideas are the essences and the perfections of things. But, on the other hand, they are distinct 77
  • 84. AN INTRODUCTION TO PHILOSOPHY from things—belong, indeed, to another world. How, then, are we to explain the relationship between things and their ideas ? Plato replied by calling them likenesses or participations of the ideas. But these terms, which later will receive in Scholasticism a profound significance, are in Plato's system nothing more than metaphors devoid of strictly intelligible content. And the further question immediately arises, why and how anything exists except ideas—that is to say, anything not pure reality. In other words, Plato has to explain what the thing is which partici-pates in the ideas, receives their likeness or reflection. Plato replied that it is matter (or the boundless, (ScTCEtpov) . And since the ideas are that which is, he was compelled to regard matter as that which is not, a sort of non-existent being : a pregnant conception which, in Aristotle's hands, was to be purged of all internal contradiction, but which, as presented by Plato, seems self-contradictory, the more so since he confuses it elsewhere with the pure space of the mathematician. {b) Undisturbed by these metaphysical difficulties, Plato proceeded to build up, in accordance with its inner logic, the edifice of his system. For the theory of ideas involved an entire system of philosophy, embracing the nature of knowledge, man, and the physical universe. Human knowledge was divided into two totally different categories : imagination (elxacria) and opinion (So^a), which are concerned with that which of its nature cannot be the object of knowledge, the visible and corruptible world and its deceptive shadows ; and 78
  • 85. PLATO AND ARISTOTLE intellectual knowledge (voyjcji,!;'), which is concerned with intelligible things and is itself subdivided into reason (Siavoia), whose object is mathematical number, and intellect (vou(^), which rises by means of dialectic to the intuitive contemplation of the idea-essences and finally of God, the super-essential good. Plato had now to explain this intellectual knowledge, the origin of the ideas in our minds, images of the eternal ideas. Since these ideas cannot be derived from the senses, which are fettered to illusion, we must receive them immediately from on high, and they must be innate in our soul. In a former existence, before its union with the body, the soul beheld the ideas and possessed intuitive knowledge. That know-ledge still remains with us, but, clouded and darkened by the Ufe of the body, it abides in the soul as a dormant memory, and it is by gradually reviving it that the quest of wisdom enables us to reconquer our original intuition of truth. Thus man is a pure spirit forcibly united with a body, as it were an angel imprisoned in the flesh {psychological dualism). The human soul lived before it animated the body, to which it is tied as a punishment for some pre-natal sin, and after death it enters another body ; for, while Plato believed in the immortality of the soul, he also held the Pythagorean tenet of transmigration, or metempsychosis. Since the physical world is not an object of know-ledge, Plato can speak of it only in fables or myths, which he develops with all the resources of a con-summate art, although they serve only to disguise the impotence of his philosophy to account for material reality. 79
  • 86. AN INTRODUCTION TO PHILOSOPHY — It is in his myths that he ascribes the production, or rather the organisation, of the world to a demiurge— whom, in the opinion of many commentators, he regarded as distinct from God and inferior to him and expounds the queer notion that man is the origin of all living organisms : the first men created by the gods were of the male sex ; those who led evil lives were changed after death into women, who, in turn, if they continued to sin, were changed into irrational brutes, perhaps even into plants. In ethics Plato, like his master Socrates, but more clearly than he, established the fundamental truth of moral philosophy : neither pleasure nor virtue nor any partial good, but God himself, and God alone, is the good of man. But how does man attain his good ? By making himself, Plato replied, as like as possible to God by means of virtue and contemplation. Plato also examined, though inadequately, the concept of virtue, and outlined the theory of the four cardinal virtues, prudence, justice, fortitude, and temperance. He taught that it is better to suffer than commit injustice, and in the Republic ^ he paints so sublime and so flawless a portrait of the righteous sufferer that it is as though he had caught a glimpse of the Divdne Face. But, as a result of his exaggerated intellectualism, he failed to distinguish the acts of the practical from those of the speculative intellect and identified virtue, which requires rectitude of the will, with knowledge, which is a perfection of the reason alone. He there-fore misapplied the principle, in itself true, that the will always follows the guidance of the understanding, and maintedned that sin is simply due to lack of know-ii, 362 A 80
  • 87. . PLATO AND ARISTOTLE ledge and that no one deliberately does evil : " the sinner is merely an ignorant person," The conse-quence of this theory, which Plato did not intend, is the denial of free will. Plato's sociology betrays the same idealist and rationaHst tendency which leads him to misapply another true principle, namely, that the part exists for the whole ; so that in his ideal republic, governed by philosophers, individuals are entirely subordinated to the good of the state, which alone is capable of rights, and disposes despotically of every possible species of property, not only the material possessions, but even the women and children, the life and hberty, of its citizens {absolute communism) {c) The radical source of Plato's errors seems to have been his exaggerated devotion to mathematics, which led him to despise empirical reahty. They were also due to an over-ambitious view of the scope of philosophy, in which Plato, like the sages of the East, though with greater moderation and discretion, placed the purification, salvation, and life of the entire man. Moreover, it is on account of these false principles latent in his system that all those philosophic dreams which tend in one way or another to treat man as a pure spirit can be traced directly or indirectly to Plato. But of Plato himself we may say that his false principles grew in an atmosphere too pure to allow them to yield their full fruit and poison the essence of his thought. St. Augustine was therefore able to extract from Plato's gold-mine the ore of truth. 8i
  • 88. AN INTRODUCTION TO PHILOSOPHY Plato's thought worked on a large scale and sought to embrace all things in a single grasp. But his superior wisdom and amazing gift of intuition pre-vented him from fixing in a final and definite state-men of doctrine many a speculation which floated vaguely before his mind. Weak points on which another philosopher would have insisted he touched lightly. Hence what in itself is a mark of imperfec-tion— vagueness, imprecision, hints, never worked out, with which he is often satisfied, a method of exposition more aesthetic than scientific, employing only m,eta-phors and symbols, a method which St. Thomas with good reason criticises severely ^—was actually his safe-guard, preserving him from a too disastrous distortion of the truths he had discovered. From this point of view it may be said that Platonism is false, if regarded in esse as a fully developed system, but, if regarded in Jieri as a progress towards a goal beyond itself, of the utmost value as a stage in the development of the true philosophy. ARISTOTLE To extract the truth latent in Platonism was the mighty reform effected by Aristotle. Aristotle success-fully took to pieces Plato's system, adapted to the exigencies of reality the formal principles he had dis-covered and misapplied, reduced his sweeping per-spectives within the limits imposed by a sublime common sense, and thus saved everything vital in his 1 Plato habuit malum modum docendi ; omnia enim figurate dicit et per symbola, intendens aliud per verba, quam sonent ipsa verba. (St. Thomas, In I De Anima, viii.) 82
  • 89. PLATO AND ARISTOTLE master's thought. He did more : he founded for all time the true philosophy. If he saved whatever was true and valuable, not only in Plato, but in all the ancient thinkers of Greece, and brought to a successful conclusion the great work of synthesis which Plato had attempted, it was because he definitively secured the attainment of reality by the human intellect. His work was not only the natural fi^it of Greek wisdom purified from Plato's mistakes and the alien elements included in Platonism ; it contained, com-pletely formed and potentially capable of unlimited growth, the body of the universal human philosophy. Before Aristotle, philosophy may be regarded as in an embryonic stage and in process of coming to birth. Thenceforward, its formation complete, it was capable of indefinite development, and knew no bounds. Inventum philosophicum semper perfectibile. In fact, Greek speculation after Aristotle had spent its force, and was unable to keep firm hold of the truth. It would receive considerable material enrich-ment, but in essentials would deform instead ofperfecting philosophy.^ {a) For twenty years Aristotle was Plato's disciple ; but he was a disciple with the equipment of a formid-able critic. No one has refuted Plato's idealism more powerfully than he, or more effectively demolished a system which places the substance of things outside themselves. It is perfectly true that the primary object of the intellect is, as Socrates taught, the essences of things ; ^ This is the reason why we have ended with Aristotle this introduc-tory sketch of the history of philosophy, or more exactly of the formation of philosophy. 83
  • 90. AN INTRODUCTION TO PHILOSOPHY perfectly true also, as Plato had perceived, that the essence of Peter, Paul, or John is humanity or human nature, abstracting from the individual characteristics peculiar to Peter, Paul, or John. But this essence as a universal exists only in the intellect—in our mind,* which extracts or abstracts it from the things in which it exists individualised ^—and, on the other hand, it is solely as an object of intelligence (inasmuch as it cannot be conceived apart from certain attributes), and not in its real existence, that it is eternal and necessary. Therefore the essences of perishable things possess no separate existence in the pure state,- and the entire Platonic world of archetypal ideas is sheer fiction. The truth of the matter is, as we shall prove later in detail, that there exists in everything an intelligible and immaterial element, which Aristotle calls form^ in virtue of which it possesses a specific nature or essence. But this principle is not separate from things ; it inheres in them as one of the factors which constitute their substance. Thus individual objects, though mutable and mortal, are no longer deceptive shadows ; they are reality. If real objects of a higher order exist, none are more immediately accessible to our knowledge. If the sensible world be, as it were, an imperfect likeness of the divine life of pure spirit, it is a being which resembles another being, not a mere image which has no existence in itself. If the world is subject to becoming, it is not pure becoming, but contains 1 And primarily in the Divine Intellect, as the Schoolmen were to explain, thus taking accovmt of the truth contained in Plato's exemplarism. » Cf. St. Thomas, In I Metaph., 1. lo, 158 (ed. Gathala). 84
  • 91. PLATO AND ARISTOTLE enduring and substantial realities. If there is no science of the individual object of sense as such, never-theless a science of sensible reaUty is possible, because there exists, incarnate, so to speak, in that reality, something intelligible and immaterial. Thus the corporeal universe is the object not of mere opinion, which can be expressed only by myth and allegory, but of scientific knowledge, the science of physics. Aristotle was the true founder of physics.^ His incomparably powerful genius viewed mobihty in the immutable Hght of intellect, showed that all change obeys unchanging laws, laid bare the nature of motion itself, of generation and corruption, and distinguished the four species of causation operative in the sensible world. In language strangely trenchant and severe, he sums up his long polemic against the doctrine of ideas. Plato, he argues, completely misconceived the nature of the formal cause when he separated it from things. While he fancied " he was stating the substance of perceptible things," he asserted " the existence of a second class of substances," and his " account of the way in which they are the substances of perceptible things is empty talk ; for ' sharing ' (participating) means nothing." He thus made it impossible to give a satisfactory account of nature, and, by attributing all causation and all true reality to the ideas, he was unable to distinguish in the activity of things the lespective parts played by the efficient and the final 1 Aristotle's experimental physics (the science of phenomena) is a magnificent intellectual construction totally ruined by mistakes of fact. But his philosophical physics (the science of moving being as such) lays down the foundations and principles of every true philosophy of nature. 85
  • 92. AN INTRODUCTION TO PHILOSOPHY cause. He thus neglected " the efficient cause which is the principle of change," He further failed to give any account of the cause of " that which we see to be the cause in the case of the arts, for the sake of which mind and nature produce all that they do produce." For " mathematics has come to be the whole of philosophy for modern thinkers, and they profess to explain all other things by mathematics." " And as to motion, if the ideas are motionless," there is no archetype of motion in the world of ideas, but in that case " whence," according to the Platonists, " did motion come ? If we cannot explain motion, the whole study of nature has been annihilated." ^ Refutation of the theory of ideas logically in-volved the criticism and correction of all the other parts of the Platonic system. In epistemology Aristotle showed that physics, mathematics, and metaphysics, or the first philosophy, are indeed three distinct sciences, but that they are distinguished by their subject-matter, not by the faculty employed, which in all alike is reason. But his most important achieve-ment in this sphere was to prove, by the marvellous analysis of abstraction which dominates his entire philosophy, that our ideas are not innate memories of pre-natal experience, but derived from the senses by an activity of the mind. In psychology, if in his reaction against Plato's metempsychosis, and from an excessive caution, he refrained from inquiring into the condition of the soul after death, at least he laid the firm foundations of the spiritualist doctrine by proving, on the one hand, in opposition to Plato's dualism, the substantial unity of 1 Meiaph., i, 9, 992 a 25-992 b 10. 86
  • 93. PLATO AND ARISTOTLE the human being, composed of two substantial parts incomplete and complementary, and, on the other, against the materialists, the spiritual nature of the operations of the understanding and will. He thus created the only psychology capable of assimilating and explaining the vast material accumulated by modern experiments. In ethics, by distinguishing between the speculative judgment (which proceeds from the understanding alone) and the practical judgment (which proceeds con-jointly from the will), he showed how free will is possible, and how the sinner does what he knows to be evil, and drew, especially in his treatment of the cardinal virtues and in his analysis of human acts, the outlines of what was to be, so far as the natural order is concerned, the ethics of Christianity. • {b) But Aristotle must be studied, not only in his attitude to Plato, but absolutely in his attitude to that which is. For Plato did no more than furnish him with the occasion to wrestle with the problem of being. Aristotle won the match, leaving us his great concepts of potentiality and act, matter and form^ the categories^ the transcendentals, the causes, as weapons wherewith to wage the same intellectual contest, and teaching us, as a irue master of wisdom, to rise above the study of visible and perishable things to contemplation of the hving, imperishable reality which knows no change. " Immovable in its pure activity, this being is in no way subject to change. . . . On such a prin-ciple depend the heavens and the world of nature. Its Ufe is such as the best which we enjoy, and enjoy for but a short time. For it is ever in this state, since 87
  • 94. AN INTRODUCTION TO PHILOSOPHY its act is also pleasure—the act of the supreme intelli-gence, pure thought thinking itself. ... If God is always in that good state in which we sometimes are, this compels our wonder ; and if in a better, this is yet more wonderful. Life also belongs to God : for the act of thought is Ufe, and God is that act ; and God's essential act is Ufe most good and eternal. We say therefore that God is a living being, eternal and perfect, so that life which endures everlastingly belongs to God, for God is this life." ^ Moreover, this God is perfectly one, absolutely single. " Those who say mathematical number is first, and go on to generate one kind of substance after another and give different principles for each, make the substance of the universe a mere series of episodes and they give us many governing principles ; but the world must not be governed badly. As Homer observed, the rule of many is not good ; one is the ruler^ * Thus Aristotle, as Alexander of Aphrodisias remarks in a fine passage of his Commentary on the Metaphysics* " leads us fi"om the things which are themselves on the lowest plane, but most familiar to us, up to the Father, who has made all things, to God the most sublime, and proves that as the founder is the cause of the unity of the globe and the brass, so the Divine Power, author of unity and maker of all things, is for all beings the cause of their being what they are." Aristotle's mind was at once extremely practical and extremely metaphysical. A rigorous logician, but also a keen-sighted realist, he gladly respected the 1 Metaph., xii, 7, 1072 b ; 9, 1074 b 35. * Metaph., xii, 10, 1076 a. ' Alexander of Aphrodisias, In Metaph., ad 1045 a 36. 88
  • 95. PLATO AND ARISTOTLE demands of the actual, and found room in his specula-tion for every variety of being without violating or distorting the facts at any point, displaying an intellec-tual vigour and freedom to be surpassed only by the crystalline lucidity and angelic force of St. Thomas Aquinas. But this vast wealth is arranged in the light of principles, mastered, classified, measured, and dominated by the intellect. It is the masterpiece of wisdom, a wisdom which is still wholly human, but nevertheless, from its lofty throne, embraces with a single glance the totaUty of things, Aristotle, however, was a profound rather than a comprehensive thinker. He took little care to display the proportions and wide perspectives of his philo-sophy ; his primary object w£is to apprehend by an absolutely reliable method and with a faultless preci-sion what in every nature accessible to human know-ledge is most characteristic, most intimate—in short, most truly itself. Therefore he not only organised human knowledge, and laid the solid foundations of logic, biology, psychology, natural history, meta-physics, ethics, and politics, but also cut and polished a host of precious definitions and conclusions sparkling with the fires of reality. It can therefore be affirmed without hesitation that among philosophers Aristotle holds a position alto-gether apart : genius, gifts, and achievement—all are unique. It is the law of nature that the sublime is difficult to achieve and that what is difficult is rare. But when a task is of extraordinary difficulty both in itself and in the conditions it requires, we may expect that there will be but one workman capable of its accomplishment. Moreover, a well-built edifice is 89
  • 96. AN INTRODUCTION TO PHILOSOPHY usually built not on the plans of several architects, but on the plan of a single one. If, therefore, the edifice of human wisdom or philosophy is to be adequately constructed, the foundations must be laid once for all by a single thinker. On these foundations thousands of builders will be able to build in turn, for the growth of knowledge represents the labour of generations and will never be complete. But there can be but one master-builder.^ For that reason, in spite of the mistakes, de-fects, and gaps which betray in his work the limita-tions of human reason,^ Aristotle is as truly the ^ Descartes remarks very truly in his Discours de la Mithode : " Works composed of many pieces and made by the hand of several workmen are not so perfect as those which are the work of a single individual." But he was wrong in believing (i) that he was the man destined to lay the foundation of philosophy, a work which the ancients had failed to accomplish ; (ii) that by himself he was competent—at least, given sufficient time and experience—not only to lay the foundation of science, but to complete the edifice ; and (iii) in rejecting contemp-tuously the entire achievement of preceding generations, together with the traditional wisdom of humanity. Aristotle, on the contrary, succeeded in his task by constant criticism and analysis of his predecessors' thought, and by making use of the accumulated results of human speculation in the past. 2 Aristotle is often credited with certain errors made by his disciples or commentators, especially about the human soul and the divine knowledge and causality. But a careful study of the text proves that when the philosopher maintained that the intellect is separate, he meant that it is separate from matter, not from the soul itself {cf. St. Thomas, In III De Anima, 4 and 5) , and therefore he did not deny, as is often asserted, the personal immortality of the human soul {cf. also Metaph., xii, 3, 1070 a 26). Nor did he teach that God is not the efficient cause of the world and moves it only as the end, or good, which it desires. (The passage in the Metaph., xii, 7, means simply that God moves as final cause or object of love the intelligence which moves the first heaven ; he does not affirm that God can act only as final cause and has not made things. On the contrary, in Metaph., ii, i , 993 b 28, he says that the heavenly bodies are dependent on the first cause, not only 90
  • 97. PLATO AND ARISTOTLE philosopher par excellence, as St. Thomas is the theo-logian.^ for their motion, but for their very being. Cf. Metaph. vi, i, 1026 b 17.) Cf. also the passage from Alexander of Aphrodisias, quoted above in the text, in which God's efficient causality in Aristotle's system is admirably brought out. As for the passage {Metaph., xii, 9) in which Aristotle investigates the formal object of the Divine Intellect, remarking that it is better not to know than to know certain things of a lower order, it most certainly does not amount to the denial that God knows the things of the world ; the statement is put forward simply to prepare for the solution of the question discussed. That solution, as indicated by Aristotle, is formally true, and consists in the proposition, which St. Thomas later affirmed more explicitly, that the Divine Intellect, because of its absolute independence, has no other formal object than the Divine Essence itself, and therefore does not know the things of the world in themselves, but in that essence in which every-thing is life. It remains true, nevertheless, that Aristotle committed serious errors (for instance, his attempt to prove the existence of the world ab aeterno) , and was also guilty of many omissions. In particular, the doctrine of creation, which follows with absolute necessity from his principles, is nowhere explicitly formulated by him (indeed, no heathen philosopher reached a clear notion of creation ex nihilo); and on those questions which, though in themselves capable of rational proof, are most difficult to solve without the aid of revelation—the relation of the world to God, the lot of the soul after death—he maintained a reserve, which was perhaps very prudent in itself, but leaves his work manifestly incomplete. 1 Goethe, repeating the theme of Raphael's wonderful School of Athens, in which Plato is depicted as an inspired old man, his face turned heavenward, Aristotle as a man in the full vigour of youth pointing triumphantly to the earth and its realities, has drawn in his Theory of Colours (Part 2, Ueberliefertes) a striking comparison between Plato and Aristotle. " Plato," he says, " seems to behave as a spirit descended from heaven, who has chosen to dwell a space on earth. He hardly attempts to know this world. He has already formed an idea of it, and his chief desire is to communicate to mankind, which stands in such need of them, the truths which he has brought with him and delights to impart. If he penetrates to the depth of things, it is to fill them with his own soul, not to analyse them. Without intermission and with the burning ardour of his spirit, he aspires to rise and regain the heavenly abode from which he came down. The aim of all his 91
  • 98. AN INTRODUCTION TO PHILOSOPHY (c) Aristotle was born in 384 at Stagira, in Thrace.' The son of a doctor, by name Nicomachus, he belonged to the family of the Asclepiadae, descended, it was said, from Aesculapius. At the age of eighteen he became a pupil of Plato, whose lectures he attended until his death (347). After Plato's death he went to live at Atarneus in Mysia where Hermias was king, and from there to Mytilene. He then spent eight years at the court of King Philip of Macedon, where he became Alexander's tutor. After his pupil's acces-sion to the throne, he settled in Athens, where he established his school at the Lyceum (a gymnasium consecrated to Apollo Lycaeus). He taught as he walked to and fro with his pupils under the trees of the Lyceum, whence the name Peripatetics (walkers) by which his disciples became known. He spent twelve years in Athens. When the party opposed discourse is to awaken in his hearers the notion of a single eternal being, of the good, of truth, of beauty. His method and words seem to melt, to dissolve into vapour, whatever scientific facts he has managed to borrow from the earth. " Aristotle's attitude towards the world is, on the other hand, entirely human. He behaves like an arcliitect in charge of a building. Since he is on earth, on earth he must work and build. He makes certain of the nature of the ground, but only to the depth of his foundations. Whatever lies beyond to the centre of the earth does not concern him. He gives his edifice an ample foundation, seeks his materials in every direction, sorts them, and builds gradually. He therefore rises like a regular pyramid, whereas Plato ascends rapidly heavenward like an obelisk or a sharp tongue of flame. " Thus have these two men, representing qualities equally precious and rarely found together, divided mankind, so to speak, between them." 1 " We must remember, moreover, that Stagira, a city of Chalcidice, was a Greek colony where Greek was spoken. It is therefore incorrect to regard Aristotle, as he is sometimes regarded, as only half Greek. He was a pure Hellene, as pure as Parmenides, for example, or Anaxagoras." (Hamelin, Le Systeme d'Aristote, p. 4.) 92
  • 99. PLATO AND ARISTOTLE to the Macedonians brought against him an accusation of impiety, on the pretext of a hymn he had once composed on the death of his friend Hermias, he retired to Chalcis, where he died at the age of sixty-three (322). The story is told of him that his love of study was so great that he devised the plan of holding in his hand, while at work, a ball of copper which, if he fell asleep, would rouse him by falling into a metal basin. To assist his researches Philip and Alexander placed their vast resources at his disposal. He wrote books to be read by the general public {dialogues), which are all lost—Cicero praised their style in the highest terms : flumen aureum orationis fundens Aristoteles {Acad., II, 38, 119)—and acroamatic books, in which he summarised the lectures given to his disciples ; the majority of these have survived.^ The history of these books, as related by Strabo, is very strange, and illustrates, as aptly as Pascal's remark on Cleopatra's nose, the part played by trifling * The following is a list of Aristotle's works : — i. The collection of works dealing with Logic and known collectively by the name of Organon (instrument of scientific knowledge). They consist of the Kary^yoplcn, the Categories ; the 'AvaXvTiKa irpdrepa Kai Offrepa, the Prior and Posterior Analytics ; the TdmKa, the Topics ; the Hfpl <To(piaTiKQ)v iXiyx'^i', On Sophistic Arguments ; and the Uepl ipfiTivelai, On Interpretation, De Interpreiatione, a treatise on the meaning of proposi-tions, which in spite of Andronicus's rejection must be accepted as authentic. ii. The Physics, ^vciKi] dxpSaais (the authenticity of Book 7 is doubtful), with which we must group the followii^ treatises : On the Heavens {De Caelo), Tlepl dvpavoO ; On Generation and Corruption, Uepl yeveaiu^ Kai <p0opS.s ; On the Parts of Animals, Uepl ^i^wv fi6pitav ; On the Soul, Uepl ^vxv^ ; On Sensation, Tlepl al(r6r,ffews kcu aladriruv ; On Memory, Uepl tj.vf)fi.r)% Kai avafivva-eta^ ; the Meteorology, MereupoXoyiKd ; the History of Animals, riepi ri fi^a l<rroplai (the authenticity of Book 10 is doubtful) ; and many other treatises, several of which are of dubious 93
  • 100. AN INTRODUCTION TO PHILOSOPHY accidents in determining the destinies of mankind. At the philosopher's death they were inherited together with his Ubrary by his disciple and successor Theo-phrastus. Theophrastus bequeathed them to a disciple authenticity, especially the De Mundo. (The treatise on Physiognomy is spurious, but is apparently a compilation from authentic fragments.) iii. The Metaphysics, Td uer^ rk (f>v<nKd, of which the second book, a iXttTTov, is the work of a disciple, Pasicles of Rhodes. iv. The Nicomachean Ethics, 'HdiKo, ^iKOfiixfia, and the Eudemian Ethics, 'KdiKct EvBrifida. The latter work was composed, not by Aristotle, but by Eudemus himself. To these we may add the Great Ethics (Magna Moralia), "ROiko, tieyaha, which is a risumi of the two preceding; and therefore not written by Aristotle ; the Politics, UoKitiko. ; the Poetics, Tlepi noti]TiKT]i ; and the Rhetoric, T^x^r) pr/TopiKifi. The treatise On Virtues and Vices, the Economics, and the Rhetoric to Alexander are spurious. In 1 891 was discovered and published the Constitution of Athens,'AO-nvalwv TToXiTela. It formed part of a collection in which Aristotle gave an outline of the constitutions of 1 58 Greek states. Of the Schoolmen who commented on the works of Aristotle, the most important are Albertus Magnus, St. Thomas, and Silvester Maurus, whose paraphrase and commentary following the text word for word may still be usefully consulted. Saint Thomas wrote commentaries : (a) on the De Interpretatione (unfinished and replaced by Cajetan's for lessons 3-14 of Book 2) ; [b) on the Posterior Analytics ; (c) on the Physics ; (d) on the De Caelo et Mundo (St. Thomas died before it was completed, and from Book 3, lesson 8, the commentary was continued by his pupil, Peter of Auvergne) ; [e) on the Generation and Corruption of Animals (St. Thomas's unfinished Commentary has been completed by passages borrowed from other writers, in particular from the Commentary of Albertus Magnus) ; (/) on the Meteorology (completed by another hand from Bk. 2, lesson 11); [g) on the De Anima (the commentary on Books 2 and 3 is by St. Thomas himself, the commentary on Bk. i compiled from his lectures by one of his pupils, Raynald of Piperno ; (A) on the Parva Naturalia [De Sensu et Sensato, de Memoria et Reminiscentia, de Somno et Vigilia) ; (i) on the Metaphysics (modern edition by Padre Cathala, Turin, Marietti, 1915) ; (j) on the Nicomachean Ethics ; {k) on the Politics (completed by Peter of Auvergne from Bk. 3, lesson 6, or, as others think firom Bk. 4) . Cf. De Rubeis, Dissert. 23 in vol. i. Op. Omn. S. Thomae Aq., Leonine edition. For the writings of St. Thomas and the authenticity of his various minor works, see Mandonnet, O.P., Des Ecrits autherUiques de Saint Thomas (reprinted fi:om the Revue Thomiste, 1909-19 10) Friburg. 94
  • 101. PLATO AND ARISTOTLE named Neleus, Neleus in turn to his heirs. The latter, fearing they might be seized for the royal library at Pergamum, hid them in an underground chamber. They died and the manuscripts were lost. They remained lost for a century and a half and were only recovered by the good fortune of a successful book-lover. About the year lOO b.g. the descendants of Neleus's heirs discovered the manuscripts (in a very bad condition, as we can well imagine) and sold them for a large sum to a wealthy collector, Apellicon of Teos, who published them, in a very faulty edition. At the capture of Athens by the Romans in 86 b.g. they came into the possession of Sulla. The gram-marian Tyrannic had access to them and made use of them, and finally Andronicus of Rhodes catalogued and republished them.^ Commentaries were composed by Alexander of Aphrodisias (second century a.d.), also by several neo-Platonic philosophers. Porphyry (third century), Themistius (fourth century), Sim-plicius, and Philo (sixth century). 1 Strabo, Geog., xiii, i, 54 ; Plut. Sulla, 26. Strabo's testimony is of considerable weight. It has, however, been proved that some of the most important scientific treatises of Aristotle were known to the Peri-patetics and their opponents in the third and second centuries B.C. We must therefore conclude that Strabo's account is reliable in its positive assertions, so far as the history of Aristotle's " acroamatic " manuscripts are concerned, but inaccurate, or at least exaggerated, in its negative statements. More or less complete copies of the Philosopher's works must have been in circulation in the Peripatetic school before Apellicon's discovery. We may nevertheless agree with Hamelin's conclusion that " the scientific writings of Aristotle were little read even by the Peripatetic school in its degeneracy. Apellicon's discovery would have had the effect of making these works once more fashionable." The truth of the matter therefore would be that, before this discovery and the works of Andronicus, Aristotle's scientific treatises were not indeed unknown, as Strabo says, but at any rate litde and badly known. 95
  • 102. AN INTRODUCTION TO PHILOSOPHY The Scholastic tradition, which grew up from the eighth century onward in the Christian West, was for long ignorant of Aristotle's original works, with the exception of the Organon (the treatises on logic), which had been translated into Latin ^ by Boethius (480-526). But it was acquainted with his thought, which had been transmitted and popularised at second hand and formed an integral part of the great philosophic synthesis of late antiquity, Platonic though it was in the main, on which the Fathers, especially St. Augustine, had drawn so largely in the service of the faith. In the Christian schools Aristotle's logic was taught in Boethius's translation. But it was not until the latter part of the twelfth century that the other writings of the Philosopher {physics, metaphysics, ethics) began to reach the Schoolmen, mainly, it would appear, as a result of the ardent polemic conducted at that date by the leaders of Christian thought against the philosophy of the Arabs, who possessed these books together with the neo-Platonic commen-taries in a Syrian version translated later into Arabic, and appealed to their authority. At first the object of considerable suspicion * on account of the source ^ Later Boethius's work was partially lost, and it was not until after H41 that certain books of the Organon, reintroduced from the Arabs, began to appear in the philosophic literature of the Middle Ages, where they formed what was then known as the Logica Nova. These were the Prior and Posterior Analytics, the Topics, and the Sophistic Arguments. Cf. de Wulf, Hist, de la Phil, midiivale, and ed. pp. 149 sqq. * Censures (issued in 12 10 by a council of the province of Sens which met at Paris, and renewed in 12 15, in the statute imposed on the University of Paris by the legate Robert de Courcon, a statute confirmed by Gregory IX in 1231 and by Urban IV in 1263) forbade the employment of Aristotle's writings in public lectures or private teaching. We must remember, however, that, as M, Forget points out {Rapp. cm 96
  • 103. PLATO AND ARISTOTLE from which they had been received and the mistakes which the Arab commentators had foisted into them, all the works of Aristotle were soon translated into Latin, at first from the Arabic text, later ^ from the original Greek. ^ Now took place the meeting of human wisdom and divine truth, of Aristotle and the Faith. All truth belongs of right to Christian thought, as the spoils of the Egyptians to the Hebrews. Quaecunque igitur apud omnes praeclara dicta sunt, nostra Christianorum sunt,* because according to that saying of St. Ambrose, which St. Thomas deHghted to quote, every truth, whoever said it, comesfrom the Holy Spirit.'^ But someone must actually take possession, someone must enrol in congr. scientif. intern, des cath., Brussels, 1894), individuals remained free to read, study, and comment on these books in private. Moreover, the censures bound only the University of Paris. In 1229 the University of Toulouse, founded and organised under the protection of the papal legate himself, attracted students by announcing lectxires on the books forbidden at Paris. Finally, even at Paris, when the faculty of arts included in its course from 1255 public lectures on the Physics and Metaphysics, ecclesiastical authority made no attempt to interfere. And, more significantly still. Pope Urban IV, a kw years later, took under his personal patronage William of Moerbeke's translation of Aristotle and St. Thomas Aquinas's Commentaries. See GhoUet, " Aristotelisme de la Scolastique " in Vacant and Mangenot's Dictionnaire de Thiologie, and de Wulf, op. cit., p. 242. 1 Some of Aristotle's works were apparently read at first in a Latin translation from Arabic, others in a direct translation from the Greek. In any case it was not long before the latter entirely superseded the former. St. Thomas used only direct versions from the Greek. 2 The best of these translations is that of the entire works of Aristotle made between 1260 and 1270 by William of Moerbeke at the suggestion and under the supervision of St. Thomas. It is an absolutely literal rendering of the Greek text. ' St. Justin, ApoL, ii, 1 3. * Omne verum a quocumque dicatur, a Spiritu sancio est. 97
  • 104. AN INTRODUCTION TO PHILOSOPHY the royal service of Christ the marvellous intellect of Aristotle. This work, begun by Albert the Great (i 193-1280), was continued and brought to a success-ful conclusion by St. Thomas Aquinas (1225-1274). Its accomplishment demanded a rare conjunction of favourable conditions—the ripe culture of the age of St, Louis, the magnificent organisation of the Domini-can order, the genius of St. Thomas.^ St. Thomas, whom the Church has proclaimed Doctor par excellence, Doctor Communis Ecclesiae, and whom she has enthroned as the universal teacher of her schools, was not content with transferring the entire philosophy of Aristotle to the domain of Christian thought, and making it the instrument of a unique theological synthesis ; he raised it in the process to a far higher order, and, so to speak, transfigured it. He purged it from every trace of error—that is to say, in the philosophic order, for so far as the sciences of observation or phenomena are concerned, St. Thomas was no more able than Aristotle to escape the errors prevalent in his day, errors which do not in any way affect his philosophy itself. He welded it into a powerful and harmonious system ; he explored * For the successful performance of such a task it was also requisite that Christian thought should have attained the high degree of elabora-tion alike in the philosophical and theological order which it had received from the Fathers and the Scholastic predecessors of St. Thomas. Therefore the work of Albertus Magnus and St. Thomas was not to change, but on the contrary to complete Christian philosophy, giving it its mature expression. If contemporaries were primarily impressed with the novelty of their work—a novelty' of completion, not of alteration —the reason is that the final process which perfects a system must always come as a shock to those who witness it, and who most likely are attached by force of habit to certain aspects of its imperfect state as such. 98
  • 105. PLATO AND ARISTOTLE its principles, cleared its conclusions, enlarged its horizon ; and, if he rejected nothing, he added much, enriching it with the immense wealth of the Latin Christian tradition, restoring in their proper places many of Plato's doctrines, on certain funda-mental points (for example, on the question of essence and existence) opening up entirely new perspectives, and thus giving proof of a philosophic genius as mighty as that of Aristotle himself Finally, and this was his supreme achievement, when by his genius as a theologian he made use of Aristotle's philosophy as the instrument of the sacred science which is, so to speak, " an impress on our minds of God's own knowledge," ^ he raised that philosophy above itself by submitting it to the illumination of a higher light, which invested its truth with a radiance more divine than human. Between Aristotle as viewed in himself and Aristotle viewed in the writings of St. Thomas is the difference which exists between a city seen by the flare of a torchlight procession and the same city bathed in the light of the morning sun. For this reason, though St. Thomas is first and foremost a theologian, we may as appropriately, if not with greater propriety, call his philosophy Thomist rather than Aristotehan. This philosophy of Aristotle and St, Thomas is in fact what a modern philosopher has termed the natural philosophy of the human mind, for it develops and brings to perfection what is most deeply and genuinely natural in our intellect aUke in its elementary appre-hensions and in its native tendency towards truth. Sum, TTieoL, i, q. i, a 3, a</ 2. 99
  • 106. AN INTRODUCTION TO PHILOSOPHY — It is also the evidential philosophy, based on the double evidence of the data perceived by our senses and our intellectual apprehension of first principles the philosophy of being, entirely supported by and modelled upon what is, and scrupulously respecting every demand of reality—the philosophy of the intellect, which it trusts as the faculty which attains truth, and forms by a discipline which is an incomparable mental purification. And for this very reason it proves itself the universal philosophy in the sense that it does not reflect a nationality, class, group, temperament, or race, the ambition or melancholy of an individual or any practical need, but is the expression and product of reason, which is everywhere the same ; and in this sense also, that it is capable of leading the finest intellects to the most subHme knowledge and the most difficult of attainment, yet without once betraying those vital convictions, instinctively acquired by every sane mind, which compose the domain, wide as humanity, of common sense. It can therefore claim to be abiding and permanent [philosophia perennis) in the sense that before Aristotle and St. Thomas had given it scientific formulation as a systematic philo-sophy, it existed from the dawn of humanity in germ and in the pre-philosophic state, as an instinct of the understanding and a natural knowledge of the first principles of reason and ever since its foundation as a system has remained firm and progressive, a powerful and living tradition, while all other philo-sophies have been born and have died in turn. And, finally, it stands out as being, beyond comparison with any other, one ; one because it alone bestows harmony and unity on human knowledge—both meta-lOO
  • 107. PLATO AND ARISTOTLE physical and scientific—and one because in itself it realises a maximum of consistency in a maximum of complexity, and neglect of the least of its principles involves the most unexpected consequences, distorting our understanding of reahty in innumerable directions. These are a few of the external signs which witness to its truth, even before we have studied it for ourselves and discovered by personal proof its intrinsic certitude and rational necessity. lOI
  • 108. DEFINITION OF PHILOSOPHY We began by calling philosophy human wisdom. Now that the history of its origins has given us further information as to the nature and object of this wisdom, we are in a position to attempt a more precise definition of philosophy. For this purpose we shall take philosophy to mean philosophy par excellence, the first philosophy or meta-physics. What we shall say of it in the absolute sense [simpliciter) will be applicable relatively [secundum quid) to the other departments of philosophy. Philosophy is not a " wisdom " of conduct or practical fife that consists in acting well. It is a wisdom whose nature consists essentially in knowing. How ? Knowing in the fullest and strictest sense of the term, that is to say, with certainty, and in being able to state why a thing is what it is and cannot be otherwise, knowing by causes. The search for causes is indeed the chief business of philosophers, and the knowledge with which they are concerned is not a merely probable knowledge, such as orators impart by their speeches, but a knowledge which compels the assent of the intellect, Hke the knowledge which the geometrician conveys by liis demonstrations. But certain knowledge of causes is termed science. Philo-sophy therefore is a science. Knowing by what medium, by what light ? Know- 102
  • 109. DEFINITION OF PHILOSOPHY ing by reason, by what is called the natural light of the human intellect. This is a quality common to every purely human science (as contrasted vsdth theology). That is to say, the rule of philosophy, its criterion of truth, is the evidence of its object. The medium or light by which a science knows its objects is termed in technical language its lumen sub quo, the light in which it apprehends the object of its knowledge (itself termed the object, quod). Each of the different sciences has its own distinctive light {lumen sub quo, medium sen motivum formale) which corresponds with the formal principles by means of which they attain their object. But these different principles are aUke in this, that they are all known by the spontaneous activity of our intellect, as the natural faculty of knowledge, in other words by the natural light of reason—and not, hke the principles of theology, by a supernatural communication made to man {revelation), and by the light of faith. We have now to consider the object quod of philosophy. Knowing what ? To answer this question we may recall the subjects which engaged the attention of the different philosophers whose teachings we have summarised. They inquired into everything—know-ledge itself and its methods, being and non-being, good and evil, motion, the world, beings animate and inanimate, man and God. Philosophy there-fore is concerned with everything, is a universal science. This does not, however, mean that philosophy absorbs all the other sciences, or is the sole science, of which all the rest are merely departments ; nor on 103
  • 110. AN INTRODUCTION TO P H I L O S O P H 'y — the Other hand that it is itself absorbed by the other sciences, being no more than their systematic arrange-ment. On the contrary, philosophy possesses its distinctive nature and object, in virtue of which it differs from the other sciences. If this were not the case philosophy would be a chimera, and those philosophers whose tenets we have briefly sketched would have treated of unreal problems.^ But that philosophy is something real, and that its problems have the most urgent claim to be studied, is proved by the fact that the human mind is compelled by its very constitution to ask the questions which the philosophers discuss, questions which moreover in-volve the principles on which the certainty of the conclusions reached by every science in the last resort depends. " You say," wrote Aristotle in a celebrated dilemma, *' one must philosophise. Then you must philosophise. You say one should not philosophise. Then (to prove your contention) you must philosophise. In any case you must philosophise." * ^ No doubt they also studied many questions relating to die special sciences, for the differentiation of the sciences had not been carried so far as in modem times. Nevertheless the primary object of their study lay elsewhere, and, at least after Socrates, those special sciences astronomy, geometry, arithmetic, music, medicine, and geography which the ancients cultivated with success developed separately, and in clear distinction from philosophy. The very .ki'ory of the special sciences, which in modern times have made such enormous progress independently of philosophy and as autonomous branches of study, plainly proves that they are no part of the former 2 " el fikv (pCKoao(ptyreov, (ptoao(pr,Tiov, Kal, el fiii cp ' TocprjTiov, (pCKoaocpt)- riov, w6.vT(t}% &pa. (pioiTo(prjTiov." This dilemma is taken from the UporftirTiKd?, a lost work of which only a few fragments have come down to us, {Cf. fr. 50, 1483 b 29, 42 ; 1484 a 2, 8, 18.) 104
  • 111. DEFINITION OF PHILOSOPHY But how can pliilosophy be a special science if it deals with everything ? We must now inquire under what aspect it is concerned with everything, or, to put it another way, what is that which in everything directly and for itself interests the philosopher. If, for example, philosophy studies man, its object is not to ascertain the number of his vertebras or the causes of his diseases ; that is the business of anatomy and medicine. Philosophy studies man to answer such questions as whether he possesses an intellect which sets him absolutely apart from the other animals, whether he possesses a soul, if he has been made to enjoy God or creatures, etc. When these questions are answered, thought can soar no higher. No problems Ue beyond or above these. We may say then that the philosopher does not seek the explanation nearest to the phenomena perceived by our senses, but the explanation most remote from them, the ultimate explanation. This is expressed in philo-sophical terminology by saying that philosophy is not concerned with secondary causes or proximate explana-tions ; ^ but on the contrary with Jirst causes, highest principles or ultimate explanations. Moreover, when we remember our conclusion that philosophy knows things by the natural light of reason, it is clear that it investigates the first causes or highest principles in the natural order. When we said that philosophy is concerned with everything, everything which exists, every possible object of knowledge, our statement was too indefinite. We determined only the matter with which philosophy ^ That is to say, approximating to the particulars of sensible phenomena. 105
  • 112. AN INTRODUCTION TO PHILOSOPHY deals, its material object, but said nothing of the aspect under which it views that object, or the attributes of that object which it studies that is to say, we did ; not define its,formal object, its formal standpoint. The formal object of a science is the aspect under which it apprehends its object, or, we may say, that which it studies primarily and intrinsically and in reference to which it studies everything else ^ that which philo-sophy ; studies in this formal sense in things, and the standpoint from which it studies everything else, is the first causes or highest principles of things in so far as these causes or principles belong to the natural order. The material object of a faculty, science, art, or virtue, is simply the thing or subject-matter—without further quahfication—with which that faculty, science, art, or virtue, deals. For instance, the material object of chemistry is inorganic bodies ; of the faculty of sight, objects within our range of vision. But this does not enable us to distinguish between chemistry and physics, which is also concerned with inorganic bodies, or between sight and touch. To obtain an exact definition of chemistry we must define its object as the intrinsic or substantial changes of inorganic bodies, and similarly the object of sight as colour. We have now defined the formal object {objectum formate quod), that is to say, that which immediately and of its very nature, or intrinsically and directly, or again neces-sarily and primarily (these expressions are equivalent renderings of the Latin formula per se primo) , is appre-hended or studied in things by a particular science, 1 Qtod per se primo haec scientia consideral et sub cujus ratione caetera omnia cogncscil. 1 06
  • 113. DEFINITION OF PHILOSOPHY art, or faculty, and in reference to which it apprehends or studies everything else. Thus philosophy, alone among the branches of human knowledge, has for its object everything which is. But in everything which is it investigates only the first causes. The other sciences, on the contrary, have for their object some particular province of being, of which they investigate only the secondary causes or proximate principles. That is to say, of all branches of human knowledge philosophy is the most sublime. It follows further that philosophy is in strictest truth wisdom, for it is the province of wisdom to study the highest causes : sapientis est altissimas causas considerare. It thus grasps the entire universe in a small number of principles and enriches the intellect without burdening it. The account we have just given is applicable in an unquaUfied sense only to the first philosophy or meta-physics, but may be extended to philosophy in general, if it is regarded as a body of which metaphysics is the head.^ We shall then define philosophy in general as a universal body of sciences * whose formal stand- ^ The ancients understood by the term philosophy the sum-total of the main branches of scientific study {physics, or the science of nature ; mathematics, or the sciences of proportion ; metaphysics, or the science of being as such ; logic ; and ethics). There could therefore be no question of distinguishing between philosophy and the sciences. The one question with which they were concerned was how to distinguish the first philosophy, or metaphysics, from the other sciences. We, on the contrary, since the enormous development of the special sciences, must distinguish from them not only metaphysics (the science of absolutely first principles) but the study of the first principles in a particular order (for instance, the mathematical or the physical) ; and the entire body of these constitutes what we call philosophy. 2 Only metaphysics and logic constitute a universal science speci-fically one. 107
  • 114. . AN INTRODUCTION TO PHILOSOPHY — point ^ is first causes (whether absolutely first causes or principles, the formal object of metaphysics, or the first causes in a particular order, the formal object of the other branches of philosophy). And it follows that metaphysics alone deserves the name of wisdom absolutely speaking {simpliciter) , the remaining branches of philosophy only relatively or from a particular point of view {secundum quid) Conclusion I.—Philosophy is the science which by the natural light of reason studies the first causes or highest principles of all things—is, in other words, the science of things in their first causes, in so far as these belong to the natural order. The difficulty of such a science is proportionate to its elevation. That is why the philosopher, just because the object of his studies is the most sublime, should personally be the humblest of students, a humility, however, which should not prevent his defending, as it is his duty to do, the sovereign dignity of wisdom as the queen of sciences. The perception that the sphere of philosophy is 1 Strictly speaking, there is no one formal object of philosophy, since philosophy as a whole is not simply one, but a compound of several distinct sciences {logic, natural philosophy, metaphysics, etc.), each specified by a distinct formal object {ens rationis logicum, ens mobile, ens in quantum ens—cf. Part II). But between the formal objects of the different philosophic sciences there is something analogously common the fact that they study, each in its own order, the highest and most universal causes, and treat their subject-matter from the standpoint of these causes. We may therefore say that the highest causes constitute the final object or the formal standpoint analogously common of philosophy taken as a whole, 1 08
  • 115. DEFINITION OF PHILOSOPHY universal led Descartes (seventeenth century) to regard philosophy as the sole science ^ of which the others were but parts ; Auguste Comte, on the contrary, and the positivists generally (nineteenth century), sought to absorb it in the other sciences, as being merely their " systematisation." It is evident that the cause of both errors was the failure to distinguish between the material and formal object of philosophy. Philosophy of Aristotle and St. Thomas Philosophy and the corpus of other sciences have the same material object (everything knowable). But the formal object of philosophy is first causes, of the other sciences secondary causes. Descartes Auguste Comte Philosophy absorbs the The sciences absorb other sciences—is the whole philosophy—there is no of science. philosophy. We said above that philosophy is a science, and that it attains certain knowledge. By this we would not be understood to claim that philosophy provides certain solutions for every question that can be asked within its domain. On many points the philosopher 1 Descartes used the term " philosophy " in its ancient sense. For the ancients, as for Descartes, the word denoted the entire body of scientific knowledge. But the ancients divided philosophy thus under-stood into several distinct sciences, among which metaphysics was distinguished as in the fullest sense philosophy. Descartes, on the contrary, regarded philosophy, still understood as the entire body of scientific knowledge, as a science specifically one (of which meta-physics, physics, mechanics, medicine, and ethics were the principal subdi.isions). He thus recognised only one science. In our view philosophy is a body of sciences which owes its imity and distinction from the other sciences to its formal standpoint (first causes). The leading member of this body of sciences is metaphysics, a science specifically one, whose formal object is luiiversal (being qua being). 109
  • 116. AN INTRODUCTION TO PHILOSOPHY must be content with probable solutions, either because the question goes beyond the actual scope of his science, for example in many sections of natural philosophy and psychology, or because of its nature it admits only of a probable answer, for example the application of moral rules to individual cases. But this element of mere probability is accidental to science as such. And philosophy yields a greater number of certain conclusions, and of those many more perfect, namely, the conclusions of metaphysics, than any other purely human science. no
  • 117. VI PHILOSOPHY AND THE SPECIAL SCIENCES W^E have now to define the relationship between philosophy (particularly the first philosophy or meta-physics) and the other sciences. Every science is mistress in her own house, inas-much as every science possesses the indispensable and sufficient means of attaining truth within its own sphere and no one is entitled to deny the truths thus proved. A science, however, or rather a scientist, may happen to make a mistake in its own domain. In such a case the science in question is no doubt com-petent to judge and correct itself, but it is obvious that a superior science has also the right to judge and correct it, if the mistake should contradict one of its own results and thus come under its jurisdiction. But philosophy, and especially philosophy in the highest sense, that is metaphysics, is the sovereign science. Therefore it is competent to judge every other human science, rejecting as false every scientific hypothesis which contradicts its own results. Take for example an hypothesis of physics which appears to contradict a truth of philosophy.^ Physics is competent to judge that hypothesis by the laws of 1 It may, for instance, be questioned whether the law of inertia, as formulated since Galileo and Descartes, can be reconciled with tlic axiom of philosophy : quidquid movetur ab alio movetur. Ill
  • 118. AN INTRODUCTION TO PHILOSOPHY physics, but philosophy is also competent to judge it by the principles of philosophy, determining whether and how far it really contradicts the philosophic truth in question. (If the contradiction is real, it is evident that the hypothesis of physics in question must be false, for one truth cannot contradict another. The physicist must therefore bow to the verdict of philosophy, revise his arguments and make further experiments.^) Let us now take a conclusion of philosophy which appears to contradict a truth estabhshed by physics.^ It is for philosophy to judge that conclusion in accord-ance with the principles of philosophy, to decide whether and how far it is really in conflict with the physical truth in question. But physics is incompe-tent to determine the question by the principles and data of physics. (If the contradiction is real, it is obvious that the alleged conclusion of philosophy is false, for one truth cannot contradict another. The philosopher will therefore bow, not indeed to the verdict of physics, but to the verdict of philosophy judging itself by means of physics, and will revise his arguments accordingly.) * It is true, no doubt, that we have actually to do, not with philosophy, but with philosophers, and that philosophers are fallible, and aphilosopher may therefore be mistaken in judging an hypothesis of physics, but this does not prove that he has no right to judge it. A physicist may therefore be justified in a particular case in maintaining an hypothesis of physics agidnst a philosopher who asserts that it contradicts a truth of philosophy. But that is because the evidence he possesses in support of his hypothesis convinces him that the philosopher is mistaken in his verdict—in other words, has not given it ut philosophus, as a mouthpiece of philosophy. But he would not therefore be justified in denying the philosopher's competence as such to determine the question. 2 For example, when the philosophical doctrine oijree will appears to the mechanists to contradict the physical law of the conservation ofenergy. 112
  • 119. PHILOSOPHY AND THE SPECIAL SCIENCES Moreover, since the laws of one science are subordi-nate to the laws of a superior science, it clearly follows that it is the office of the superior science to govern the inferior. But since the principles of philosophy (the first philosophy or metaphysics) are the absolutely first principles of all human knowledge, they possess an authority over the principles of all other human sciences, which are in a certain sense dependent upon them. That is to say, philosophy (the first philosophy or metaphysics) governs the other sciences. Since the principles of philosophy (the first philo-sophy or metaphysics) are the absolutely first prin-ciples of all human knowledge, the principles or postulates of all human sciences are in a certain sense dependent upon them. They do not, it is true, depend directly on the principles of metaphysics, as the truth of a conclusion depends on the truth of its premisses. They are self-evident by the light of natural reason [principia per se nota) . But they are not absolutely speaking {simpliciter) first principles. Therefore, although they carry con-viction independently of metaphysics, nevertheless they presuppose in fact the principles of metaphysics and can be resolved into them. They can be known without an exphcit knowledge of the principles of metaphysics, but they could not be true, unless the latter were true. And in this sense they are indirectly subordinate to the latter. For instance, the mathe-matical axiom, two quantities which are equal to a third quantity are equal to one another^ can be resolved into the metaphysical axiom of which it is a special case : two beings identical with a third are identical with one another. It is for this resison that all the sciences are said to 113
  • 120. AN INTRODUCTION TO PHILOSOPHY be indirectly subordinate to metaphysics. Moreover, they are obliged on occasion to employ the universally valid principles of metaphysics. In this sense they are said to be subordinate to metaphysics in a particular aspect or relatively {secundum quid). To govern or direct anything is to prescribe its object or end. The sciences are not directed by philo-sophy to their end, in the sense that they cannot attain it without the aid of philosophy. Arithmetic, for example, has no need of philosophy to investigate the numerical truths which it investigates of its very nature. Philosophy, however, assigns the distinctive ends of the special sciences in the sense that it deter-mines speculatively the distinctive object of each, and what constitutes their specific unity and differen-tiation from the rest (classification of the sciences).^ And so doing it assigns the order in which they stand one to another. Thus all the sciences are ordered by wisdom : sapientis est ordinare. If therefore a science, or more correctly a scientist, should happen to lose sight of its true object by encroaching on the domain and rights of another science,* it is the duty of philosophy to redress the disorder involved. In this capacity philosophy governs or directs the sciences * A problem discussed in major logic. 2 Such divagations and encroachments are only too frequent. For example, the non-EucIidcan geometries may be so treated as to divert mathematics from its proper end. On the other hand, since Descartes, mathematics has usurped the domain of all the sciences, and in our time physics and chemistry are constantly encroaching upon the domain of biology, medicine upon psychology ; while the incursions made by physics or biology into the province of philosophy itself are innumerable : for example, the pseudo-scientific theories of the non-existence affinal causes, the unreality of qualities, determinism, atomism, or the biological dogmas of fransformism and mechanism. 114
  • 121. PHILOSOPHY AND THE SPECIAL SCIENCES (to these distinctive ends), not by positive prescription, but by setting them right, if they transgress their boundaries. On the other hand, that knowledge in which the mind attains its ultimate good, that is to say, the highest knowledge, may be regarded as the common transcendent goal towards which all the special sciences converge. But this knowledge is bestowed by philo-sophy— the science of first causes—which in this capacity governs or directs the sciences in view of the common end to which their particular objects are subordinate. All the sciences are thus directed to wisdom. From all we have just said it follows that to be proficient in the sciences it is not necessary to be a philosopher or to base one's work on a philosophy ; neither need the scientist while engaged in his special task seek advice from the philosopher or attempt to play the philosopher himself; but " philosophy alone enables the man of science to understand the position and bearings of his special science in the sum-total of human knowledge " or "to acquire a notion either of the principles implicit in all experimental knowledge or the true foundations of the special sciences." ^ It follows, further, that a period in the history of human culture in which philosophy is not allowed her rightful suzerainty over the sciences as scientia rectrix ° inevitably ends in a condition of intellectual chaos and a general weakening of the reasoning faculty. Descartes, just because he absorbed all the sciences 1 T. Richard, Philosophic du raisonnement dans la science d'apris Saint Thomas, p. 14.. ' St. Thomas, In Metaph., Inlroduction.
  • 122. AN INTRODUCTION TO PHILOSOPHY in philosophy, and regarded science as absolutely and without quaUfication one, believed that the prin-ciples of all the sciences depend directly on the principles of the first philosophy (metaphysics). In consequence he held that the study of the sciences and of philosophy itself must begin with metaphysics, that is to say, with the coping-stone of the entire edifice. The contrary error is the belief that the principles of science are absolutely independent of the principles of philosophy. There is therefore no place for a scientia rectrix, and the sciences are no longer a structure built on a definite plan, but a formless agglomeration. It is surprising that Auguste Gomte, who wished to reduce philosophy to a mere systematisation of the special sciences, failed to see that this very function of classifying and systematising the sciences (in what he terms their objective synthesis) is only possible if philo-sophy is a distinct science of a higher order on which the others are in a certain respect dependent.^ Philosophy of Aristotle and St. Tfiomas The principles of the special sciences are subordinate to the principles of philosophy, but only indirectly. Philosophy therefore governs the other sciences, but its government is such that it may be termed constitutional. (The special sciences are autonomous.) The study of the first philosophy (metaphysics) should be undertaken, not at the beginning, but at the end of intellectual research. Philosophy of Philosophy of those who reject Descartes Philosophy The principles of the The principles of the special sciences arc directly special sciences are not subordinate to those of subordinate to the prin- * For Gomte, indeed, sociology takes the position of scientia rectrix, only, however, as ordering the sciences in reference to the human subject, not in themselves. {Subjective synthesis.) 1x6
  • 123. PHILOSOPHY AND THE SPECIAL SCIENCES philosophy. The latter ciples of any science of a therefore exercises over the higher order. These sci-other sciences a govern- ences therefore are in no ment v^hich may be termed sense governed but are in despotic. a condition which may be The study of the first termed anarchy. There is philosophy (metaphysics) no supreme science or first should be undertaken at philosophy (metaphysics), the beginning of intellec-tual research. Finally, if a science bases its demonstrations on certain postulates or data, which it can neither explain nor defend, there must be a superior science whose function it is to defend these postulates or data. In this sense the science of architecture defends that of building. It is, however, obvious that every science, except the highest, bases its demonstrations on postu-lates or data it is incapable of explaining or defending. For instance, mathematics does not inquire what is the nature of quantity, number, or extension, nor physics what is the nature of matter. And if an objector should deny that the sensible world exists, that two quantities equal to a third are equal to one another, or that space has three dimensions, neither physics nor mathematics can refute his objection, since they on the contrary assume these postulates or data. There-fore it must be the function of philosophy (the first philosophy or metaphysics) to defend against every possible objection the postulates of all the human sciences. It is from common sense, or from the natural evidence of the intellect and experience, that the sciences derive their postulates. This is no doubt their sufficient warrant to build on these postulates, but it is insufficient to safeguard and protect them against 117
  • 124. AN INTRODUCTION TO PHILOSOPHY errors which call the postulates in question. And it is also insufficient to provide for the perfecting stability and essential needs of human knowledge. Human knowledge would remain excessively imperfect and weak, and would fail to reach its final end, if the postulates of the sciences were not scientifically explained, discussed, and defended. Philosophy, therefore, and particularly the first philosophy or metaphysics, because it is wisdom and the supreme science, judges, governs, and defends the other sciences. But the ruler is certainly not dependent upon those whom he governs. We therefore conclude that philosophy is independent of the inferior sciences, or at any rate depends on them only in the sense that a superior, when he is not strong enough to be self-sufficient, depends on the servants or instruments which he employs. It was for this reason that Aristotle regarded philosophy as the science pre-eminently free. Philosophy appeals to the facts, the data of experi-ence. To obtain the necessary materials it uses as instruments the truths provided by the evidence of the senses and the conclusions proved by the sciences. It depends on both, as a superior who cannot do his own work depends on the servants he employs. A dependence of this kind is a purely material dependence, since the superior depends on the inferior to be served by him, not to do him service. He therefore judges by his own light of whatever his servants bring him to supply his needs. For example, one of the most successful students of bees, Francois 1x8
  • 125. PHILOSOPHY AND THE SPECIAL SCIENCES Huber, who was blind, interpreted by the light of his intellect the facts seen by his servants' eyes. But further, this purely material dependence of philosophy, though absolutely necessary in respect of the evidence of the senses, is relative and contingent in respect to the conclusions of the sciences. It is in fact from the evidence of the senses that philosophy derives the fundamental principles which—interpreted by its own Hght—it employs as premisses in its demon-strations and as the means to prove its special truths. For instance, the truth, perceived by the senses and interpreted by the light of philosophy, there is motion in the universe, served Aristotle as the premiss from which he proved that being is divided into act and potentiality, and that there is a first mover that is pure act (God). It is obvious that philosophy is absolutely unable to dispense with data of this kind, and that the data thus employed as premisses must be absolutely true. Otherwise the conclusions which philosophy deduces from them would be uncertain. But it is otherwise with the propositions and conclusions of the sciences. No doubt these conclusions, if true, contribute to the store of materials utilised by philo-sophy— but philosophy (and particularly the first philosophy or metaphysics) is under no necessity to employ them, indeed ought not to employ them to establish its own conclusions, at any rate not its certain conclusions, though it may make use of them as confirmatory evidence. It must certainly have at its disposal some scientific conclusions, indeed as plentiful a supply as possible, because it cannot develop its principles satisfactorily until it sees them embodied, so to speak, in concrete examples which 119
  • 126. AN INTRODUCTION TO PHILOSOPHY the senses can perceive. But it does not need one particular scientific proposition rather than any other, provided, that is to say, that, true to its own nature and maintaining the Hberty due to a superior science, it draws its proofs from its own principles and from the fundamental truths supplied by the evidence of the senses and not from the conclusions supplied by the sciences. These latter, in fact, should not be premisses but simply illustrations which assist philo-sophy to attain its own truths. A sound philosophy can therefore dispense with the particular system of scientific explanations of which it makes use in accord-ance with the state of science at a particular epoch, and if that system were one day proved false the truth of that philosophy would not be affected. Only its language and the sensible illustrations with which it clothed its truths would require modification. These remarks are important. They show how the datum of experience on which philosophy is primarily based suflSces for the requirements of a supreme and universal science. This datum is provided by an instrument—the evidence of the senses—earlier than scientific observation, infinitely more certain than the inductions of the sciences, and placed by nature at the disposal of every man, and consists of truths so simple that they are universally and absolutely valid, so immediate and evident that their certziinty exceeds that of the best established scientific conclusions.^ ^ To this fundamental datum we may add—but as secondary matter and at times valuable confirmation—the facts of a more special description discovered, controlled, and measured by the observations and experiments of science. We should bear in mind that the absolutely evident truths which constitute the primitive and funda-mental datum of pliilosophy must be carefully distinguished from 120
  • 127. PHILOSOPHY AND THE SPECIAL SCIENCES From what has been said we can also understand why the purely scientific mistakes to be found in older statements of AristoteUan and Thomistic philo-sophy, statements which inevitably bear the stamp of the scientific beliefs of their period, do nothing to discredit the truth of that philosophy. For no philo-sophy has observed more faithfully than that of Aristotle and St. Thomas the laws of thought which guarantee its purity and freedom. On the other hand philosophy, though distinct from the special sciences, is not unrelated to or isolated from them. On the contrary, it possesses the duty to exercise its office as scientia rectrix by con-stantly throwing its light on the discoveries and hypotheses, the unceasing activity and the development of the sciences, and one of the most essential requisites for its life and progress in the world is to maintain an intimate contact with the lower branches of study whose data it interprets and renders fruitful. To the extent to which philosophy thus concerns itself with interpreting by the aid of its own truths the facts or hypotheses which positive science regards as proved, the errors or lacunae of positive science may introduce accidentally into a true system of pliilosophy elements of error which are, so to speak, the token and price of the human development of philosophy—but certain interpretations of experience drawn from unscientific observation which are nothing more than pseudo-scientific statements. If, for example, in natural philosophy, to prove the reality of substantial change, it were argued that whereas water is a liquid body, hydrogen and oxygen are gaseous bodies, the argument would rest, not on a truth attested by the senses, but on a scientific error, for in reality the same inorganic bodies are found in the three states solid, liquid, and gaseous. Obviously an adequate scientific training helps the philosopher to avoid pitfalls of this kind. 121
  • 128. AN INTRODUCTION TO PHILOSOPHY they can only falsify a philosophy itself to the extent to which it is untrue to its own nature and enslaves itself to the lower branches of study.^ It is clear from everything which has been said that the nature and needs of philosophy make it incumbent upon the philosopher to keep himself as fully ac-quainted as he can with the scientific knowledge of his period, provided, however, that he preserves intact the freedom of philosophic truth. For though the philosopher as such need not use the affirmations of the special sciences to establish his own truths, he ought to make use of them, (i) to illustrate aptly his principles, (ii) to confirm his conclusions, (iii) to interpret, throw Hght upon, and assimilate, the assured results of the sciences so far as questions of philosophy are involved. And finally he should use the affirmations of science (iv) to refute objections and errors which claim support from its results. From yet another point of view the study of the sciences is necessary for the philosopher : his own education must of necessity, owing to the very condi-tions ofhuman nature, be a progress from the imperfect to the perfect, so that before undertaking the study of wisdom he should undergo the training of the sciences.* 1 The '* crime " of the decadent Scholastics of the sixteenth and seventeenth centuries was that they beUeved, and made others believe, that the philosophy of Aristotle and St. Thomas was in this sense bound up with the mistakes of ancient science, of which it is in reality wholly independent. 2 The philosopher also requires a scientific training to be in a position to distinguish readily between the primary evidence of experience and certain popular but really pseudo-scientific interpretations of experience, such, for example, as the hypothesis of the sun's motion around the earth or the over-hasty belief that a particular inorganic body is essentially liquid and another essentially solid or gaseous. 122
  • 129. PHILOSOPHY AND THE SPECIAL SCIENCES It is not therefore surprising that all the great philosophers have been thoroughly acquainted with contemporary science. Some have even been great scientists (for example Aristotle, Albertus Magnus, and Leibniz), and several scientific discoveries of the first magnitude has been made by philosophers, for instarrce the mathematical discoveries of Pythagoras, Descartes, and Leibniz. In this connection we may observe that a profound and practical knowledge of a single science with which the student is directly acquainted contributes far more to a philosophic training than a superficial and second-hand knowledge of a large number. Though owing to the degree to which specialisation has been carried in modern times he cannot hope ever to possess that complete knowledge of all the sciences which is possessed by the scientist in his particular department, the philosopher should nevertheless aim at acquiring a sufficiently thorough knowledge of the entire body of the sciences, an ideal in itself not beyond the bounds of possibility, as is proved by the example of several powerful minds. Conclusion II.—Philosophy is the highest of all branches of human knowledge and is in the true sense wisdom. The other (human) sciences are subject to philosophy, in the sense that it judges and governs them and defends their postulates. Philosophy on the other hand is free in relation to the sciences, and only depends on them as the instruments which it employs. 123
  • 130. VII PHILOSOPHY AND THEOLOGY — Philosophy is the highest of the human sciences, that is, of the sciences which know things by the natural light of reason. But there is a science above it. For if there be a science whch is a participation by man of the knowledge proper to God himself, obviously that science will be superior to the highest human science. Such a science, however, exists ; it is theology. The word theology means the science of God. The science or knowledge of God which we can attain naturally by the unassisted powers of reason, and which enables us to know God by means of creatures as the author of the natural order, is a philosophic science—the supreme department of metaphysics and is known as theodicy or natural theology. The know-ledge or science of God which is unattainable naturally by the unassisted powers of reason, and is possible only if God has informed men about himself by a revelation from which our reason, enlightened by faith, subsequently draws the implicit conclusions, is supernatural theology or simply theology. It is of this science that we are now speaking. Its object is something wholly inaccessible to the natural apprehension of any creature whatsoever, namely, God known in himself, in his own divine life, or in technical language sub ratione Deitatis, not, as in 124
  • 131. PHILOSOPIiY AND THEOLOGY natural theology, God as the first cause of creatures and the author of the natural order. And all theo-logical knowledge is knowledge in terms of God thus apprehended, whereas all metaphysical knowledge, including the metaphysical knowledge of God, is knowledge in terms of being in general. The premisses of theology are the truths formally revealed by God {dogmas or articles of faith) , and its primary criterion of truth the authority of God who reveals it. Its light is no longer the more natural light of reason, but the Ught of reason illuminated by faith, virtual revelation in the language of theology, that is to say, revelation in so far as it implicitly (virtually) contains whatever conclusions reason can draw from it. Alike by the sublimity of its object, the certainty of its premisses, and the excellence of its light, theology is above all merely human sciences. And although it is unable to perceive the truth of its premisses, which the theologian beHeves, whereas the premisses of philosophy are seen by the philosopher, it is never-theless a science superior to philosophy. Though, as St. Thomas points out, the argument from authority is the weakest of all, where human authority is con-cerned, the argument from the authority of God, the revealer, is more solid and powerful than any other.^ And finally as the object of theology is he who is above all causes, it claims with a far better title than * Licet locus ab auctoritate, quae fundaiur super ratione humana, sit infirmissimus, locus tarrum ab auctoritate quae fundatur super revelatione divina est efficacissimus. St. Thomas, Sum. Theol. i, q. i, a 8, ad 2.
  • 132. AN INTRODUCTION TO PHILOSOPHY metaphysics the name of wisdom. It is wisdom par excellence.^ What relations, then, must obtain between philosophy and theology ? As the superior science, theology jW^^j philosophy in the same sense that philosophy judges the sciences.^ It therefore exercises in respect of the latter a function of guidance or government, though a negative govern-ment, which consists in rejecting as false any philosophic affirmation which contradicts a theological truth. In this sense theology controls and exercises jurisdiction over the conclusions maintained by philosophers. Thej&r^mfjj^j^ of philosophy, however, are independent of theology, being those primary truths which are self-evident to the understanding, whereas the premisses of theology are the truths revealed by God. The premisses of philosophy are self-supported and are not derived from those of theology. Similarly the light by which philosophy knows its object is independent * Theology is theoretical wisdom, par excellence the wisdom which knows God by the intellect and its ideas, that is to say, by the normal processes of human knowledge. There is another wisdom of a still higher order which is a gift of the Holy Spirit, and enables lis to know God experimentally and by means of charity. It enables xis to judge of divine things instinctively, as the virtuous man judges of virtue {per modum inclinationis) , not scientifically as the moralist judges of virtue {per modum cognitionis) . Cf. St. Thomas, Sum. Theol. i, q. i, a 6, a</ 3. 2 See above, pp. 1 1 1 sqq. The philosopher and the scientist are never entitled to deny the rights which theology possesses over philosophy and the sciences. They may, however, be justified in rejecting in a particular instance, not indeed the authority of the Church, but the judgment of an individual theologian, since the individual theologian does not necessarily speak as the mouthpiece of theology, and may therefore be mistaken. 126
  • 133. PHILOSOPHY AND THEOLOGY of theology, since its light is the light of reason, which is its own guarantee.^ For these reasons philosophy is not positively governed by theology,^ nor has it any need of theology to defend its premisses (whereas it defends those of the other sciences). It develops its principles autonomously within its own sphere, though subject to the external control and negative regulation of theology. It is therefore plain that philosophy and theology are entirely distinct, and that it would be as absurd for a philosopher to invoke the authority of revelation to prove a philosophical thesis as for a geometrician to attempt to prove a theorem by the aid of physics, for example, by weighing the figures he is comparing. But if pliilosophy and theology are entirely distinct, they are not therefore unrelated, and although philo-sophy is of all the human sciences pre-eminently the free science, in the sense that it proceeds by means of premisses and laws which depend on no science superior to itself, its freedom—that is, its freedom to err—is Umited in so far as it is subject to theology, which controls it externally. I In the seventeenth century the Cartesian reform ^ This light is its own evidence and in philosophy is sufficient of itself. But this does not prevent it serving also—in theology, however, not in philosophy—as the instrument of a superior light ; neither, of course, does it imply that human reason is not subordinate in its very principles to the First Intellect. 2 Theology can turn the investigations of philosophy in one direction rather than in another, in which case it may be said to regulate philosophy positively by accident {per accidens) . But absolutely speaking theology can regulate philosophy only negatively, as has been explained above. Positively it does not regulate it either directly, by furnishing its proofs (as faith for apologetics), or indirectly, by classifying its divisions (as philosophy itself classifies the sciences). 127
  • 134. AN INTRODUCTION TO PHILOSOPHY resulted in the severance of philosophy from theology,^ the refusal to recognise the rightful control of theology and its function as a negative rule in respect of philo-sophy. This was tantamount to denying that theology is a science, or anything more than a mere practical discipline, and to claiming that philosophy, or human wisdom, is the absolutely sovereign science, which admits no other superior to itself. Thus, in spite of the religious beliefs of Descartes himself, Cartesianism introduced the principle of rationalist philosophy, which denies God the right to make known by revela-tion truths which exceed the natural scope of reason. For if God has indeed revealed truths of this kind, human reason enlightened by faith will inevitably employ them as premisses from which to obtain further knowledge and thus form a science, theology. And if theology is a science, it must exercise in respect of philosophy the function of a negative rule, since the same proposition cannot be true in philosophy, false in theology.^ * It may, it is true, be replied that Descartes's intention was simply to emancipate philosophy from the authority of a particular theological system—Scholasticism—which he regarded as worthless, because it took its philosophic or metaphysical principles from Aristotle. In reality, however, it was with theology itself that he broke, when he broke with Scholasticism, which is the traditional theology of the Church. And moreover his conception of science implied the denial of his scientific value of theology. In any case the result of his reform was the assertion of the absolute independence of philosophy in relation to theology. {Cf. Blondel, " Le Christianisme de Descartes." Revtie de Mitaph. et de Morale, 1896.) 2 The theory of a double truth, by which the same thing may be true in philosophy, but fake in theology, was invented by the mediaeval Averroists, who sought in this way to evade the censures of the Church. In various forms it has been revived in modem times by all who, like the modernists, wish to keep the name of Catholics while freely professing in philosophy opinions destructive of some particular dogmatic truth. 128
  • 135. PHILOSOPHY AND THEOLOGY On the other hand, philosophy renders to theology services of the greatest value where it is employed by the latter. For in fact theology employs in its demonstrations truths proved by philosophy. Philo-sophy thus becomes the instrument of theology, and it is in this respect and in so far as it serves theological argument that it is called ancilla theologiae. In itself, however, and when it is proving its own conclusions, it is not a bond-servant but free, subject only to the external control and negative ruUng of theology. As was shown above, philosophy is from the very nature of things obhged to employ as an instrument the evidence of the senses, and even, in a certain fashion, the conclusions of the special sciences. Theo-logy* considered in itself as a science subordinate to the knowledge of God and the blessed, is not in this way obliged to make use of philosophy, but is absolutely independent. In practice, however, on account of the nature of its possessor, that is to say, on account of the weakness of the human understanding, which can reason about the things of God only by analogy with creatures, it cannot be developed without the assistance of philo-sophy. But the theologian does not stand in the same relation to philosophy as the philosopher to the sciences.^ We have seen above that the philosopher ^ This distinction between the relationship of theology to philosophy and that of philosophy to the special sciences derives from the fact that, since theology is a participation of the divine wisdom, the human subject is too weak for its unaided study and to draw conclusions from it is comj)elled to employ as premisses conclusions established by an inferior discipline. Since, however, philosophy is a human wisdom, at which reason can 129
  • 136. AN INTRODUCTION TO PHILOSOPHY should employ the propositions or conclusions which he borrows from the sciences, not to estabUsh his own conclusions (at any rate not conclusions for which metaphysical certainty is claimed), but merely to illustrate his principles, and therefore that the truth of a metaphysical system does not depend on the truth of the scientific material it employs. The theologian, on the contrary, makes use at every turn of philosophic propositions to prove his own conclu-sions. Therefore a system of theology could not possibly be true if the metaphysics which it employed were false. It is indeed an absolute necessity tiiat the theologian should have at his disposal a true philosophy in conformity with the common sense of mankind. Philosophy taken in itself normally precedes theo-logy. Certain fundamental truths of the natural order are indeed what we may term the introduction to the faith {praeambula Jidei). These truths, which are naturally known to all men by the light of common sense, are known and proved scientifically by philo-sophy. Theology', being the science of faith, pre-supposes the philosophical knowledge of these same truths. Philosophy considered as the instrument of theology serves the latter, principally in three ways. In the first place theology employs philosophy to prove the truths which support the foundations of the faith in that department of theology which is termed apolo-arrive, though with difficulty, by its unassisted natviral power, the human mind should be able to draw from it certain conclusions (especially metaphysically certain conclusions) without employing as premisses the conclusions of sciences to which it is superior in d^[nity and in certainty. 130
  • 137. PHILOSOPHY AND THEOLOGY 1 getics,^ which shows, for example, how miracles prove the divine mission of the Church ; secondarily to impart some notion of the mysteries of faith by the aid of analogies drawn from creatures—as for instance when theology uses the philosophic conception of verbum mentale, the mental word,' to illustrate the dogma of the Trinity ; and finally to refute the adversaries of the faith—as when theology shows by means of the philosophic theory of quantity * that the mystery of the Eucharist is in no way opposed to reason. We must not forget that, if philosophy serves theology, it receives in return valuable assistance firom the latter. In the first place, so far as it is of its nature subject to the external control and negative ruhng of theology, it is protected from a host of errors ; and if its freedom to err is thus restricted, its freedom to attain truth is correspondingly safeguarded.* In the second place, in so far as it is the instrument of theology, it is led to define more precisely and with more subtle refinements important concepts and theories which, left to itself, it would be in danger of neglecting. For example, it was under the influence See Garrigou-Lagrange, De Revelatione, i, 2, 19 18. 2 A theory studied in psychology. ' An explanation given by cosmology. * Unassisted reason can indeed avoid error on any particular point whatsoever within the sphere of philosophy, but in view of the weakness of human nattire it is unable without the assistance of grace to avoid error on some point or other ; that is to say, without a special grace or the negative control of revelation and theology it cannot achieve a perfect human wisdom. {Cf. St. Thomas, Sum. Theol., i, q. i, a. I ; Sum. contra Cent., i, 4 ; Garrigou-Lagrange, De Rev., i, pp. 41 sqq.)
  • 138. AN INTRODUCTION TO PHILOSOPHY of theology that Thomism elaborated the theory of nature and personality, and perfected the theory of the habitus, habits, etc. Conclusion III.—Theology, or the science of God so far as He has been made known to us by revelation, is superior to philosophy. Philosophy is subject to it, neither in its premisses nor in its method, but in its conclusions, over which theology exercises a control, thereby constituting itself a negative rule of philosophy. 132
  • 139. VIII PHILOSOPHY AND COMMON SENSE Before wc know things with a scientific or perfect knowledge by reflecting upon them and by their causes, we know them imperfectly {unscientific know-ledge, the knowledge of everyday life). We must remember that we are obliged not only to begin with this unscientific knowledge of everyday life ; we must be content with it to the end, improving it more or less by study and reading, in that enormous number of cases where science in the strict sense is unattain-able. For, so far as the knowledge of secondary causes is concerned, no man can possibly attain, with the per-fection required of the genuine scientist, universal knowledge ; in other words, he cannot specialise in all branches of science, a contradiction in terms. He is fortunate, indeed, if he can make himself master of a single science. For all the others he must be satisfied with a knowledge which, however enriched and improved it may be in the case of what is known as a cultivated man, that is to say, a man well acquainted with the scientific knowledge of other people, is always inferior to science in the strict sense. But in the domain of first causes, the science of all things is within a man's grasp, for it is precisely the distinguishing character of the science called philosophy to know 133
  • 140. AN INTRODUCTION TO PHILOSOPHY all things by their first causes/ and it is to the philo-sopher or the sage, the wise man, that we have the right to apply Leonardo da Vinci's aphorism : facile cosa e farsi universale ; it is easy for a man to make himself universal. Ordinary knowledge- consists for the most part of mere opinions or beliefs, more or less well founded. But it implies a solid kernel of genuine certainties in which the philosopher recognises in the first place data of the senses (for example, that bodies possess length, breadth, and height), secondly, self-evident axioms (for example, the whole is greater than the part, every event has a cause, etc.), and thirdly, consequences immediately deducible from these axioms (proximate conclusions). These certainties which arise spontaneously in the mind when we first come to the use of reason are thus the work of nature in us, and may therefore be called an endowment of nature * as proceeding from the natural perception, consent, instinct, or natural sense of the intellect. Since their source is human nature itself, they will be found in all men aUke ; in other words, they are common to all men. They may therefore be said to belong to the common perception, consent, or instinct, or to the common sense of mankind. The great truths without which man's moral life is impossible—for example, knowledge of God's existence, the freedom of the will, etc.—belong to this domain of ^ It is therefore obvious what a stupendous delusion is involved in the positivist view of philosophy. Were philosophy merely the co-ordination or systematisation of the sciences, its attainment would presuppose a perfect mastery of all the sciences, that is to say, specialisa-tion in every science, which amounts to saying that philosophy is beyond the reach of man. * Kleutgen, La philosophic scolastique, i, p. 439.
  • 141. PHILOSOPHY AND COMMON SENSE common sense, as consequences immediately deducible (proximate conclusions) from primary data appre-hended by observation and first principles apprehended by the intellect. All men, unless spoiled by a faulty education or by some intellectual vice, possess a natural certainty of these truths. But those whose understanding has never been cultivated are not able to give any account or at least any satisfactory account of their convictions ; that is to say, they cannot explain why they possess them. These certainties of common sense, conclusions of an implicit reasoning, are as well founded as the certainties of science. But their possessor has no knowledge, or an imperfect knowledge, of the grounds on which he bases them. They are therefore imperfect not in their value as truth but in the mode or condition under which they exist in the mind. Of the self-evident truths {the whole is greater than the part, every event has a cause, etc.) which are the object ofwhat is termed the understanding ofprinciples, and whose certainty is superior to that of any conclusion of science, common sense possesses a knowledge whose mode is equally imperfect, because it is confused and implicit. Common sense therefore may be regarded as the natural and primitive judgment of human reason, infalUble, but imperfect in its mode. The wholly spontaneous character of common sense, and its inability to give an account of its con-victions, have led certain philosophers to regard it as a special faculty purely instinctive and unrelated to the intellect (the Scottish school, end of eighteenth and beginning of nineteenth century ; Reid, Dugald Stewart, and in France, JoufFroy), or as a sentiment 135
  • 142. AN INTRODUCTION TO PHILOSOPHY . distinct from and superior to reason (the intuitive or sentimentalist school ; for instance, Rousseau, Jacobi, and in our own time Bergson). But in that case it would necessarily be blind, for we possess no other light than that of the intellect or reason. The light of common sense is fundamentally the same light as that of science, that is to say, the natural light of the intellect. But in common sense this light does not return upon itself by critical reflection, and is not perfected by what we shall learn to know as a scientific habit {habitus) We must now define the relations which obtain between philosophy and common sense. Philosophy cannot, as the Scottish school main-tained, be founded on the authority of common sense understood simply as the common consent or universal witness of mankind, or as an instinct which in fact compels our assent. For it is in fact founded on evidence, not on authority of any kind. But if by common sense we understand only the immediate apprehension of self-evident first principles, which is one of its constituents, we may say with truth that it is the source of the whole of philosophy. For the premisses of philosophy are indeed the evident axioms which in virtue of its natural constitution implant in the mind its primary certainties. It is important to be quite clear that, if philosophy finds its premisses already enunciated by common sense, it accepts them not because they are enunciated by common sense, or on the authority of common sense understood as the universal consent or common instinct of mankind, but entirely and solely on the authority of the evidence. 136
  • 143. PHILOSOPHY AND COMMON SENSE Finally, if we take into account the entire body of truths (premisses and conclusions) known by common sense with certainty but in an imperfect mode, we must conclude that philosophy is superior to common sense, as the perfect stage of anything (in this case the scientific stage of knowledge) is superior to the im-perfect or rudimentary stage of the same thing (in this case the pre-scientific stage of the same knowledge, which is yet true and certain at both stages). If in common sense we consider not the conclusions which it reaches but the premisses alone, it is still inferior to philosophy in respect of its mode of know-ledge, but superior alike to philosophy and to all the sciences in respect oi its object and of the light in which it knows. For, as we have said above, philosophy and all the sciences are ultimately founded on the natural evidence of first principles (to which philosophy returns—in criticism—to study them scientifically, whereas the other sciences are content to accept them from nature). Philosophy studies scientifically the three categories of truths to which common sense bears instructive witness : (i) the truths of fact which represent the evidence of the senses ; (ii) the self-evident first prin-ciples of the understanding, in as much as it clears up their meaning by critical reflection and defends them rationally ; (iii) the consequences immediately de-ducible (proximate conclusions) from these first principles, inasmuch as it provides a rational proof of them. And, further, where common sense yields to the mere opinions of popular belief, philosophy continues to extend indefinitely the domain of scientific certainty. Thus philosophy justifies and continues 137
  • 144. AN INTRODUCTION TO PHILOSOPHY common sense, as, for instance, the art of poetry justi-fies and continues the natural rhythms of language. It is also the province of philosophy to decide what are the genuine certainties affirmed by common sense, and what is their true significance ; a function which common sense is incapable of performing, for the very reason that it does not understand, or does not under-stand clearly, the grounds ofits knowledge. In this sense philosophy controls common sense, as, for example, the art of poetry controls the natural rhythms of language. Nevertheless, common sense has the right and duty to reject any philosophic teaching which denies a truth of which it possesses natural certainty, as the inferior has the right and duty to oppose a superior who acts in a manner evidently unjust. For as soon as a truth becomes known to us, by whatever channel, it is a sin not to accept it. Common sense may there-fore accidentally judge philosophy. It is related of Diogenes that when Zeno the Eleatic was arguing in his presence against the possibility of motion, his sole reply was to get up and walk. Simi-larly, when Descartes taught that motion is relative or " reciprocal," so that it makes no difference whether you say the moving object is moving towards the goal or the goal towards the moving object, the English philosopher Henry More retorted that when a man runs towards a goal panting and tiring himself,^ he has no doubt which of the two, the moving object or the goal, is in motion. These protests of common sense bzised on the evidence of the senses were perfectly justified. It must, however, be added that they were insufficient 1 Letter of March 5, 1649. 138
  • 145. PHILOSOPHY AND COMMON SENSE —not indeed to confute the respective theses of Zeno and Descartes but to confute them as errors in philo-sophy. That would have demanded a philosophic refutation of the arguments adduced by these philo-sophers, and explanations showing why and at what point they went wrong. It must be observed that though in itself and in order to establish its demonstrations philosophy does not depend upon the authority of common sense, understood as the universal consent or common instinct of mankind, nevertheless it is dependent upon it in a certain sense {materially, or in respect of the subject), in its origin as a human activity and in its development in the mind of philosophers. From this point of view philosophy may be compared to a building, and the great pre-scientific conclusions of common sense (the existence of God, the freedom of the will, etc.) to the scaffolding which nature has erected beforehand. Once the edifice has been completed it supports itself on its rock-bed, the natural self-evidence of its first principles, and has no need of scaffolding. But without the scaffolding it could not have been built. It is now evident how unreasonable that philosophy is, which priding itself on its scientific knowledge of things despises common sense a priori and on principle, and cuts itself off from its natural convictions. Descartes (who in other respects and in his very conception of science concedes too much to common sense) began this divorce, on the one hand, by admitting as the only certain truths those scientifically established, thus denying the intrinsic value of the convictions of common sense, and on the other hand, by professing as part of his system several doctrines 139
  • 146. AN INTRODUCTION TO PHILOSOPHY incompatible with those convictions. His disciple Malebranche, and above all the critical philosophers of the Kantian school, as also certain modernist philosophers, have carried this tendency to its extreme, until for some of these philosophers it is suflSdent that a proposition should be acceptable to common sense for it to be questioned or denied by science, which would be contaminated by the " credulity " of the common herd, unless it taught the contrary of what mankind at large believes to be true. Yet the greater the natural strength of a man's intelligence, the stronger should be his grasp of these natural certainties. He therefore who professes to condemn common sense shows not the strength but the weakness of his understanding. It is now obvious that in its attitude to common sense, as in its solution of the majority of the great philosophic problems, Thomism keeps the golden mean between two opposing errors Hke a mountain summit between two valleys. Philosophy of Aristotle and St. Thomas The convictions of common sense are valid, and science is untrue to itself if it rejects them. But the basis of philosophy is the natural witness of the intellect, not the authority of common sense. Rationalist, Critical, and Scottish School Modernist Schools Not only are the convic- Not only is the authority tions of common sense of common sense incapable valid, but the authority of of furnishing the basis of common sense imposing philosophy, but the con-itself as a blind instinct on victions of common sense the mind is the foundation are destitute of any specu-on which philosophy should lative value, be based. 140
  • 147. PHILOSOPHY AND COMMON SENSE From all that has been said it is evident what an important part the certainties of common sense play as an introduction to philosophy. Those who are beginning the study of philosophy and about to acquaint themselves with the most recent problems, and even perhaps the most misleading systems, ought to repose an absolute trust in the convictions of common sense of which they find their minds already possessed, for they will help them to rise to a higher and more perfect knowledge, conclusions scientifically established. Conclusion IV.—Philosophy is not based upon the authority of conunon sense understood as the universal consent or common instinct of mankind ; it is nevertheless derived fi-om common sense considered as the imderstanding of self-evident first principles. It is superior to common sense as the perfect or " scientific " stage of knowledge is superior to the imperfect or ordinary stage of the same knowledge. Neveitheless philosophy may be accidentally judged by common sense. For the purposes of this present outline we need only add that philosophy is not constructed a priori on the basis of some particular fact selected by the philosopher (Descartes's cogito), or principle arbitrarily laid down by him (Spinoza's substance, Fichte's pure ego, SchelUng's absolute, Hegel's idea) whose conse-quences he ingeniously develops. Its formal principles are the first principles apprehended in the concept of being, whose cogency consists wholly in their evidence 141
  • 148. AN INTRODUCTION TO PHILOSOPHY for the intellect,^ and on the other hand its matter is experience, and its facts * the simplest and most obvious facts—the starting-point from which it rises to the causes and grounds which constitute the ultimate explanation. Not a whimsy spun out of his own brain, but the entire universe with its enormous multitude and variety of data must be the philosopher's teacher. And he must always bear in mind that, if philosophy enables the human intellect to apprehend with absolute certainty the highest and most profound reaUties of the natural order, it cannot therefore claim to exhaust those realities by making them known to the utmost extent of their inteUigibiUty. From this point of view science does not destroy the mystery of things, that in them which is still unknown and unfathomed, but on the contrary recognises and deUmits it ; ' even what it knows it never knows completely. The wise man knows all things, inasmuch as he knows them in their ultimate causes, but he does not know, is infinitely removed from knowing, everything about everything. Ignorance, however, is not the same as error. It is sufficient for the philosopher that he knows with This is what the Positivists fail to see. 2 This is what the pure intellectttalists—from Parmenides to Hegel — who construct their metaphysics wholly a priori, have failed to grasp. ' Aristotle {Metapk., i, 2) remarks that the occasional cause of philosophy is ro Oavfidi^eiv, admiratio, by which he means wonder mingled with dread, in other words awe, a wonder which knowledge tends to remove. But we must be careful to understand his meaning of the wonder which does not understand, not of the admiration, indeed the awe, bom of understanding. The wise man is astonished at nothing because he knows the ultimate causes of all things, but he admires far more than the ignorant man. Cf, De Part. Arum., i, 5, 645 a 16 : ^c waffL rois <pvaiKoU ivearl u Oav/xaffrdv. 142
  • 149. PHILOSOPHY AND COMMON SENSE certainty what it is his province to know and what it is of the first importance for us to know. Indeed, it is better not to know things which divert the mind from the highest knowledge, as Tacitus remarks : nescire guaedam, magna pars sapientiae. We have considered the nature of philosophy ; it remains to distinguish its departments. We shall thus obtain a clear notion of its sphere, and at the same time become acquainted with its principal problems. 143
  • 151. PART TU'O THE CLASSIFICATION OF PHILOSOPHY
  • 153. THE MAIN DIVISIONS OF PHILOSOPHY When a man undertakes a work, he begins by testing his tool in various ways to learn the use he can and should make of it. The philosopher's work is to acquire knowledge ; his tool, reason. Therefore the philosopher before he begins his work must examine reason to discover the use he should make of it. The study of reason as an instrument of acquiring knowledge or means of discovering truth is called logic. Logic is therefore, strictly speaking, not so much a department of philosophy as a science or art, of which philosophy (and indeed all the sciences) makes use, and the introduction to philosophy. It is a propae-deutic to science.^ The other sciences are dependent upon logic inasmuch as it teaches the method of procedure in the acquisition of knowledge, and we are obUged to possess the means or tools of knowledge before we can acquire knowledge itself. It is thus evident that the study of philosophy must 1 " Res autem de quibus est logica, non quaeruntur ad cognoscendum propter seipsas, sed ut adminiculum quoddam ad alias scientias. Et idea logica non continetur sub philosophia speculativa quasi principalis pars, sed quasi quoddam reductum ad earn, prout ministrat speculationi sua instrumenta, scilicet syllogismos et dejinitiones, et alia hujusmodi, quibus in speculativis scientiis indigemus. Unde et secundum Boethium in Comment, sup. Porphyrium, non tarn est scientia quam scientiae instrumentum." St. Thomas, Sup. Boet. de Trin., q. 5, a. i, a^ 2. It is therefore only reductively that logic belongs to theoretical philosophy.
  • 154. AN INTRODUCTION TO PHILOSOPHY from the very nature of things begin with logic, although on account of its difficulty and extremely abstract character logic usually repels rather than attracts beginners.^ A few of the moderns rebel against this order of study, and maintain that logic should be studied only in the course of learning the other branches of philosophy, or after they have been learned. But this is like arguing that the surgeon should only study anatomy by the practice or after the practice of his art upon the sick, it is absurd, Aristotle remarks, to study at the same time a science and its conditions or method of procedure. axoTxov a^xa J^YITetv £7ti,cJTYi[i.ir)v xal XpOTTOV erutcTTYifXYjc;'.'' When by the study of logic he has made himself master of his tool, the philosopher can set to work. What that work is we know already : to acquire a knowledge of things by their first principles. If, however, we consider the aim of learning, there are two distinct types of knowledge. We can, for example, make use of our eyes simply in order to see • Cf. St. Thomas, Sup. Boet. de Trin., q. 6, a. i, ad 3. " Dicendum quod in addiscendo incipimus ab eo, quod est magisfacile, nisi necessitas aliud requirat. Quandoque enim necesse est in addiscendo non incipere ab eo quod estfacilius, sed ab eo a cujus cognitione cognitio sequentium dependet. Et hac positione oportet in addiscendo incipere a logica, non quia ipsa sit facilior scientiis ceteris ; habet enim maximam difficultatem, cum sit de secundo intellectis ; sed quia aliae scientiae ab ipsa dependent, in quantum ipsa docet modum procedendi in omnibus scientiis. Oportet enim primum scire modum scientiae quam scientiam ipsam, ut dicitur II. Metaph. 2 Metaph., ii, 995 a 12. St. Thomas, In II Metaph., 1. 5. "Quia enim diversi secundum diversos modos veritatem inquirunt, idea oportet quod homo instruatur per quern modum in singulis scientiis sint recipienda ea quae dicuntur. Et quia non est facile, quod homo simul duo capiat, sed dum ad duo attendit, neutrum capere potest : absurdum est, quod homo simul quaerat scientiam et modum qui corwenit scientiae. Et propter hoc debet prius addiscere logicam quam alias scientias, quia logica tradit communem modum procedendi in omnibus scientiis." 148
  • 155. THE MAIN DIVISIONS OF PHILOSOPHY and enjoy the sight of things, and we can also make use of them for the practical purposes of life. In the same way we can employ our reason scientifically, solely for the pleasure of knowledge. The sciences thus acquired exist solely for the sake of knowledge (the theoretical sciences). And if there be a theoretical science which seeks to account for things by their first principles, its object will be that which is the first principle in the theoretical order, namely, the first causes of everything which exist (that is to say, the first causes naturally knowable). That science is theoretical philosophy. We can, on the other hand, employ our reason scientifically for our profit and the improvement of our hfe ; the sciences thus acquired exist to procure by some kind of activity the good of man (the practical sciences). And if there be a practical science which seeks to regulate human acts by first principles, its object will be that which is the first principle in the practical order, namely, the absolute good of man (the absolute good naturally knowable).^ Such a science is practical philosophy—otherwise termed moral science or ethics.^ There are, indeed, other practical sciences besides ethics ; for example, medicine, which seeks to procure the health of man. But the object of these sciences is not good, pure and simple (the sovereign good), but some particular human good. They do not, * That is to say, the sovereign good of man as it would be, if his end were simply natural happiness. See below, pp. 265-267. * Observe that this division of philosophy into theoretical and practical relates to the end, the aim, not to the object itself of the science, which as such is always necessarily theoretical. Therefore it does not enter into the specification in the strict sense of the philosophic sciences. See below, p. 271.
  • 156. AN INTRODUCTION TO PHILOSOPHY therefore, refer in the practical order to which they belong to the first principle of action, and for that reason are not philosophies. Ethics, or moral science, is thus the only practical science which deserves the name of philosophy.^ We must bear in mind that although the object of ethics is to procure a good which is not solely a good of the intellect, that is to say, does not consist in know-ledge alone, its rule of truth is that which is, and it proceeds by way of demonstration, resolving conclusions into their premisses. In other words it is practical in virtue of its object (to know in order to procure the good of man in the ordering of his acts) , but as science in the strict sense it is theoretical knowledge.* We must further remark that the practical sciences are obviously subordinate to the theoretical, (i) as presupposing (if not in the order of their origin in 1 It may be added that of the practical sciences, only one, ethics, is in fact vere et proprie scieniia, that is to say, proceeds by way of demon-stration, in a necessary matter, and imparts a truth, which consists in knowing things in conformity with what is, and not in properly directing a contingent action. The other practical sciences (medicine, architec-ture, strategy, etc.) are arts, not sciences in the strict sense. {Cf. John of St. Thomas, Cursus Philos., i. Log. ii, q. i, a. 5). But though ethics is in the strict sense a science, it is for that very reason only in an indirect sense practical : for its procedure consists in providing knowledge {speculabiliter) not in producing action {opera-biliUr), and though it certainly supplies rules immediately applicable to concrete cases, the right application and good use of these rules in practice is the effect not of ethics but of the virtue of prudence. On the other hand, as we shall see later, the philosophy of art is also, in a sense, a practical philosophy. But it is very far from being a practical science, even hke ethics in the indirect sense, for it treats only of principles and is unable to descend to the rules immediately applicable to the concrete work to be executed. * Hence even practical philosophy is a theoretical wisdom which proceeds by way of knowledge (see above, p. 102). 150
  • 157. THE MAIN DIVISIONS OF PHILOSOPHY time, at least in the nature of things) the truths proved by these sciences, which they apply for the benefit of man—for example, medicine, as the science of heaUng, presupposes anatomy (ii) as sciences inferior ; in dignity to the theoretical sciences. For the latter are studied for their own sake, and are therefore good in themselves, whereas the practical sciences are studied for the good or utiUty of man, and are therefore good only in relation to that good or utility. It follows that philosophy in the strictest sense is theoretical philosophy (especially the first philosophy or metaphysics), logic being the science introductory to it, and ethics the science detached from it to treat specially of that which concerns the good of man. We are now in a position to define more exactly the object of these three main divisions of philosophy. A science which seeks to procure man's sovereign good must before all else treat of those things which are the indispensable conditions of its attainment. But these are the actions which man performs in the free exercise of his faculties ; in other words, human acts as such. We may therefore say that human acts are the formal object (subject-matter) of moral philosophy. A science whose aim is to know things by their first causes must treat primarily of that in things which depends immediately upon those causes. But that in things which depends immediately upon the first or highest causes is that which is most essential in them, their being, and that which is the most widely, indeed universally, distributed, being, which everything whatsoever possesses. We therefore conclude that the 151
  • 158. AN INTRODUCTION TO PHILOSOPHY formal object of theoretical philosophy is the being of things. Now theoretical philosophy studies the being of things in different fashions and from higher or lower points of view (degrees of abstraction) . It may study the being of things with their sensible properties {ens mobile) , or the being of things with the sole properties of quantity {ens quantum) , or the being of things with the sole properties of being (being qua being, ens in quantum ens) . Hence arise the three principal divisions of theoretical philosophy (see below, Ghs. Ill & IV). Finally, a science which studies reason as the tool for the attainment of truth must treat before all else of that which we handle or manipulate when we reason. But that which we handle or manipulate when we rezison is the things themselves. For example, when we affirm that man is superior to the other animals because he possesses intellect, it is indeed the thing itself, man, which we hold in our mind and to which we join or attribute those other things, intellect and superiority. But the man that we thus handle in our mind is obviously not the man as he exists, or can exist in reality ; there is no question of seizing some man who passes in the street to stick an attribute on to his back. That our mind may work on them, things possess in the mind a manner of being which they do not and cannot jjossess in reality. They exist there so far as they are known. Predicated one of another divided, reunited, linked together according to the necessities of knowledge, they lead there a distinct life, with its own laws. It is this life and its laws, the order to which things must submit so far as they are objects of knowledge, if they are to guide the mind 152
  • 159. THE MAIN DIVISIONS OF PHILOSOPHY to truth, that logic primarily studies, and since it is concerned with something which exists and can exist only in the mind or with what philosophers term an ens raiionis, a conceptual being, we may say that the formal object of logic is that conceptual being, ens rationis (the order which should prevail among conceptual objects) which directs the mind to truth. As opposed to conceptual being, the ens rationisy which can exist only in the mind—for example, the genus animal or the species man {the genus animal comprises mankind and the brutes, man is the species of Peter)—we term real being, ens reale, that which exists or can exist in reality—for example, animals, man, human nature {all animals are mortal, human nature is fallible). Conclusion V.—Philosophy is divided into three principal parts : (i) logic, which is the introduc-tion to philosophy in the strict sense, and which studies the conceptual being {ens rationis) which directs the mind to truth ; (ii) theoretical philosophy or simply philosophy, which studies the being of things (real being, ens reale) ; (iii) practical philosophy or ethics, which studies human acts. 153
  • 160. II LOGIC Logic studies reason as the tool of knowledge. To study any complicated machine, a reaper for instance, we must begin by making it work in the void, while we learn how to use it correctly and without damaging it. In the same way we must first of all learn how to use reason correctly, that is to say, in conformity with the nature of ratiocination, and without damaging it. Hence arises our first problem : What are the rules which we must obey in order to reason correctly ? We should next study our reaper no longer in the void, but as applied to the actual material with which it was designed to deal, learning how to use it, not only correctly, but profitably and efficiently. In the same way we must study reasoning as applied to facts, asking ourselves under what conditions reasoning is not only correct but also true and conclusive, and productive of knowledge. It is in this department of logic that we study the methods employed by the different sciences. But before this a far graver problem will arise and demand solution. It is by our ideas that things are presented to the mind, so that we can reason about them and acquire knowledge. Everybody knows by experience what an idea is. It is sufficient for a man to reflect on what is in 154
  • 161. LOGIC a ; his mind, when he makes a judgment of any kind. For instance, philosophers have made many mistakes philosophers^ mistakes, have made, many—all these are present in the mind as so many ideas. Nevertheless to clear up any possible obscurity, we will try to describe what everybody means by the word. We will, for instance, define ideas as images or interior reproductions of things, by which the latter are pre-sented to us in such a way that we can reason about them (and thus acquire knowledge). No doubt the words we employ express our ideas. But they bring with them something besides. If, for example, I pronounce the word angel, I have in my consciousness two images of the being in question. In the first place an idea, in virtue of which, strictly speaking, I know that particular being (the idea of a pure spirit), but in addition a sensible representation (the image of some figure more or less nebulous and winged) which possesses no Ukeness whatever to the being in question, for a purely spiritual being is invisible. If, again, I pronounce the word square I have in my consciousness the idea of the square, by means of which I can reason about the thing concerned (the idea of a rectangular polygon offour equal sides) and at the same time a sensible representation—which in this case adequately represents the thing in question— particular figure which I imagine drawn in chalk on the blackboard. The idea and the representation are quite distinct, as is shown by the fact that I can vary the latter in a host of different ways (the imaginary drawing can be larger or smaller, white, red, yellow, etc.) without any variation of the former. Moreover, 155
  • 162. — AN INTRODUCTION TO PHILOSOPHY if I were to pronounce, for example, the word myriagon instead of square^ I should possess as definite and as clear an idea of it as I had of the square (the idea of a polygon of ten thousand sides) ^ whereas the only sensible representation I could form of it would be extremely vague and confused. It is evident that if the sensible representations assist me to reason, they are not the instrument with which I reason to acquire a knowledge of things. For I can reason as accurately about the angel or the myriagon as I can about the square. And my reasoning is in no way dependent on the thousand alterations I can make in my sensible representations of an angel, myriagon, or square. From this we conclude that things are presented to our consciousness in two different fashions, either by an idea or by a sensible representation. By the first we think (intelligimus) the thing, by the second we imagine it. The representation is simply a species of phantom, an image of what we have pre-viously seen, heard, touched, etc., in short, of what has been originally made known to us by a sensation. Formerly called a phantasm, it is now called simply an image. In future, then, we shall reserve to denote it the term image, whose meaning we accordingly restrict. (But we must no longer use the same word to denote an idea.) We conclude, therefore, that Conclusion VI.—Ideas are the internal likenesses of things by which the latter are presented in such a way that we can reason about them (and thus acquire knowledge) ; images are the internal likenesses of things by which the latter are pre- 156
  • 163. LOGIC sented to us as our sensations have first made them known to us. Words directly signify ideas, at the same time evoking images. If now we compare objects as they are presented by ideas and as they are presented by sensations or images, we see at once that they are distinguished by a character of the very first importance. If, for example, I call up before my mind the image of a man, I see present in my imagination with outlines more or less vague and more or less simplified some particular man. He is fair or dark, tall or short, white or black, etc. But if I form the idea of man, as, for instance, when I state the proposition man is superior to the irrational animals, or whites and blacks are alike men, that idea does not bring before me any man in particular. It leaves out of account all the individual characteristics which distinguish one man from another ; in the language of philosophy it abstracts from them. This is proved by the fact that while remaining absolutely the same and without need of any modifica-tion whatsoever, it can be applied to the most dis-similar individuals ; Sancho Panza is just as much a man as Don Quixote. Moreover, when we cast our mind over the different sciences, that is to say, the different systems of ideas by which we know reality, we find that none of them is concerned with the individual as such. Chemistry, for instance, only studies in chlorine or nitrogen what is common to all the individual molecules of chlorine or nitrogen. And this must necessarily be the case. The individual as such explains nothing (for, since it represents only 157
  • 164. AN INTRODUCTION TO PHILOSOPHY itself, it obviously cannot account for anything else).^ Again, we have only to take any idea whatsoever and fix our attention on what it presents to us, comparing it with the images which form and dissolve around it, to perceive at once the abstract nature of the idea. In the transition from the image to the idea whatever is individual evaporates, so to speak, slips between our fingers, and vanishes. Take, for example, the idea of weapon which I employ when I state that man is the only animal obliged to manufacture its weapons. As I pronounced the word weapons, I was no doubt conscious of a halo, so to speak, of fluctuating images surrounding the idea thus expressed, to any of which I can at pleasure give a more definite shape, di javelin very shadowy no doubt, a flint axe^ a cross-bow, a gun. . . . But of the individual characteristics of the particular javelin, axe, bow, or gun as they appear in my imagination with their distinctive form, colour, and dimensions, nothing whatever remains in my idea of weapon. Everything of the sort has disappeared. Though what I apprehend by the idea is certainly some thing, that some thing is of an entirely different order (immaterial), it is simply a certain determination of being, a certain nature, an instrument of attack or defence ; and that is devoid of any individual character. That is to say, objects as presented to us by our sensations and images are presented in a state which is individual, or in technical language singular. On the contrary, objects as presented to us by our ideas, by the internal likenesses which enable us to reason about them, are presented in a state which is non-individual, abstract, or in technical language universal. 1 Cf. T. Richard, op. cit., p. 21. 158
  • 165. LOGIC — We call universal that which is the same in a multi-tude of individuals, one in many, unum in multis.) We shall therefore hold as an established truth that Conclusion VII.—Our sensations and images present to us directly or by themselves the indi-vidual, our ideas directly and by themselves the universal. But the question at once arises : Since real objects are individual or singular, how can the knowledge we obtain by means of our ideas be true, since our ideas directly present only the universal ? This problem, which will compel us to investigate carefully in what exactly consists the universality of that which our ideas present, is, not indeed in itself, but at any rate for us men, the first and most im-portant of philosophic problems.^ For it is concerned with the nature of the intellect itself and of our ideas, that is to say, with the instrument by which all our knowledge is obtained ; and the solution propounded by different philosophers dominates their entire system. From this point of view, and taking no account of many differences of secondary importance, we may classify philosophers in three great schools : {a) The nominalist school, for which universzds have 1 Does the problem of universals belong to logic, psychology, or metaphysics (criticism or epistemology) ? To all three, in fact, according as it is studied from three different standpoints. We may inquire what constitutes the nature of a universal (standpoint of the formal cause), or the manner in which a universal is formed in the mind (standpoint of the efficient cause), or the epistemological value of the universal (standpoint oi the final cause).
  • 166. AN INTRODUCTION TO PHILOSOPHY no existence except as names or ideas with which nothing in reality corresponds ; for instance, there is nothing in the reality of human nature which is equally present in Peter^ Paul, and John. This position amounts to sheer negation of the possibility of intellectual knowledge, and reduces science to a figment of the mind. The most typical representatives of this school are, in antiquity the sophists and the sceptics, in modem times the leading English philosophers, William of Occam in the fourteenth century, Hobbes and Locke in the seventeenth, Berkeley and Hume in the eigh-teenth, John Stuart Mill and Spencer in the nine-teenth. It may be added that the majority of modem philosophers (that is to say, of those who ignore or oppose the scholastic tradition) are more or less deeply, and more or less consciously, imbued with nominalism. (b) The realist school (" absolute realism ") for which the universal as such, the universal taken separately, as it exists in thought, constitutes the reality of things. This position reduces sense-knowledge to mere illusion. That which is real is, for example, a human nature existing in itself and separately outside the mind, a man in himself (Platonism), or a universal being existing as such outside the mind and regarded as the sole and unique substance (doctrine of Parmenides, Vedantism) . The systems of certain modern philosophers (Spinoza, Hegel) approxi-mate more or less closely to realism.^ {c) The school which is usually called that of > It mast be borne in mind that realism understood in this sense, far from being incompatible with idealism, is essentially an idealist doctrine. For realism of this type regards as the reality of things that which is distinctive of our ideas as such. Plato is thus at once the most typical representative both of idealism and absolute realism. i6o
  • 167. LOGIC moderate realism. (Its doctrine, however, is in the most strict sense original, and keeps the just mean between realism and nominalism, not by watering down or modifying absolute realism, but by a view of things which transcends the opposing errors.) This school, distinguishing between the thing itself and its mode of existence, the condition in which it is presented, teaches that a thing exists in the mind as a universal, in reality as an individual. Therefore that which we apprehend by our ideas as a universal does indeed really exist, but only in the objects themselves and therefore individuated—not as a universal. For example, the human nature found alike in Peter, Paul, and John really exists, but it has no existence outside the mind, except in these individual subjects and as identical with them ; it has no separate existence, does not exist in itself. This moderate realism is the doctrine of Aristotle and St. Thomas. Philosophy of Aristotle and St. Thomas {Moderate Realism) That which our ideas That which our ideas present to us as a universal present to us as a universal does not exist outside the exists outside the mind mind as a universal. individuated. Nominalism Realism That which our ideas That which our ideas present to us as a universal present to us as a universal has no real existence what- really exists as a universal, soever. It is impossible to over-emphasise the importance of the problem of universals. It is for want of atten-tion to it that so many philosophers and scientists of modem times cling to the naive belief that science i6i
  • 168. AN INTRODUCTION TO PHILOSOPHY must be a copy pure and simple, a tracing of the individual reality ; serve up the stock arguments of ignorance against abstraction, the essential pre-condition of all human knowledge ; and when treating of the principles of the sciences, especially of mathe-matics, spin elaborate theories, devoid of solid founda-tion, whose sole result is to render knowledge totally impossible. -i 162
  • 169. Ill THE PHILOSOPHY OF MATHEMATICS THE PHILOSOPHY OF NATURE The distinctive object of theoretical philosophy is the being of things. The things which are immedi-ately observed are corporeal things, bodies. But the term body may be taken in two distinct senses. It may mean a mathematical body, or a natural or physical body. A mathematical body is simply that which possesses three-dimensional extension, breadth, length, and height. A natural or physical body is that which is perceived by the senses as possessing certain active and passive properties. THE PHILOSOPHY OF MATHEMATICS If the philosophy of mathematics studies the being of bodies in the first sense of the term body, it is obvious that the first problem it must consider is in what does the primary object of mathematics consist ; in other words, what is the nature of quantity, extension, and number ? ^ 1 Questions relating to the philosophy of mathematics are usually treated in natural philosophy or in metaphysics. We believe, however, that if classification is to be scientific, we are obliged to maintain in what is now known as philosophy (scientific knowledge of things by their first causes) the fundamental division of the sciences (the whole group of which constituted for the ancients theoretical philosophy) into three parts : physica, mathematica, metaphysica, corresponding to 163
  • 170. AN INTRODUCTION TO PHILOSOPHY The enormous progress made by modern mathe-matics has rendered more indispensable than ever before the philosopliic study of the first principles of the mathematical sciences, which alone can provide a rational account of the true nature of mathematical abstraction and the mentzd objects which it considers, the properties- and mutual relationships of the con' tinuous and the discontinuous, the real meaning of surds and transjinite numbers, the infinitesimal, non-Euclidean space, etc., and finally of the validity of mathematical transcripts of physical reahty, and of such hypotheses, for example, as the theory of relativity. THE PHILOSOPHY OF NATURE * Since the philosophy of sensible nature studies the being of bodies in the second sense of the term body, it the three grades of abstraction (see p. 1 52) . Cf. Aristotle, Metaph., vi, i, 1026 a 18. Tjoeii iv eUv <pio(Xo<plai dewptjTiKal, ixad7)/j,aTiKT^, <(>v<nKi/i, deciXoyiKTi. It is true, as we shall see later, that the philosophy of mathematics, for the very reason that it studies the essence of quantity and is thus at least reductively metaphysical, transcends the strict sphere of the mathematical sciences and is specifically distinct firom them. This, however, does not alter the fact that it is concerned with the second degree of abstraction and must therefore be studied as a separate branch of philosophy. 1 In the logical order of the sciences, the natural sciences which correspond to the first degree of abstraction (see p. 152) precede the mathematical sciences, which correspond to the second, so that in accordance with this order we should be obliged to divide theoretical philosophy into (i) The philosophy of nature (corresponding to the first degree of abstraction), (ii) The philosophy of mathematics (corresponding to the second degree), (iii) metaphysics (corresponding to the third degree). Nevertheless the philosophy of mathematics should precede natural philosophy for two reasons. On the one hand, truths of the mathematical order are easier to 164
  • 171. MATHEMATICS AND NATURE must deal with a large number of problems. We can, however, pick out the most important of these for mention here. The most universal and obvious characteristic of the corporeal world, which is involved in every physical event, is change. Philosophers, in whose vocabulary change of every description is termed motion, must therefore inquire what motion is. It is at once obvious that if motion exists, something must be moved, namely, bodies. Further, certain changes seem to affect the very substance of bodies ; as, for instance, when the chemical combination of hydrogen and oxygen produces a new body, water. How is this possible ? We are compelled to ask what corporeal substance is. (a) The mechanists—whether in their doctrine of the human soul they are materialists (Democritus, Epicurus, Lucretius, among the ancients, Hobbes in the seventeenth century, etc.) or spiritualists Uke apprehend than truths of the natural order, which presuppose experience. For this reason children should be taught the elements of mathematics before the natural sciences, the study of whic^equires a more advanced age. {Cf. Aristotle, Mc. Eth., vi ; St. T'homas, Sup. Boet. de Trm., q. 5, a. i, a// 3.) We should therefore follow the same order in philosophy and lead the mind up to the study of natural philosophy by the study of the philosophy of mathematics. On the other hand, and this is the more important consideration, natural philosophy with the last and highest of its sub-divisions, namely psychology, touches the frontier of metaphysics. It would be a breach of continuity to insert the philosophy of mathematics between natural philosophy and metaphysics. In the seventeenth century Sylvester Maurus maintained—and in so doing was faithful to the Aristotelian tradition—that the natural order of study is as follows : (i) logic, (ii) mathematics, (iii) physics, (iv) metaphysics. {Quaest. philos., i, q. vii.) 165
  • 172. AN INTRODUCTION TO PHILOSOPHY Descartes—reduce corporeal substance to matter, which in turn they confuse with quantity or geometrical extension. They can therefore admit no essential or specific difference among bodies, which are all modifications of one single substance. Moreover, the physical universe is for them devoid of quality and energy, since space and local motion alone are real, and the union of matter and spirit in a being such as man becomes absolutely unintelligible. {b) Another school {dynamism) tends on the contrary to get rid of matter as a constituent of bodies. It culminates in the system of Leibniz, who reduced corporeal substance to units of a spiritual character {monads) analogous to souls. For Leibniz extension, indeed sensible reaUty as a whole, is nothing more than an appearance or a symbol, and the corporeal world as such is absorbed in the spiritual. The dynamism of Boscovich (eighteenth century), who reduced corporeal substance to points offorce, and the modern physical theory which claims to explain everything in the physical universe as manifestations of one sole reality, energy (of which, however, its ex-ponents fail to give a philosophic definition), may be regarded as degradations and materialisations of Leibniz's doctrine. {c) The Aristotelian philosophy recognises in corporeal substance two substantial principles : (i) matter {first matter, materia prima), which, however, in no way represents, as in the conception of the mechanists, the imaginable notion of extension, but the idea of matter (that of which something else is made) in its utmost purity—it is what Plato called a sort of non-entity, simply that of which things are i66
  • 173. MATHEMATICS AND NATURE made, which in itself is nothing actual, a principle wholly indeterminate, incapable of separate existence, but capable of existing in conjunction with something else (the form) ; (ii) an active principle, which is so to speak, the living idea or soul of the thing, and which determines the purely passive first matter, somewhat as the form imposed upon it by the sculptor determines the clay, constituting with it one single thing actually existent, one single corporeal substance, which owes to it both that it is this or that kind of thing, that is to say, its specific nature, and its existence, somewhat as the form imposed by the sculptor makes a statue what it is. On account of this analogy with the external form of a statue (its accidental form) Aristotle gave the name of/orm {substantialform) , which must be understood in a sense altogether special and technical, to this internal principle of which we are speaking, which determines the very being of corporeal substance. The Aristotelian doctrine, which regards a body as a compound o{ matter [xikri) andform ([xopcprj), is known as hylomorphism. It accepts the reahty, on the one hand, of matter, the corporeal world, and extension,^ on the other of physical qualities,' also a distinction of nature or essence between the bodies which we regard as belonging to different species. It reveals the presence, even in inanimate bodies and Hving things devoid of reason, of a substantial principle, immaterial in its nature, which, however, differs from spirits in ^ Extension of quantity is not, as the mechanists hold, the substance of bodies, but their first accident. * Qualities are also accidents of corporeal substance. (See below, pp. 217-232 (substance and accident).) 167
  • 174. AN INTRODUCTION TO PHILOSOPHY the Strict sense, in its incapacity to exist apart from matter. And it renders intelligible the union in the human being of matter and a spiritual soul which is the form of the human body, but differs from the Other substantial forms inasmuch as it can exist apart from matter. Ptdlosophy of Aristotle and St. Thomas {Hylomorphism) Every corporeal substance is a compound of two substantia! and complementary parts, one passive and in itself wholly indeterminate [matter), the other active and the principle of determination {form). Mechanism Dynamism Ck>rporeal substance is Corporeal substance is regarded as something explained either as units simple, a matter itself belonging to the category identified with geometrical of pure forms and spirits extension. (Leibnizian monadism) or as a manifestation of force or energfy. We have now to consider a class of bodies which possess a peculiar interest for us, and seem to be superior to all the others. They are living bodies, from the lowliest micro-organism to the human organism. The property which distinguishes them from all other bodies is self-movement. On that account common sense recognises in them a soul or principle of life, irreducible to any combination of physico-chemical factors or elements. If this is indeed the case, we must inquire whether there are not different kinds of soul, whether vegetables and animals possess a soul, etc. On the other hand certain philosophers (known by the general appellation of mechanists) claim that science will one day explain all the phenomena of life by the forces of lifeless matter, that is to say, that the living i68
  • 175. MATHEMATICS AND NATURE organism is simply a very complicated physico-chemical machine. This involves a problem of the first importance. What is Ufe ? What are the first principles which constitute the living organism ? But of all living things which possess a body the highest is man. Man is as it were a world apart, for the study of which we are in a peculiarly favourable position, because we know him from within by what is called self-consciousness. His most distinctive charac-teristic is the possession of intelligence or reason. If, however, intelligence is indeed something wholly immaterial, it follows that the science which studies man, though a branch of natural philosophy which treats of moving or sensible being, is in a sense intermediate between this department of philosophy and metaphysics which treats of the wholly immaterial.^ If it is the possession of intelligence or reason which makes man man, the problems which relate to his intellectual activity must, it would seem, dominate the entire science of man.^ And in fact the 1 The science of man occupies therefore a singular position (due to the very nature of its object) astride two distinct sciences, nattiral philosophy and metaphysics. It is for this reason that all questions involving the intellect and the strictly spiritual portion of psychology display in the case of man such extreme complexity and are, so to speak, overshadowed by matter. It is not surprising, therefore, that when the Thomists wished to investigate these questions in their purity they studied not man but the angels. Hence the extreme importance of the treatise De Angelis, not for theology alone, but also for philosophy and metaphysics. ' Observe that psychology as understood by the moderns does not correspond exactly to the ancients' treatment of the soul. Aristotle's Trepl fvxv^> De Anima, studies not only the human soul, but also the soul in general as the principle of life, whether vegetative, sensitive, or intellectual. Such a treatise therefore belongs to what we now call biology as well as to psychology. 169
  • 176. AN INTRODUCTION TO PHILOSOPHY fundamental problem of psychology is that of the origin, of ideas : how we are to explain the presence in us of the ideas which enable us to reason about things and which present things to us as universals. At this point we have been brought back by a different approach to the problem of universals which we have lately considered. We then noted that what our ideas immediately present to us is something non-individual or universal. We have now to ask how this knowledge of the universal is acquired by our minds. We saw above that things as they are known by the senses and tha imagination are presented as individuals. It is this particular man that I see, with this particular appearance actually impressing itself on my retina and distinguishing him from the other man I see beside him. Sense-knowledge is thus knowledge of the individual alone. The object as object of sensation or the object reproduced by an image is the object apprehended in its individuality. Since, therefore, what we know immediately by our ideas is not individual, the reason must be that our ideas are in fact extracted by us from our sensations and images, but in such a fashion that there enters into them nothing whatsoever of the object as it exists as an object of sensation or reproduced by an image (that is to say, as we shall see later, as the object of a know-ledge steeped in materiality). Arising from images, but of a higher order than the image, and apprehending nothing of the object as reproduced by the image, our ideas must necessarily be unable to give us any know-ledge of the object in its individuality. 170
  • 177. MATHEMATICS AND NATURE Moreover, we could not possibly derive our ideas from things, except by way of our senses, which are in immediate contact with things. And we have only to observe the mental development of a child to be convinced that all our knowledge begins with the senses. Therefore intellectual knowledge (knowledge by means of ideas) must undoubtedly be derived from sense-knowledge. On the other hand, since everything apprehended by sensations and images is characterised by individu-ality, our ideas must be extracted from images in such a way that nothing of the image, as such, enters into the idea. But how is this process of extraction conceivable ? If nothing whatever of the object as it is reproduced by the image enters into the object as it is apprehended by the idea, it is obvious that the idea is not the result of any combination or distillation of sensations or images. We are therefore compelled to postulate an agent of a higher order, the voii<; 7roL7]Ti.x6(;, as the Peripatetics termed it, the intellectus agens—a kind of intellectual light (we may perhaps compare it to X-rays) which, when applied to the object presented to us by the image, draws out of it for our understanding something already contained in it but hidden, which the image by itself could never reveal. The some-thing thus extracted and Uberated from that which constitutes the individuality of the object (because it is liberated from that which constitutes the materi-ality of sense-knowledge) is the form or intelligible likeness of the object, which is, so to speak, imprinted on the intellect to determine it to know, by making it produce within itself by a vital reaction the idea 171
  • 178. AN INTRODUCTION TO PHILOSOPHY in which it apprehends the object as a universal ; for example, the idea of man or living being, of Aryan or Semite. We must, however, remember that what our ideas present thus as a universal is in itself (abstracting fix)m its existence either in things or in the mind) neither individual nor universal, being purely and simply that .which the thing is.^ We must also bear in mind that, if our intellect does not directly know the individual as such, it knows it indirectly. For, at the very moment when it thinks of an object by means of an idea, it turns to the images from which the idea has been drawn, which present the thing as an individual. And by thus reflecting on the images it apprehends, though indirectly and in a manner wholly superficial and totally inexpressible, the individuality of the thing. Conclusion VIII.—Our ideas are extracted {ab^ straded) from the sensible datum by the activity of a special faculty (the intellectus agens or active intellect) which entirely transcends the sensible order and is, as it were, the light of our under-standing. Philosophers term abstraction the operation by which we thus extract our ideas from the store of images accumulated by sense experience, ideas which repre-sent that which the thing is, abstracting from its individuality. Here we may add that abstraction admits of lesser and greater degrees. For instance, though the idea 1 I.e. the nature, essence, or quiddity, of the object. See pp. 20i, 206. 172
  • 179. MATHEMATICS AND NATURE of horse is, like every idea, abstract, when we think of horse we can at the same time see or imagine horses, and thus know in the sensible order what we know at the same time by means of our idea in the intelligible order. If, on the other hand, we think of angel or spiritJ the sole function of the more or less vague images which accompany the thought is, as we have already observed, to assist the intellect to function. In their own order they have no value as knowledge, for they tell us nothing. We can neither see nor imagine an angel or a spirit ; in this case, therefore, we cannot at the same time know by our senses the thing we know by our intellect. It is important to bear in mind that the things with which philosophy is primarily concerned belong to this second category. They cannot be known either by the senses or the imagination, but solely by the intellect. It is to this higher degree of abstraction that the study of philosophy owes its special difficulty. Be-ginners are often perplexed when they suddenly exchange the literary studies on which they have been engaged hitherto, studies in which the imagination was employed equally with the intellect, for an exercise wholly intellectual. But this difficulty will soon pass, if they will not try to represent by the imagination objects of pure thought which are entirely unimagin-able, such, for instance, as essence, substance, accident, potentiality, and act ; a chimerical attempt, which will only cause needless headaches and effectually prevent them from understanding anything of philosophy. If abstraction is indeed such an operation as we have described, it follows in the first place that man 173
  • 180. AN INTRODUCTION TO PHILOSOPHY is endowed with a spiritual soul, the first principle of this function (for its result, our ideas, is incommensur-able with sensations and images and of a purely immaterial order) ; and on the other hand, that it is of the very nature of this spiritual soul to be united to a body (for our ideas cannot be formed except by means of sensations and images, which in turn neces-sarily suppose bodily organs). We thus perceive how the problem of abstraction, or the origin of ideas, is bound up with another fundamental problem of psychology, which concerns the very essence of man : in what does the human being consist ? Does man possess a spiritual soul, wholly different from that of the beasts ? And if so what is the relationship between this soul and the human body ? On the problem of the origin of ideas philosophers may be divided roughly into three main groups : (a) The sensualists, who hold that ideas are derived from the senses, but reduce ideas to sensations, (b) The partisans of innate ideas,^ who recognise the essential distinction between ideas and sensations or images, but deny that we extract our ideas fi-om the sense datum, (c) The school of Aristotle and St. Thomas^ which holds that our ideas differ essentially from sensations and images, but that they are extracted ^ We may so term this second group in default of a more suitable title, but only if we considerably widen its meaning. For in this class of philosophers we must include not only those who teach that our ideas exist in our minds from birth in the same way as our soul exists (the doctrine of innate ideas in the strict sense), but those who hold that they are immediately implanted in us by God or are seen by us in GJod (Berkeley, Malebranche), or are the arbitrary product of our mind imposing its laws on things (Kant).
  • 181. MATHEMATICS AND NATURE from them by the operation of the spiritual light in us (voiji; noirixiy(.6c„ intellectus agens). The principal representatives of sensualism are Locke (seventeenth century) and John Stuart Mill (nineteenth century) in England, and Condillac (eighteenth century) in France. The sensualists are, as a rule, also nominalists, but the converse does not hold, and many philosophers whom we class here among the defenders of innate ideas betray, in modem times at least, the influence of nominalism. In the second class (the defenders of innate ideas) we must reckon Plato among the ancients, Descartes (seven-teenth century) and Leibniz (seventeenth to eighteenth century) among the modems. Though their explana-tions differ, all these hold that our ideas are innate. Kant (end of eighteenth century) also belongs to this group, though for him what is innate is not our ideas, but the categories, rules, or forms in accordance with which our mind manufactures the objects of knowledge. Philosophy of AristotU and St. Thomas Our ideas are derived Our ideas are essentially from the senses (and there- different from sensations fore from things) but by and images, but are ab-the operation of a spiritual stracUd from them by the faculty, and are essentially operation of a spiritual different from sensations faculty, and images. Sensualism Doctrine of Innate Ideas Our ideas are derived Ideas differ essentially from the senses, which are from sensations and images sufficient to produce them, and are not derived from and do not differ essenti- the senses (nor therefore ally from images and sen- from things, with which sations. our senses alone are in immediate contact).
  • 182. . AN INTRODUCTION TO PHILOSOPHY The answers which philosophers have given to the problem of human nature correspond strictly with the position they adopt towards the problem of abstrac-tion. The sensualists, at least so far as they are faithful to the logic of their doctrine (Gondillac, for example, was not), deny either that the soul existe {materialists), or that we can in any case know its existence {pheno-menalists) . The defenders of innate ideas, on the other hand, tend to regard man as a pure spirit which happens to be joined to a body—how, they find it difficult to explain {dualism or exaggerated spiritualism).^ Finally, the school of Aristotle and St. Thomas teaches that man is a composite of two substantial principles, each incomplete in itself and the complement of the other, one of which is a spiritual and immortal soul {animism) Philosophy of Aristotle and St. Thomas (Animism) Two principles each incomplete in itself, one of which (the rational soul) is spiritual, form together a single substance (the human composite). Error of Defect Error of Excess The human soul does not Man is a spirit acci-exist {materialism) or is dentally united to a body unknowable {phenomenal- {exaggerated spiritualism) : ism). the soul and the body are two substances each com-plete in itself {dualism) . We should remark further that the position adopted by philosophers towards the origin of ideas also deter-mines their attitude to the general problem of the * This tendency recurs even in Kant (especially in ethics), though he, like the phenomenalists, denies that reason can demonstrate the / existence of the soul. >i 176
  • 183. MATHEMATICS AND NATURE . existence of things known by the senses (the sensible or corporeal world) and of things invisible and spiritual, accessible to reason alone. Philosophy of Aristotle and St. Thomas {also of common sense) It is impossible without absurdity to doubt either the existence of corporeal objects (attested by the senses) or the existence of spiritual objects (proved by reason) Systems more or less Materialist Nothing exists which is not perceptible by the senses and material {abso-lute materialism) ; or at least its existence is un-knowable {phetwmenalist materialism and positivism). Systems more or less Idealist The world perceived by the senses has no real existence {absolute idealism); or at least its existence is unknowable and doubtful {phenomenalist idealism). 177
  • 184. IV CRITIGISM (ePISTEMOLOGY) In studying man philosophy is dealing with an object which already by an entire portion of itself transcends the corporeal world, that is to say, the world of sensible nature. But it has the power and duty to go further, and since its distinctive object is the being of things, it must study that being no longer as corporeal, sensible, or moving (the subject-matter of the philosophy of sensible nature) , but simply as being ; consequently it must study being under an aspect absolutely universal, and as it is present not only in visible things but also in things which possess no corporeal, sensible, or mobile being ; that is to say, in things which are purely spiritual. This is the object of that branch of philosophy which is philosophy or wisdom par excellence, and is known as the first philosophy or metaphysics.^ CRITICISM (ePISTEMOLOGY) But before undertaking this study, the philosopher must secure against all possible attack or distortion the 1 The name metaphysics originated in the fact that in the catalogue of Aristotle's works drawn up by Andronicus of Rhodes, the treatise dealing with the first philosophy (Ilepi rrjs irpuTtp <t>ioao<f>la^, the title probably which Aristotle himself would have given to it) comes after the books which treat of naturi (Meri tA tpvaiKi). It would seem, however, that chronologically Aristotle followed the same order in the actual comp>osition of his works. 178
  • 185. CRITICISM ( EPISTEMOLOGY) principles of this sovereign science, which are also the principles of all human knowledge. For it is the office of wisdom to defend its own principles and those of the other sciences. It will therefore be necessary, before studying being in itself, as such, to study the relation ofhuman thought to being. This is the object of a special department of metaphysics, known as criticism, because it has the function of judging knowledge itself Logic shows how and in accordance with what rules reason attains truth and acquires knowledge ; this in turn pre-supposes the pKDSsibility of true knowledge (a possibility attested by common sense and evident by the light of nature). Criticism submits this presupposition to scientific treatment, showing in what the truth of knowledge consists, and establishing by a reflex argument that true, certain, and scientific knowledge is undoubtedly attainable.^ What is the truth of knowledge, and is it possible to refiite those who question the veracity of our organs of knowledge, particularly of the intellect or reason ? This clearly is the double question which arises at the outset. The answer, however, is sufficiently plain. As to the first question, there is no difficulty in under-standing what is meant by the notion of truth. What is a true or truthful word ? A word which expresses, as it really is, the speaker's thought ; a word in 1 By thus distinguishing criticism (epistcmology) from logic, and making it the first part, special introduction, or if you prefer, apologetic introduction, to metaphysics, we are faithful to the arrangement and divisions of Aristotle himself, who discusses criticism briefly {Metaph., iv) before studying the great problems of being as such.
  • 186. AN INTRODUCTION TO PHILOSOPHY conformity with that thought. What, then, is a true thought ? A thought which represents, as it really is, the thing to which it refers ; a thought in conformity with that thing. We therefore conclude that truth in the mind consists in its conformity with the thing. It is impossible to define truth otherwise without lying to ourselves, without falsifying the notion of truth of which in practice we make use, in the living exercise of our intelligence, each time that we think. We may further remark that a thought false in all its constituents is an impossibility for, being in conformity with nothing whatsoever, it would be the zero of thought. If, for instance, I affirm that stones have a soul^ this is undoubtedly a complete error. But it is true that stones exist, true also that certain beings have a soul ; that is to say, all the constituents which compose this false thought are not false. Therefore error itself presupposes truth. ^ We may also observe that if man were really and seriously to doubt the veracity of his organs of knowledge he simply could not live. Since every action or abstention from action is an act of trust in that veracity, action and inaction would alike become impossible. A man therefore who attempted to carry out in his Ufe the thought truth is impossiblefor me would inevitably lose his reason. Nietzsche, who was a great poet but regarded belief in truth as the ultimate bondage from which the world should be deUvered, made the experiment to his cost. i Cf. Sum. TheoL, ii-ii, q. 172, a. 6 : Sicut se habet bonum in rebus, ita verum in cognilione. ImpossibiU est autem inveniri aliquid in rebus, quod totaliter bono privetur : unde etiam impossibile est esse aliquam cogrdtionem quae totaliter sit falsa absque admixtione alicujus veritatis. 180
  • 187. CRITICISM (ePISTEMOLOGY) As for the sceptics, who doubt, at least theoretically and in words, the reliability of our organs of knowledge, especially of the intellect or reason, it would obviously be waste of breath to attempt to demonstrate its reliability to them. For every demonstration rests on some previously admitted certainty, and it is their very profession to admit of none. To defend human knowledge against their attack it is sufficient (i) to show in what that knowledge consists and how it is attained ; (ii) to refute the arguments they adduce ; (iii) to make a reductio ad absurdum. When they say that they do not know whether any proposition is true, either they know that this proposition at any rate is true, in which case they obviously contradict themselves, or they do not know whether it is true, in which case they are either saying nothing whatever, or do not know what they say. The sole philosophy open to those who doubt the possibility of truth is absolute silence—even mental. That is to say, as Aristotle points out, such men must make themselves vegetables. No doubt reason often errs, especially in the highest matters, and, as Cicero said long ago, there is no nonsense in the world which has not found some philosopher to maintain it, so difficult is it to attain truth. But it is the error of cowards to mistake a difficulty for an impossibihty. Conclusion IX.—The truth of knowledge consists in the conformity of the mind with the thing. It is absurd to doubt the reliability of our organs of knowledge. On this question of the reliability of our organs of i8i
  • 188. AN INTRODUCTION TO PHILOSOPHY knowledge philosophers may again be divided — roughly—into three groups : {a) The sceptics, who, impressed by the enormous number of errors put forward by men, and especially by philosophers, doubt the trustworthiness of reason, and affirm that truth is impossible of attainment. The principal representatives of scepticism are, among the ancients, Pyrrho (360-270), the neo-Academics (Arcesilas 315-241 ; Cameades, 214-129) and the later Greek sceptics (Aenesidemus, first century a.d., and Sextus Empiricus, end of the second century) ; in modern times Montaigne and Sanchez in the sixteenth century, and pre-eminently David Hume in the eighteenth. The philosophers called anti-intellectualists, because they despair of intellect and reason, and look for truth to the will, to instinct, feeling, or action (Rousseau, Fichte, Schopenhauer, Bergson, William James, the modernist and pragmatist school), must be classified with the sceptics, because, although they do not, like the sceptics strictly so called, declare truth unattainable, they maintain that it is unattainable by the organ whose distinctive nature it is to discover truth, and because by rejecting the intellect and reeison they effectually deprive man of his sole normal means of attaining it. (b) The rationalists, on the contrary, are of opinion that truth is easy to attain, and therefore undertake to bring all things within the compass of reason, a human reason which has no need to submit humbly and patiently to the discipline, whether of reaUty itself, a teacher, or God. In the first case they tend to subjectivism, which takes as its criterion of truth the knowing subject, not the object to be known ; a position which is the dissolution of knowledge. In the 182
  • 189. . CRITICISM ( EPISTEMO logy) second they tend to individualism, which calls upon each philosopher to work out a philosophy entirely his own, and create an original and novel view of the universe {Weltanschauung). In the third, they tend to naturalism^ which claims to attain to a perfect wisdom by the unassisted powers of nature, and rejects all divine teaching.^ The father of modem rationalism was Descartes (seventeenth century) , to whom Malebranche, Spinoza, and Leibniz traced more or less directly their philosophic pedigree. But its first principles and true spirit were revealed by Kant (end of eighteenth century), who completed the Cartesian revolution ; his pantheistic successors, Fichte, Schelling, and Hegel, deified the human subject of knowledge. Through Kant, and the subjectivist philosophy which traces its origin to him, rationaUsm, as before in the era of the sophists, has joined hands with its opposite (scepti-cism), and become absorbed in the anti-intellectualism of the modernists (end of the nineteenth and beginning of the twentieth century) {c) The school of Aristotle and St. Thomas teaches that truth is neither impossible nor easy, but difficult for man to attain. It is thus radically opposed alike to scepticism and to rationalism. It sees in the multitude of errors put 1 Naturalism rejects divine teaching in these two different ways : ( I ) It denies God the right to teach men truths in themselves inaccessible to the unassisted reason {supernatural mysteries). (2) It also denies him the right to teach men by revelation truths in themselves accessible to unassisted reason (truths of the natural order, philosophic truths—for example, the immortality of the human soul) which reason can indeed discover by its unaided powers, but always with the risk of mingling error with truth, whereas revelation brings them within the reach of all, easily and without any admixture of error. 183
  • 190. AN INTRODUCTION TO PHILOSOPHY forward by men and particularly by philosophers a sign indeed of the weakness of the human under-standing, but a reason to prize the intellect the more dearly and to embrace truth the more ardently, and an instrument for the advancement of knowledge by the refutations and explanations which these errors call forth. And, on the other hand, it recognises that reason is our sole natural means of attaining truth, but only when formed and disciplined, in the first place and pre-eminently by reality itself (for our mind is not the measure of things, but things the measure of our mind), secondly by teachers (for science is a collective, not an individual, achievement, and can be built up only by a continuous living tradition), and finally by God, if he should please to instruct mankind and bestow upon philosophers the negative rule of faith and theology.* Philosophy of Aristotle and St, Thomas {Moderate Intellectualism) That which really is the cause of truth in the mind. Reason is capable of attaining with complete certainty the most sublime truths of the natural order, but with difficulty and only when duly disciplined. Error by Defect Error by Excess Reason is incapable of Reason attains truth in attaining truth, which every sphere easily and either is wholly inaccessible without any need of sub-to man (scepticism) or must mitting to any external be sought otherwise than discipline (rationalism). by the intellect (anti-intel-lectualism). Synthesis of these Two Errors The mind of man makes the truth of that which he knows (namely, phenomena), and that which really is, the thing in itself, is unknowable by reason (criticism or Kantian agnosticism). 1 See above, p. 124. 184
  • 191. CRITICISM (EPISTEMOLOGY) Another question, among those with which criticism is concerned, demands consideration here. The intellect or reason being the instrument of philosophy, what is the formal object of the intellect, to which intellectual knowledge relates directly and in itself? To answer this question it is suflficient to ask oneself whether there does not exist an object which is always present to the mind when the intellect functions ? Such an object does exist. Whatever I know by my intellect, there is always some being or mode of being present to my mind. There is, however, nothing else except being which is always present in this way. If, for example, I think of a quality, a magnitude, or a substance, in all these cases alike I think of some being or mode of being ; but there is nothing except being which is common to these three objects of thought, and therefore present in all three alike. We therefore conclude that being is the formal object of intellect, that is to say, the object which it apprehends primarily and in itself (per se primo) and in function of which it apprehends everything else. To know the cause of a thing, its purpose, origin, properties, and relations with other things, is in these various ways to know what it is, to apprehend its being under those different aspects. To use the understanding without the notion of being arising is an impossibility. The intellect, moreover, is able to apprehend the being of bodies in their sensible appearances (pheno-mena). It is thus, for example, that in physiology it studies the properties of living organisms in reference to causes which themselves belong to the sensible order. Of this nature are the sciences of secondary 185
  • 192. AN INTRODUCTION TO PHILOSOPHY causes or the sciences of phenomena. But the intellect can also apprehend the being of things in their first principles. This is the function of philosophy as a whole, which in turn is subdivided into natural philosophy and metaphysics^ according as the being apprehended in its first principles by the intellect is the being of bodies as such or being simply as being. Psychology indeed deals with this question of the formal object of the intellect. But the distinctive function of criticism is to make clear that the being with which we are here concerned is indeed the actual being of things, which exists in them independently of the knowing mind. To maintain on the contrary that the object of our intellect is not the being of things but the idea of being which it forms in itself, or more generally that we apprehend immediately only our ideas,' is to deliver oneself bound hand and foot to scepticism. For if that were the case, it would be impossible for our mind under any circumstances to conform itself to that which really is, and truth would therefore be unattainable. Moreover, the intellect would stand convicted of falsehood, for what the intellect professes to know is what things arc, not what its ideas are. In reality ideas, as the conscious-ness of every man witnesses immediately, are our instruments of knowledge. If, therefore, knowledge did not apprehend the things themselves, knowing would be an operation or activity without end or object, which is absurd. For to form an idea or judgment is to know, just as to make use of a knife is to cut. And, just as it is impossible to cut without cutting something 1 The doctrine of Descartes and after him of all subjective philosophy. 1 86
  • 193. — CRITICISM (e PISTEMO LO G y) —the end or object of the act of cutting, which is not the knife, but the thing cut by it—so it is impossible to know without knowing something—the end or object of the act of knowing, which is not the idea, but the thing known by it.^ Conclusion X.—The formal object of the intellect is being. What it apprehends of its very nature is what things are independently of us. From the two truths just enunciated, the intellect is a truthful faculty, and being is the necessary and immediate object of the intellect, there arises as a corollary a funda-mental truth. By intelligible we mean knowable by the intellect. But to affirm that being is the necessary and immediate object of the intellect, and that the intellect attains true knowledge, amounts to saying that being, as such, is an object of which the intellect possesses true knowledge ; that is to say, that it is intelligible. And to say that being as such is intelligible is to say that intelligibihty accompanies being, so that everything is inteUigible in exact proportion to its being. We therefore conclude Conclusion XL—Being as such is intelligible. Everything is intelligible in exact proportion to its being. * Intellectual knowledge comes into existence by means of ideas. But ideas are simply that by means of which {id quo), not that which {id quod) we know directly, a pure medium of knowledge, not (unless reflexively) an object or term known. This is why we say that the being of things is the immediate object of our intellectual knowledge (by immediate we mean known without the intermediary of another term or object pTiviously known). 187
  • 194. AN INTRODUCTION TO PHILOSOPHY It must be borne in mind that when we affirm that everything is intelligible in exact proportion to its being we mean intelligible in itself, to intellect, not intelUgible to us, to our intellect. If, indeed, as a result of the inferiority of human nature, our intellect is disproportioned to a being which exceeds it because it is superior to man, that being, though in itself more intelhgible, will be less intelUgible to us. This, how-ever, is the case with all wholly spiritual natures, and pre-eminently of God. In himself he is the most intelligible of beings, but his intellect alone is pro-portionate to this supreme intelligibility. 1 88
  • 195. ONTOLOGY : ESSENCE Having in criticism examined and defended the principles of knowledge in general, whether scientific or philosophic, we can proceed to the study of meta-physics in the strict sense, that is to say, the science of being qua being. This is the very heart of philo-sophy. We have now to consider being as such, and the great truths it contains in itself; to inquire how it enters into all things without being exhausted by any ; to study its inseparable properties, unity, truth, and goodness, to which we may add beauty ; and finally to treat of it in its activity, and attempt to penetrate the nature and modes of causation. We must also examine how throughout the entire created universe being is divided, whether we consider the constitution of all created being (division of being into potentiality and act^ essence and existence) or the different kinds of created beings (division of being into substance and accident). We shall then realise that the concepts elucidated by ontology are the key to everything else. Certain among them are indeed so indispensable that we must consider them here, for indeed at every turn we are obliged to invoke the primary concepts of essence, of substance and accident, oi potentiality and act. Though it is obviously impos-sible in a mere introduction to give an analysis and complete defence of these concepts, we shall try to 189
  • 196. AN INTRODUCTION TO PHILOSOPHY establish them with all due care, employing, it is true, examples rather than developed arguments, and simplifying matters considerably, but following never-theless the order demanded by a strictly scientific study. Although the notion of being, since it is the first and best known of all, is evidently too clear in itself to admit of definition in the strict sense, the first task incumbent on a man who wishes to think seriously is to clarify this notion in his mind, and with that object to discover the primary conceptions into which it is divided.' We shall therefore begin by asking the following question : What are the objects of thought which inevitably and from the very outset impose themselves upon the intellect when it considers being as such, or to put it in another way, since being is the primary object of intellect, what are absolutely the first data of the intellect ? * We shall see that this one fundamental question admits of three answers according as we adopt the 1 Cf. Aristotle, Metapk., v. ' The notions explained in pp. 1 9 1 sqq. present some difficulty to beginners on account of their extremely abstract character. It is, however, impossible to omit them, for they are literally of primary importance. And in particular we are convinced of the urgent necessity to define with the utmost care, from the very outset, the fundamental concept of essence. Materials for the study of this concept are scattered in different places, but is it not because we have forgotten to collect them that the term essence, when we meet it on the threshold of metaphysics, arouses to-day such suspicion, and, even if it forces itself on our acceptance, leaves such vagueness in the mind ? The student must therefore devote particular attention to the study of the notions here explained, without, however, attemjiting to comprehend them perfectly. For the moment it will be sufficient to make his first acquaintance with them. Later when he meets them again in ontology, after he has become more familiar with philosophy, they will seem much easier. 190
  • 197. ontology: essence . standpoint of intelligibility, of existence, or of action. The consideration of the first of these standpoints will lead us to determine what is meant by essence, of the second to determine what is meant by substance (as opposed to accident), of the third to determine what is meant by act (as opposed to potentiality) ESSENCE We shall first consider being from the standpoint of intelligibility ; that is to say, we shall consider being so far as it is adapted to enter the mind, or is capable of being apprehended by the intellect. This is the most universal standpoint we can adopt, for we have seen that being as such is intelligible and hence that intelligibility is co-extensive with that which is or can be. To the primary datum of the intellect from this point of view we shall give the name of essence. (a) To consider being from the standpoint of intelUgibiUty or as it is capable of being apprehended by the intellect is in the first place to consider it so far as it can be simply presented to the mind without affirmation or negation—in so far as it can be the object of simple apprehension, as we shall term it later. Triangle, polygon, seated, this man, are so many objects simply presented to the mind without being affirmed or denied. From this point of view the primary datum of the intellect is quite simply that which is placed at tlie outset before our mind when we form the conception of anything, or conceive the idea of it. Since we have agreed to use the term essence in this sense, we conclude 191
  • 198. AN INTRODUCTION TO PHILOSOPHY that an essence is that which in any object of thought whatsoever is immediately and primarily {per se primo) presented to the intellect : id quod in aliqua re per se primo intelligitur. Every idea whatsoever, unless it be, like the idea of a square circle, a pseudo-idea involving a contradiction, brings immediately before the mind something. The something thus immediately presented to the mind is an essence or a nature. When I think of man, humanity, animal, goodness, white, whiteness, seated, triangle, etc., each of the objects thus immediately presented to my mind, each of these intelligible units is by definition an essence in the wide sense of the word.^ An essence therefore is simply an object of thought as such. Every essence, however, possesses its intelli-gible constitution which distinguishes it from others and involves certain attributes. Here, however, an important observation must be made. If I consider the triangle with its properties, man, humanity, etc., they remain exactly what they are as objects of thought, whether I suppose them actually to exist or not. The fact of existence does not in any way affect essences as such. To conceive them I 1 We have already seen that the individual as such is not directly apprehended by our intellect. When indirectly, by a reflection on the images (see above, p. 172), we form an individual concept, the object presented to our mind by this concept, Peter, this man, this tree, is abo, so far as it is an object of thought, an essence in the wide sense of the term. That is to say, the concept of essence in the wide sense must be extended even to individual objects of thought. As for those conceptual beings {blindness, for example, or nothingness) which present nothing that really exists, the name essence is inapplicable to them for the reason that a privation as such has obviously no essence. (See St. Thomas, De Ente et Essentia, i.) Nevertheless, from our present standpoint, we may call them improperly essences, in the wide sense. 192
  • 199. ontology: essence abstract from the fact that they do or do not actually exist. We thus perceive that being in the sense of existence and being in the sense of essence belong to two distinct categories.^ The term being has two wholly different meanings. For example, in the quotation "to be or not to be, that is the question," being means existence^ but on the contrary in the phrase a living being it means essence. In the first case the term being signifies the act of being, the act, if I may so put it, which posits a thing outside nonentity, and outside its causes {extra nihil, extra causas) ; and in the second case it signifies that which is or may be, that which corresponds to some existence actual or possible. We may therefore say that being is didded into essence and existence. BEING (that which is : essence in the wide sense (essentia) (ens) {entitas) act of being : existence (existentia) The relationship which obtains between these two terms is a problem which we shall study later ; it is, beyond question, not simply with reference to ourselves, Uke the problem of universals, but in itself the fundamental problem of philosophy : are essence and existence really distinct in all things except God ? Actual existence, the fact of existing actually, is not ^ Observe that in existence itself we may distinguish two things : existence as the fact of existing {existere in actu exercito) and existence as an object of thought {existentia ut quod quid est) . Regarded from the latter point of view existence itself assumes the objective status of every object of thought and confronts the intellect as a particular essence or quiddity. Esse dupliciter sumi potest, scilicet in actu exercito ipsius existentiae, et per modum quidditatis ; et ut exercet existentiam, addit supra seipsum ut quod quid est ; et consequenter ut objectutji intellectus est abstractius {quam ut objectum voluntatis) : quia est objectum voluntatis secundum quod stat in actu exercito existentiae, intellectus autem secundum quod rationem habet quidditatis cujusdam in seipso. Cajetan, in I, q. 8a, a. 3.
  • 200. AN INTRODUCTION TO PHILOSOPHY included in the object of any of our ideas as such. Our intellect can ascribe actual existence to a particular object of thought only by basing itself directly or indirectly by means of ratiocination on the witness of our senses (or reflexively of our consciousness). Thus it immediately judges sensible objects exist, I exist, and demonstrates the existence of God by arguing, for example, from the observed fact of motion. It cannot by itself alone attain the actual existence of the objects of its thought. Those essences, on the other hand (such as triangle, even numbers, humanity), which of their nature connote only a possible existence (for which reason they are also called possibles) , are data furnished immediately by our intellect and ideas. We must now examine more closely this notion of essence, or being understood as that which is or can be. We have just defined an essence : that which in any object of thought whatsoever is immediately and primarily presented to the intellect : id quod in aliqua re per se primo intelligitur. Let us see whether this extremely wide concept (for it is applicable to any object of thought) may not be subdivided and qualified in such a way that the same definition taken in a more restricted sense shall henceforward be applicable, in each particular instance, only to a particular object of thought. {b) The mere presentation to the mind of an object of thought {man, white) is but the beginning of intellectual knowledge, which is perfect only in the judgment by which the mind affirms or denies this object of thought in reference to another {Peter is a man, this flower is white). If then we would consider 194
  • 201. ontology: essence being from the standpoint of intelligibility, to discover what is from this point of view the absolutely primary datum of the intellect, we must consider objects of thought so far as they can be apprehended by the intellect when it judges, for example when it affirms that Peter is a man. From this point of view which, among the various objects of thought which can be realised in a given subject, is that which the intellect apprehends immediately and before everything else ? We shall call it essence in the strict sense of the term. Consider any object of thought, for example, Peter, Paul, this dog, this bird: Peter is tall, Paul is laughing and moving, this dog is barking, this bird is flying. Each of these is a particular whole, individual, concrete, and independent, completely equipped for existence and action. It is individual subjects of this kind that our mind apprehends before anything else (from the standpoint of existence) when we think of that which is. When applied to objects of this kind the expression that which is acquires a more definite and special force. It no longer simply means that which corresponds to some actual or possible existence, but that which fulfils in the strictest sense and before everything else the act of being. These objects are all, though in very diverse respects, actors in the drama of the universe. When, however, we adopt the standpoint of intelligibility, our mind does not among the different objects of thought which things can present apprehend in the first place these individual subjects as such. On the contrary the individual, as we have seen above, escapes the direct grasp of the intellect. What I know of Peter is what I know he is—for example, a 195
  • 202. — AN INTRODUCTION TO PHILOSOPHY man. It is such objects of thought as man or humanity which it perceives in Peter, or such as white or whiteness which it perceives in thisflower, it is what a thing is that, from this point of view, our mind primarily apprehends, and it is therefore in tliis direction that we must look for the absolutely first datum of the intellect in relation to intelligibility (essence in the strict sense). The concept of essence in the wide sense has thus been subdivided into two. There is in the first place that which in the strict sense is, that which. And in the second place what a thing is, what. {that which is : essence j what in the wide sense that which act of being : existence That which is in the strict sense we shall entitle the primary subject of existence and action. It is what philosophers also term suppositum and person. For the moment we may neglect it, for, as we shall see, it does not concern our present inquiry. Let us, on the contrary, consider what a thing is. In the notion of what a thing is there are further distinc-tions and exclusions to be made, to determine more precisely what is actually the absolutely primary datum of the intellect from the standpoint of intelHgi-bility and therefore deserves to be entitled essence in the strict sense of the term ; what, for example, is the essence of Peter. Peter is seated. Peter is capable of laughter. Peter is a man. Is what is here predicated of Peter seated, capable of laughter, man—in each of these three cases, or in one alone, the being which the intellect apprehends in Peter immediately and primarily from the standpoint of intelligibility ? We said above that every object of thought is, as such, 196
  • 203. ontology: essence an essence {essence in the wide sense). Now we are studying what Peter is, and inquiring what is the object of thought which constitutes the essence of Peter {essence in the strict sense). The following are the characteristics of the object of thought thus defined, that is, of the being primarily apprehended by the intellect when it considers what a thing is. It is at once plain that the being to which the intellect is directed in the first place when it thinks what a thing is, is a being which the intellect cannot conceive that thing lacking or deprived of. It is in fact in terms of that being that the intellect immediately conceives, apprehend?, grasps, sets before itself and names the object in question. To deprive the thing of that being, or to alter its constitution in any way, would be to set before the intellect, by definition, a different thing. It is thus a being which that thing so far as it exists cannot lack or be deprived of (otherwise the intellect would not be truthful). For example, Peter, so far as he exists, cannot be other than a man ; on the other hand, he can be not seated. The being in question is therefore a being which the thing considered by the intellect is necessarily and immutably. Moreover, it is obviously the being which in the thing possesses primary importance for the intellect, since it is that to which the intellect is first directed. It is thus the being which before anything else ^ the It Is plain that the word before denotes in this connection a priority of nature, not of time.
  • 204. AN INTRODUCTION TO PHILOSOPHY object is, and is, so to speak, the ground of what the object is in other respects. It is the first being of the thing. Peter, for example, is a man before he is capable of laughter or mortal. We conclude that the being to which the intellect is in the first place directed when it thinks what a thing is, is its necessary and first being, or, in short, the being which constitutes the thing, what it necessarily and primarily is. This is the first characteristic of what we h^ve agreed to call essence in the strict sense. There is a second. It was the standpoint of intelligi-bility, it will be remembered, which we adopted when we undertook this study of essence. Peter is a man {rational animal) before being mortal. That is, man includes animal, and in the notion of animal the intellect finds the necessary characteristic, mortal. The characteristics mortal and capable of laughter— necessarily possessed by Peter—have in him a principle and ground, which by its very notion, or by what it is, or its own intelligibiUty, compels the intellect to posit these characteristics, and this principle or ground is one of the elements or aspects which con-stitute the being man. It is from the standpoint of intelligibility that Peter is a man before being mortal or capable of laughter. Thus if the being man is, as we have said, first, it is in the order of intelligibility that it is first. In other words, it is in Peter the first principle of intelligibility.^ However long our formula, we must say, if we would express this truth exactly, that the being man is in 1 Non enim res intelligibUis est nisi per suam definitionem et essentiam, St. Thomas, De Ente et Essentia, i. 198
  • 205. ontology: essence virtue of its constituent elements or aspects the root of all the characteristics necessarily possessed by Peter ^ which have in Peter a principle which by its very notion requires them. This, then, is the second characteristic of what we have agreed to call essence in the strict sense ; of the being to which the intellect is directed in the first place when it considers what things are. It is in the thing the first principle of intelligibility. Our intellect apprehends this being which is the first principle of intelligibility in two ways, one imperfect, the other perfect. If, for example, we know that an object is a man, without, however, being able to state what man is, we possess a confused knowledge of the being in question. Our intellect grasps that being, has truly apprehended it, and really perceives it, but, so to speak, after the fashion in which our eyes see an opaque object. If, however, we know this same object and are able to define what it is {an animal endowed with reason) we now possess a distinct knowledge of the being in question. Our intellect not only perceives it, but also perceives its principles or constituent aspects. In the first instance the being in question is pre-sented to us imperfectly, in the second perfectly, with the perfection demanded by science, so that we can employ it as a first principle of intelligibility. (For example, from the knowledge that this thing is endowed with reason I can deduce that it is capable of speaking, laughing, worshipping God, etc.) But in both instances it is obviously the same being which is presented to 1 These characteristics are termed properties.
  • 206. AN INTRODUCTION TO PHILOSOPHY US. Therefore, though I do not yet know or even can never know this particular being distinctly as a rational animal, in itself it will be none the less (though in this case I do not know how) in virtue of its constituent elements the root of all the characteristics which possess in Peter a principle requiring them by its very notion ; it will be none the less in itself the primary being of the thing as the first principle of its intelligibility. We now know what are the characteristics of essence in the strict sense and are in a position to define it as follows : Essence is the necessary and primary being of a thing as the first principle of intelligibility, or, in other words, what a thing necessarily and primarily is as intelligible, in short, the primary intelligible being of a thing : id quod per se primo intelligitur in aliqua re.^ fthat which is : (what a thing is primarily i essence in the essence in the - as intelligible i strict sense BEING-! wide sense that which : ihc sih}tc. oi action {suppositum, person) act of being : existence 1 Essence considered as an attribute of the thing (for example, man when we say Peter is a man) is strictly what the thing is necessarily and primarily as intelligible. Essence considered separately and in the pure state (for example, when we speak of humanity or the being man, we cannot say Peter is humanity or Peter is tlie being man) is strictly that in virtue of which a thing is what it is necessarily and primarily as intelligible, or, to put it in another way, that in virtue of which it is constituted in a determinate degree of primarily intelligible being. If therefore we consider essence in the pure state, we must substitute in our synopses for the expression what the expression that in virtue of which : (that which is : (that in virtue of which a thing sessence in the essence in the ! what it is primarily as intelligible/ strict sense wide sense that which : the subject of action {suppositum, person) nct of being : existence 200
  • 207. ontology: essence Conclusion XII.—The essence of a thing is what that thing is necessarily and primarily as the first principle of its intelUgibility. This primary datum of the intellect is termed by philosophers not only the essence but also the quiddity and nature. It is what Aristotle and the schoolmen called the to xt, ^v elvat, the quod quid est,^ and which they defined as id quod per se primo intelligitur in aliqua re,'^ a definition with which we were already acquainted, but to which we have now attached a completely definite sense. The definition, when used of essence in the wide sense, meant what a particular idea first presents to the intellect. When employed of essence in the strict sense, it means what a particular ?,ih]tct primarily isfor the intellect. Observe that every object of thought, every essence whatsoever [essence in the wide sense) is in fact the essence of something [essence in the strict sense) appre-hended more or less completely (in some or other of its properties) . When I think of animal, I apprehend the essence of^ Peter in one part of its properties. When I think of man I apprehend it as a whole. When I think of Aryan, Breton, or Peter I apprehend it as a whole with the addition of certain characteristics or attributes derived from the matter (see below, pp. 207-2 1 6) . When I think of a living body endowed with sensibility I apprehend the entire essence of the subject 1 The Latin equivalent of the Greek term is quod quid erat esse—as St. Thomas explains {De Ente et Essentia, i), id est Iwc per quod aliquid habet esse quid, that which makes any object of thought this or that particular thing. 2 Or stated more fully : id quod primo in re concipitur, sine quo res esse non potest, esique fundamentum et causa ceterorum quae sunt in eadein re : ut animal rationale est hominis essentia. 201
  • 208. AN INTRODUCTION TO PHILOSOPHY animal (and at the same time the essence of the subject Peter in one part of its properties) . When I think of white or prudence, I perceive the essence of a particular quality. When I think of goodness, unity, being, I apprehend a certain created participation of the Divine Essence (or I apprehend, by analogy, if I think of subsistent goodness, etc., the Divine Essence itself). Observe, further, that every subject capable of forming part of any proposition whatsoever ^ has an essence distinctively its own, whether it be an individual subject such as Peter {substantia prima, subject par excellence), an abstract and universal subject {sub-stantia secunda) such as animal, an accident, for example, a particular colour or virtue, or a transcendental, for instance the one, the good, etc. The primary intelligible being of a thing is called essence {essentia) because since the intellect is modelled on being, what a thing primarily is for the intellect must be that which is of primary importance in it from the standpoint of being itself ; in fact, as we shall see later, it is by and in its essence that a thing possesses being or existence {esse).^ It is called quiddity {quidditas) because it is that which the definition expresses and declares, which in turn answers the question quid est hoc ? What is this ? It is called nature {natura) because it is the first principle of the operations for the per-formance ofwhich the thing has come into being {nata) .* 1 With the exception of conceptual beings, which do not, strictly speaking, possess an essence (see above, p. 192, note). 2 Essentia dicitur secundum quod per earn et in ea res habet esse. St. Thomas, De Ente et Essentia, i. 3 Qjtidditas est ipsa rei entitas considerata in ordine ad definitionem explicantem quid ilia sit. Entitas vera rei considerata in ordiru ad esse, dicitur essentia ; in ordine ad operationem dicitur natura. 202
  • 209. — ontology: essence The formal object of the intellect is being. On the other hand, what we have agreed to term essence is nothing but the primary intelligible being of a thing. Our intellect can, therefore, really apprehend the essences of things.^ To deny this would be to deny the intellect itself, and to say that it is bound to miss what is peculiarly its object. Moreover, our intellect claims to give us knowledge of the essences of things. The sciences which are its work have no other aim than to grasp these essences, either distinctly, to deduce from them the properties of a thing (as when we know that a particular figure is a right-angled triangle, or that Peter is a rational animal), or confusedly, simply in order to place a thing in its species and describe it (as when we know that a particular body is sulphurid acid, a particular plant alisma plantago). If therefore our intellect were incapable of really attaining the essences of things, it would deceive us. It is therefore an absolutely necessary consequence of the fundamental axiom that our intellect is trustworthy, that Conclusion XIII.—Our intellect is capable of knowing the essences of things. We do not maintain that the intellect always knows the essences of things (in the totaUty of their properties). The specific essences of things are often 1 It can also attain directly (by an appropriate concept) these complete, i.e. completely determined, essences, at least in the case of things immediately accessible to us, namely, bodies. (We attain, for example, the complete, i.e. completely determined, essence of Peter, when we know Peter not only as a living being or as an animal, but as a man.) 203
  • 210. AN INTRODUCTION TO PHILOSOPHY unknown to us, and undefined. This is due to the imperfection of the human intellect. But we maintain that our intellect is capable of knowing them—and therefore does actually know them in many cases. Neither do we maintain that the intellect can always know the essences of things perfectly, that is distinctly.^ That it can often know them only confusedly * matters Uttle. What is certain is that it is capable of appre-hending them. The eye, to take a parzdlel case, sees the coloured objects within its compass with more or less distinct detail ; it may require the assistance of a magnifying glass, but it can see them. It is important to bear in mind that the experimental sciences are very far from being able to know perfectly the essence of the things which they study. They are, in fact, unable to attain a truly distinct notion of their essences, and never possess more than a confused or purely descriptive notion of them. They know them, so to speak, after the fashion of a blind man by means of indirect signs. For example, we know distinctly the essence or nature man when we distinguish man from the other animals by the specific difference endowed with reason. But we cannot know in the same fashion how the dog, for example, differs from the lion ; we know it only by differences of a purely descriptive kind. Often even when we have before us a series of concepts of diminish-ing generality, for instance, a living body, animal, irrational animal, vertebrate, mammal, canine, dog, poodle, 1 Even so it can determine them only by means of an evident character previously known by us (for instance, the faculty of reason in human nature) which it perceives to be a necessary factor of its constitution. * See above, p. J99. 204
  • 211. ontology: essence etc., ending with Gyp or Fido, we may not know what concept {canine ? dog ? poodle ?) designates (in the totality of its properties) the essence of Gyp or Fido. This, however, does not alter the fact that somewhere in the series of concepts in the Ust just given, and any others which might be inserted among them, there must necessarily be a concept which designates that essence. (In fact, in the example we have chosen, it is the concept dog, as zoology discovers by indirect signs, and without being able to give us a truly distinct knowledge of the essence thus apprehended.) When we think of man, for example, or any other object directly presented to the mind by a human idea (an abstract idea), we put before ourselves something stripped of individuality, something which, being apprehended by a single concept, constitutes in our mind a single, and solitary object of thought—which is therefore in our mind something belonging to one {man) and capable of existing in many (in all men), that is something universal.^ Thus everything directly apprehended by an idea of our intellect—and con-sequently the essence of a thing—is in our mind as a universal. No doubt, considered as it exists in reality, the essence is individuated, for it is then identical with a subject, Peter for example, who is himself individual.* 1 See above, p. 159. 2 If, for instance, I can say Peter is a man, it is because the thing (the material object) apprehended under the object of thought man is identical with the thing apprehended under the object of thought Peter. When I thus proceed from the existence of things in my mind to their existence in reality, I must say that the object of thought man, single in my mind, is multiplied in all tlie individuals in which it is realised and is identical with each. 205
  • 212. AN INTRODUCTION TO PHILOSOPHY But this condition of individuality is no part of the very nature or inmost constitution of the essence, does not belong to the essence of Peter as such, to its character as an essence. If indeed the essence considered in itself {secundum se) were individual, our intellect could never know it, for everything directly apprehended by an idea of the intellect is apprehended as a universal. Considered in itself {secundum se) the essence is neither universal nor individual. It abstracts from every condition and mode of existence, being purely and simply what the object is primarily as intelligible and what the definition expresses.. Thus it is equally present in the actual thing, individuated (in order to exist) and, in our mind, universalised (in order to be known). For example, we see a man only in public, therefore in complete dress, whereas in his bedroom he wears pyjamas. Nevertheless the man we know, when we see him in the street, is the same man, because his pyjamas are no more part of his nature than liis suit of clothes neither belongs ; to the man considered in himself. Similarly considered in itself the essence is not universal, but neither is it individual ^ that ; is to say, the essence as such, the essence of Peter taken in itself, abstracts from all the characters which distinguish Peter from Paul or John.' 1 Cf. Aristotle, Metaph., vii, 8, 1033 b 22 ; 10, 1035 b 14. Here wc are speaking only of corporeal things, which alone are immediately accessible to us (being connatural to the human intellect), consequently the only things whose essence is directly knowable (otherwise than by analogy) and can be known complete, i.e. completely determined. 2 From all that has been said it follows that, when, for example, we say Peter and Paul possess the same essence or the same nature, the word same refers to the essence of Peter and Paul as it exists in the mind (for then it is one and the same object of thought), not as it exists in reality 206
  • 213. . ontology: essence Conclusion XIV.—The essences of things are universal in the mind, and considered in them-selves neither universal nor individual. This proposition is of the first importance. To deny it inevitably involves suspicion of the human intellect, which cannot directly apprehend in its concepts the individual as such ; ^ we shall either demand from it what it cannot give, a knowledge strictly superhuman —intellectual intuition of the individual—or deny its objective reference and fall into subjectivism. We must therefore bear firmly in mind that to know the essence or nature of anything it is not necessary to know the principles which constitute its individuality,* since the essence, considered in itself, is, in fact, nothing individual. Misconception of this fundamental (for then it is identical with Peter and with Paul, two different individuals). But since the essence in question is not individual in itself {secundum se), in other words is not distinct in Peter and in Paul qua essence, it foUovre that it is in Peter and Paul such that it can be appre-hended by the mind in a single concept and constitute in the mind one and the same object of thought. This is expressed by the statement that the essence formally universal in the mind is fundamentally universal in things or in reality. (The nature of anything exists in the mind either in a condition of logical or formal universality, as, for example, when we say man is the species of Peter and Paul, or in a condition of metaphysical or fundamental universality, as when we say man is mortal. The term fundamental here refers to the proximate foundation of universality. When, on the other hand, we say that the nature or essence is fundamentally universal in reality, we are speaking of the remote foundation of universality) 1 We are now speaking of things known by our intellect and do not deal with the question how the intellect knows its own individual and material act. 2 Cf St. Thomas, De Verit., q. 2, a. ^, ad i ; Intellectus noster singularia non cognoscens propriam habet cognitionem de rebus, cognoscens eas secundum proprias rationes speciei. 207
  • 214. AN INTRODUCTION TO PHILOSOPHY truth is at the bottom of the errors of several great modern metaphysicians, Spinoza, for example, and Leibniz (exaggerated intellectuaiists) , also Bergson and the anti-intellectualists of the present day. (a) The essence of corporeal things is universal in the sense just explained. That is to say, in this category of being there are a multitude of individuals possessing the same essence. Individuals possessing the same essence, for example, Peter, Paul, and John, are on the same level in respect of primarily intelligible being ; they are essentially equal. Nevertheless these individuals differ one from another. Peter is fair, short, and sanguine, John dark, tall, and choleric, etc.^ Such characteristics peculiar to a particular individual are not derived from the essence. Otherwise they would be identical in all the individuals which by hypothesis possess the same essence. They are therefore non-essential characters. Nevertheless they are, in fact, unalterable and necessary. 2 If he were not fair, sanguine, etc., Peter would no doubt be a man, but he would not be Peter. We must therefore conclude that these characteristics have their ground in what the object is necessarily and primarily, but as an individual, or, in what we may term the individual nature of the thing. (By individual 1 We are not now speaking of those purely contingent characteristics which distinguish one individual from another, for example Peter is in Paris, Paul in Rome, Peter is rich, Paul poor, etc. We are speaking of those characters which arise out of the constituent being of the individual, the innate characters, which are, radically at least, unalterable. * But in a fashion altogether different from the characters derived from the essence (properties). See below, p. 212, note. 208
  • 215. ontology: essence nature we mean incommunicable to any other object or, if you prefer, wholly circumscribed.) {by In this individual nature we find, as in the essence, the notes necessary and first being. But, on the other hand, and this is the important point, it is not the necessary and first being of the thing as first principle of intelligibility ; it is not the first principle of intelligibility. The individual characteristics such as fair, sanguine, etc., are not, as we pointed out, derived fi'om Peter's essence ; they are not required by it. That is to say, they do not possess in Peter a principle or ground which requires them by its ver/ notion, or in virtue of what it is, that is to say, of its own intelligibility (as, for example, rational requires capacity for laughter). Nevertheless, since they are necessarily possessed by Peter, they have their root in Peter, in Peter's individual nature ; they have there a principle. They must then have as their principle something which does not require them by its very notion, in virtue of its being or of its own intelligibility, something in the notion of which the intellect cannot discover a necessity for these characteristics rather than any others. Therefore, his principle is in itself wholly indeterminate. If neither by its notion nor in virtue of its being or its own intelligibility it requires this rather than that, it is because in itself it has no notion, being, or intelligibility. We are thus led to a principle which of itself is absolutely nothing conceivable, to first matter as Aristotle understood it, something which can enter into the constitution of a being, but is not itself a being. 209
  • 216. AN INTRODUCTION TO PHILOSOPHY If it be admitted that non-being of this sort is part of all corporeal things, and that when itself individuated by some qualification ^ it is the primary root of their individuaUty, it is easy to see that the characters which are derived from the individual nature of the thing, since their primary root is the individual matter with the dispositions it happens to possess at the moment when the thing comes into existence, have, as their first principle in the thing, a principle which does not require them by its very notion—for in itself it has neither notion nor intelligibility ; it requires them solely in virtue of the accidental dispositions it happens to possess at a particular moment. Thus, the individual nature is not the first principle of intelligibility, because it is by its matter that it is the principle of the individual characters.' {c) We have merely sought to indicate here how the obscure notion o^first matter, the study of which belongs In so far as it materia signata quantitate. Obviously beings wholly incorporeal or immaterial {pwre spirits) cannot derive their individuation from first matter. They must therefore be individuated by their essence itself, and each individual in consequence differs from the rest as a horsef for example, differs from a man, each being by himself a specific essence. For this reason in the order of pure spirits there are no two beings essentially equal. And consequently in the case of pure spirits (but only in their case) the essence is something individual and the concept of complete essence identical with that of individual nature. * To avoid any possible confusion be it observed that an individual nature is not unintelligible in itself. It is first matter that is unin-telligible in itself. Though the individual nature is not the first principle of intelligibility, is not the primarily' intelligible being of the thing, it is nevertheless the primary principle of its being, for it is the essence as individuated by matter, and is therefore intelligible in itself. That is why an intelligence more perfect than ours, the Divine Mind for example, can know it directly. 210
  • 217. ontology: essence — to natural philosophy, arises naturally in the mind as soon as it is understood that considered in itself the essence of corporeal objects is not individual, a proposition itself demanded by the fundamental axiom of the trustworthiness of the intellect. We may further point out that since matter, this species of non-being, is present as the ground of individuation (and consequently as the primary root of certain qualifications)^ only in the individual nature (in Peter's nature as such) and not in the essence {humanity), we may regard the essence, the primary intelligible being, as free from all the qualifications due to matter as their primary root or as immaterialised being, ^ in other words as the archetypal being of the thing,' 1 That is the sense of St. Thomas's dictum : formae et perfectiones rerum per materiam determinantur {De Verit., q. 2, a. 2). 2 Aristotle, Metaph., vii, 7, 1032 b 14 : Xtyu 5'ovalav (Lvev CXrji rb tI Jjv ilvai. ^ Cf. St. Thomas, De Ente et Essentia, ii : Haec materia (signata) in definitione honiinis in quantum homo non ponitur ; sed poneretur in definitione Socratis, si Socrates definitionem haberet. In definitione autem hominis ponitur materia non signata . . . from this it follows : (i) That Socrates possesses his essence not precisely as Socrates, but as man, for the essence is that which the definition expresses {cf. De Verit., q. 2, a. 2, at/g) and Socrates, as Socrates, is indefinable, Socrates's individual nature is the essence of man individuated by the materia signata. (ii) That essence taken in the pure state or separately, as for instance when we speak of humanity or the being man, may be regarded as the immaterialised being (stripped of the qualities derived from materia signata), or as the formal being of the thing as a whole (comprising both matter—not individual—and form). It is in this sense that the ancients gave to the essence (itself comprising the matter—not individual and the form) the name oiform {forma totius) : Et ideo humanitas significatur ut forma quaedam. Et dicitur quod est forma totius . . . sed magis est forma quae est totum, scilicet formam complectens et materiam, cum praecisioru tamen eorum per quae materia est nata designari (St. Thomas, De Ente et Essentia, 3). It is important to observe that, although the individual matter (e.g. haec ossa, hae carnes) is no part of the essence or specific 211
  • 218. AN INTRODUCTION TO PHILOSOPHY an ideal being which in the pure state or separately has no existence except in the mind, and exists in reality only as individuated by matter, in the concrete state of the individual nature. We must therefore conclude that there is nothing more in the individual nature than in the essence from the standpoint of primarily intelligible or archetypal being.^ From this point of view all individuals of the same species are on the same level of being ; to nature, on the other hand the unindividuated or common matter {ossa, carnes) is part of it. What constitutes man is not the soul alone, but soul and body together. {Cf. Aristotle, Metaph., vii ; St. Thomas, In VII Metaph., 1. lo, 1492 and 1496, ed. Cathala). This unindividuated or common matter, taken simply as receiving the form and determined by it, and not as the primary root of certain characters (the individual characters) of the subject, is made known to us by the form : materia cognoscitur per formam, a qua sumitur ratio universalis (St. Thomas, loc. cit. Cf. De Verit., q. 10, a. 4 fi? 5, and is part of that which we have here termed the immaterialised being (archetypal being) or formal being of the thing (forma totius, seu potius forma quae est totum). ^ The individual nature contains more (the qualities peculiar to the individual, for example, a particular temperament) than the essence, but only from the standpoint of matter, not from the standpoint of purely intelligible or immaterialised being. The individual characters are no part of that being and add nothing to it in its own order. Remark in this connection that the individual characteristics (fair, sanguine, etc.) from the very fact that they are derived from the matter, are necessary, and unalterable in a totally different sense than are the characters derived from the essence (properties) . The latter are necessary de jure, as derived from a prmciple constitutive of the essence which demands them in virtue of its very concept ; it is absolutely impossible that Peter should exist without being mortal. The individual characters, on the contrary, are only necessary de facto, as derived from particular dispositions of the matter which they presuppose. If it is impossible that Peter should exist without possessing a particular temperament, the existence of that characteristic presupposes certain material condi-tions in virtue of which Peter possesses a particular individual nature, but which are not themselves necessary. Hence these characters can be to a certain extent modified, and are unalterable only in their ground. 212
  • 219. ontology: essence know their (universal) essence is to know all there is to know in them, for the being of Peter as Peter is no more complete or determinate than the being of Peter as man. It is merely more closely circumscribed. We can now understand how, although the human intellect cannot directly know the being of objects in its individuality, its nature as an intellect is not frustrated on that account, nor does it miss its formal object, for it truly knows the being of things so far as it is primarily intelligible or archetypal being.^ Hence, though imperfect, it is neither useless nor untrust-worthy. {d) Be it observed that the synonyms essence, quiddity, and nature, all of which denote a universal, may be stretched to denote some:hing singular, when we consider the essence {humanity, for example) as individuated by matter (in Peter, for instance), or as it possesses in reality a singular mode of existence. Nevertheless, strictly speaking, the term nature alone is compatible with the predicate individual, whereas the expressions individual essence or individual quiddity are incorrect.^ In all this we have in mind things immediately accessible to us, namely corporeal things, which the human intellect cannot apprehend directly in their individuality, because, since it is obliged to abstract from images its wholly immaterial ideas, it is by that very fact compelled to abstract from that which constitutes the materiality of sense know-ledge, namely the individual m.atter. As regards immaterial things {pure spirits), our intellect is equally incapable of apprehending them in their individuality, but for an entirely different reason ; because pure spirits are not immediately "ccessible to us, and we can know them only by analogy, not in their essence, and are unable to apprehend their complete essence. ^ So far at least as the order of corporeal things is concerned. In the order of pure spirits, on the contrary, the essence is individual (see 213
  • 220. AN INTRODUCTION TO PHILOSOPHY We have indeed seen that the terms essence and quiddity are used in reference both to the existence and the definition of anything. The definition, however, can express only the primarily intelligible being of the thing, for it states its constituent elements which are by their very notion principles of intelligibility in it. Hence the definition cannot express the material individuating principles of the thing, and for that reason the individual nature as such is indefinable. Therefore, since the quiddity, what the object is as definable, can only consist in the primarily intelligible being of the object, it must be universal. Similarly, that in virtue of which anything invites that supreme perfec-tion which consists in existence can clearly be nothing but its immateriaHsed being. For it is not in virtue of that in it of which the first principle is matter that it invites existence. Its individuality is merely a condition in which it must be in order to exist. And since the essence, what a thing is, taken precisely as that in virtue of which it receives existence, can consist in nothing but its immateriaUsed being, it must be universal. The term nature, on the contrary, is used in reference above, p. 2 1 o, note i ) . And if we know the essences of spiritual beings after the fashion of a universal, it is because we only know them inadequately and by analogy with the corporeal objects previously known. The expression individual nature is not uncommon in St. Thomas {cf. De Verit., q. 2, a. 5, nature singularis ; Sum. Theol., i-ii, q. 51,3, i, natura individui, etc.). He also uses, though exceptionally, the expression essentia singularis {cf. De Verit., q. 2, a. 7). Whatever may be thought of the propriety of the term, in any case St. Thomas understands by it simply the essence individuated by the matter (not in Spinoza's sense, the essence complete, as an essence, only in the in-dividual). 214
  • 221. ontology: essence to the operations which anything is adapted to perform. A thing, however, does not act solely in accordance with its archetypal or primarily intelligible being, but also as it is subject to particular material conditions and possesses a particular individuaUty. Nothing therefore prevents our diverting the term nature from its primary significance to denote secondarily what a thing is as individual. (what primarily as intelligible : essence, quiddity, nature what as wholly circumscribed (essence individuated by matter) : individual nature [that which is : essence in the wide sense BEING -. that which : the subject of action {suppositum, person) ^act of being : existence Finally we may remark that in a series of concepts such as substance, living body, animal, man, Aryan, Breton, etc., only the concept man, strictly speaking, denotes Peter's essence. The concepts substance, living body, animal, denote only certain elements or intelligible aspects which enter into the constitution of that essence ; in other words, they denote that essence only in one part of its qualifications, and the concepts Arjyan or Breton only as circumscribed and differentiated by certain additional notes arising from the dispositions of matter. Aryan and Breton are thus, like the essence man, universal objects of thought apprehended by the mind in the individual Peter and liberated by abstrac-tion from the conditions of individual matter ; but they are universals whose extension is less than that of the essence, and which belong to a particular class 215
  • 222. AN INTRODUCTION TO PHILOSOPHY (race) divided into a multitude of individuals possess-ing the same essence ; and, since they can be distinguished only by means of characters rooted in certain dispositions of matter they cannot be the subject of a notion strictly speaking distinct, or a true definition. -J 216
  • 223. VI ONTOLOGY : SUBSTANCE AND ACCIDENT Adopting the standpoint of intelligibility, we asked ourselves in the preceding paragraph what is the being primarily apprehended by the intellect from that point of view. We were thus brought to the notioi. of essence strictly so-called, or nature (the two terms may be regarded as synonymous) ; ^ what an object primarily is as intelligible. Let us now consider the being of things no longer in reference to intelligibiHty but in reference to existence. What from this new standpoint is the being which immediately presents itself to the consideration of the intellect, that to which the latter is directed before anything else ? In other words, what is the being primarily apprehended by the intellect as existing ? We have already answered the question.* What the mind apprehends first of all as existing, is beings such as Peter, Paul, this man, this dog, this bird, individual concrete and independent subjects, fully equipped to be and to act, and which we have termed the primary subjects of action, supposita, or persons.^ It is they who primarily fulfil the act of being. 1 See above, pp. 201, 213. ^ See above, p. 195. 2 The name person is reserved for supposita of an intellectual nature, who are therefore masters of their actions and possess the maximum of independence. 217
  • 224. — AN INTRODUCTION TO PHILOSOPHY The subject of action may be thus defined from the standpoint of existence. It exists wholly by itself alone or by its own means, not in the sense that it has no need of a cause [Peter has been engendered and many causes combine to keep him in being) but in the , sense that it is by itself sufficiently disposed to be drawn from nothingness by the causes of its being taken ; separately it possesses in itself or in its own nature everything necessary to receive existence.^ In this sense we may say that it is a being existing by itself [per se) or in virtue of itself, in virtue of its own nature, ens per se existens. Since a being of this kind exists as a whole and in no wise as part of another being or subject in which it exists, we may also say that it exists in itself, in se. A being which exists per se ^ ,or rather a being immediately disposed to exist per se, is thus from the standpoint of existence the first datum of the intellect. Observe further, that when the intellect makes being of this kind its object, it transcends the limits which define the essence in the strict sense or the nature {what a thing is, or rather—if we take the pure essence, abstracted from the subject which possesses it that in virtue of which an object is what it is) . We are now concerned, as we have already * We are speaking here of created subjects. An uncreated (divine) Person possesses in himself everything necessary to exist with an underived existence. When we say that the suppositum is in no way a part of the whole in which it exists, the term whole obviously means a whole tliat is one in itself (see p. 250), not a collective whole, for example the universe. 2 This formula is preferable, because existence itself cannot enter as a constituent part into the definition of anything created. See further St. Thomas, Qpodlib., 2, q. 2, a. 4, arf 2 : ipsttm esse non est de ratione suppositi. 218
  • 225. ontology: substance and accident hinted, with that which is in the strict sense, Peter for example, and not with that in virtue of which Peter is what he is {humanity, the property of Peter in virtue of which he is a man, or his individual nature, the Petrine humanity, so to speak, in virtue of which he is Peter). ' that which is : ( what (or that in virtue of which) BEING. essence in the J wide sense I that which primarily exists : primary subject of action {suppositum, person) act of being : existence To be sure, that which is, Peter, possesses no distinctive characters other than those which constitute what he is or his individual nature. But when I say Peter I conceive this nature as constituting the whole which exists in nothing other than itself.^ When, on the other hand, I say Petefs nature, I conceive that nature as distinct from the whole which it serves to constitute and as existing in him, in that whole.* In short, the 1 Because I conceive it as possessed of a certain mode or fashion of being which philosophers term subsistence or personality, and which terminates it, somewhat as a point terminates a line. In this introduction we make no claim to solve the problem of subsistence (the distinction between nature and person) which constitutes one of the most important problems of ontology. Adopting the pedagogical standpoint of coherent exposition, we have merely sought to present and classify notions, so that their sense and mutual relationship may be understood by a synopsis which is complete from the outset. 3 I conceive it in abstraction from the modality called subsistence or personality which terminates it. Similarly I can consider a line abstracting from the point which terminates it, in which case the line thus considered is simply a part of the whole constituted by the line and point taken together, and exists in that whole. 219
  • 226. AN INTRODUCTION TO PHILOSOPHY subject of action possesses a nature or essence ; the concept of that nature or essence taken as such {what or that in virtue of which) is not the concept of the subject of action {that which) . We will now turn to this nature or essence of the subject of action. We have just observed that the subject of action exists (is capable of existence) in virtue of its own nature or its own essence,^ The nature or essence of the subject of action is there-fore that in virtue of which it is capable of existence pure and simple {simpliciter) ; the nature of Peter considered as the subject of action is that in virtue of which I can say simply Peter exists.^ Existence pure and simple is undoubtedly Peter's primary or tirst existence. But it is not his sole mode of existence. He is sad to-day, yesterday he was cheerful ; to-day he exists as sad, yesterday he existed as cheerful. He has lost the former existence and acquired the latter, but he has not therefore ceased to exist purely and simply. That is to say, he possesses a host of secondary qualifications in virtue of which he exists not only simply {simpliciter) but also under a particular aspect {secundum quid). It is thus that he is a musician or a philosopher, ill or in good health, happy or unhappy, etc. All these qualifications have accrued {accidere) to that which he is primarily as existing, are increments, or accretions, accidents. Philosophy, health, happiness, sorrow ; all these are so ' In virtue of Its essence in the strict sense of the term in the case of a purely spiritual subject, in virtue of its nature in the sense of individual nature in the case of a corporeal subject. (See p. 235, note I.) * That is to say, Vk'ithout regard to any particular point of view, w^ithout modifying my thought by any addition. 220
  • 227. ontology: substance and accident many essences ^ to which our attention has not hitherto been directed, and which are not sell-subsistent in being, but on the contrary subsisting, so to speak, only as coverings of the subject of action. Employing the analogy of sensible objects we may say metaphorically that the latter exists beneath the accidents [substat) and supports them. From this point of view it may be termed a substance.* For example, we say that Peter is a substance. Moreover, since his nature considered precisely as such {what he is, that in virtue of which he is what he is, that in virtue of which he is capable of existence pure and simple), Uke himself exists beneath the accidents, it also is entitled to the name of substance, and we can speak of Peter's substance. We have now distinguished the notion of substance as opposed to that of accident.^ 1 The definition of essence given above (p. 201) is applicable to accidents, if the subject is considered in a particular aspect. Understood in a concrete sense, as attributed to the object (for example sad when we say Peter is sad), the accident is what a thing primarily is as intelligible in a particular respect {being sad is the ground on which Peter possesses certain characteristics which necessarily follow from sadness). Understood abstractly, and separately or in the pure state (for instance when we speak of sorrow) the accident is that in virtue of which a thing primarily is what it is as intelligible in a certain respect. We may further use the term essence no longer in respect of the subject Peter, but in respect of the accidents themselves, and say that sorrow is that in virtue of which a particular passion primarily is what it is as intelligible. 2 The subject of action is also called vwdcrraffis (hypo-stasis), -irpwrov viroKflnevov, primum subjectum attributiorns. 3 Observe that the term substance {substantia) corresponds to the Greek word ovala understood in a restricted sense. The term ovcla primarily denotes essence or nature, but since substances are the first object which the intellect attains, when it considers what exists, they are also for that reason the first object which confronts the intellect with the notion of essence ; in other words they are the first to merit the denomination of essence or nature. Hence the term oiitria, which taken in its most 221
  • 228. AN INTRODUCTION TO PHILOSOPHY 'thai which is : rwhat a thing is (essence or Tin a particu-essence in the nature) and in virtue of J lar aspect : accident wide sense - which it receives exist-j ence y absolutely : ^ Ahat which primarily . , /., , . ^ v ysubstance exists (the subject of | aiction,suppontum, person) j act of being : existence SUBSTANCE The name substance is given, as we have just seen, both to the subject of action itself {that which primarily exists) and to its nature considered precisely as a nature or essence {what a thing is, that in virtue of which the subject of action is what it is and claims existence pure and simple).^ What then shall be our definition general sense denotes essence, and is afterwards divided into substance ajid accident, has most naturally served to denote in a special sense the first member of the pair, substance. 1 The subject of action (suppositum or person) is nothing but the substantial nature completed by a particular modality {subsistence or personality) which terminates it, as a point terminates a line (without adding anything to it in its order of nature) and renders it absolutely incommunicable. The term substance (corresponding to the Greek ovffia, which primarily denotes essence—see the preceding note) denotes the substantial nature without defining whether or no it is terminated by subsistence. It is therefore applicable alike to the nature (apprehended by the mind without the subsistence which terminates it) and to the subject of action (the terminated nature). But when we distinguish and contrast the nature (not terminated) and the subject of action, the term substance remains attached to the nature (not terminated) and is then contrasted with the subject of action taken as such. Thus when we speak of Peter's substance, we mean precisely the nature in virtue of which the subject of action Peter possesses primary being, and which is part of him. And theologians use the term in this sense when they teach that in the Divine Trinity the Father and the Son (two distinct Persons) possess the same substance, are consubstantial, ojuoovaioi. On the other hand, the Greek term vwixTTaai.^ {hypo-stasis, etymo-logically the same formation as sub'Stantia) after a certain vacillation in 222
  • 229. — ontology: substance and accident of substance ? Being receiving existence of itself {per se) or in virtue of itself in the unqualified sense adopted above ? No. For so understood the definition applies only to the subject of action ; absolutely speaking, it alone Peter for example—exists as a whole, and not as part of a being or subject in which it exists. Its nature, on the contrary, is part of itself and exists in itself. Peter's nature exists in Peter, and is part of Peter. It is true that since Peter is himself constituted by it, and exists in virtue of it {per earn), this nature does not exist in something previously existing which receives it (as, for example, sorrow exists in Peter, who was previously in existence). We can therefore say that it exists (is capable of existence) per se, in the precise sense that in order to exist it has no need to be part of another being previously existing which receives it in itself, but that on the contrary it constitutes the whole (the subject of action) which exists by itself. In this sense and, if we are careful to quaUfy and explain our meaning, the description ens per se existens is appUcable not only to the subject of action, but also to its nature, and may therefore serve as the definition of substance.^ (The same is true of the expression ens in se existens.) its employment, came finally to denote the subject of action taken as such [person) and is used exclusively in this sense. It is thus contrasted with substance understood as a nature not terminated by subsistence. Were we tempted to deny the vital importance of these abstract concepts and distinctions, we might recollect that for the word b/j-oovaios, on which the true understanding of the Trinity depends, but which differs only by an iota from the unorthodox term dfioiovcrioi, Catholics were willing, when the Arian heresy flourished, to suffer every kind of persecution, and sometimes even death. * Cf. John of St. Thomas, Cursus philos., i, Log. ii, q. 15, a. i. 223
  • 230. AN INTRODUCTION TO PHILOSOPHY We shall therefore ^ define substance as a thing or nature that can exist by itself or in virtue of itself {per se)—and not in another thing [in alio), that is to say, in a subject previously existing.^ (Alternatively we may define substance as a thing or nature whose property is to exist in itself.) Conclusion XV.—Substance is a thing or nature whose property is to exist by itself, or in virtue of itself {per se) and not in another thing. It is evident that the idea of substance represents something which really exists. If no substance existed, no ix.ture capable of existing in itself, all natures would be such as could exist only in something else. But in that case, since nature A could exist only in nature B and nature B in turn only in nature C, there would be an infinite regression which could never 1 Existence itself cannot be a constituent part of any created nature. It is for that reason that substance must be defined as a thing or nature capable ofexisting per se or apt to exist per se. The same observation was made above in reference to the suppositum (p. 218, note 2). We must make the sense of the suggested definition clear. If per se (or in se) is understood in the restricted sense explained in the text, our definition will mean : substance is a nature apt to exist per se {or in se) qua nature or essence in the suppositum which it constitutes when terminated by the subsistence. If, however, per se (or in se) is understood in the absolute sense in which it was taken above (p. 218), the oposed definition will mean : substance is a nature apt to exist per se {or in be) qua subject of action (suppositum or person). ' The term substance signifies a thing capable of existing in itself, or of subsisting ; that is to say, of being self-contained as an existent thing (its function subsistere), so that, once it exists, it sustains in being the additional qualities or accidents with which it is invested (its function substare). But it is only as a suppositum that substance is immediately capable of performing these two functions. Considered as a nature or essence it merely seeks to perform them. 224
  • 231. ontology: substance and accident reach a nature in which all these natures existed ; they therefore could not exist. Those philosophers who, like Fichte (nineteenth century), denounced the " dead substance of the Latins " to oppose to it " Teutonic action or becoming " were fighting against the intellect itself, which is simply unable to dispense with the notion of substance and imposes it upon us as an absolutely primary and immediate datum. Moreover, that which they took for substance and declared " dead," " inert," etc., was a mere figment of their own imagination. For substance is not an " empty receptacle," " an inert and dead support." It is the absolutely primal being of a thing, the radical principle of its activity and all its actuality. As Aristotle said, substantia est primum ens.^ But to perceive this a philosopher must make use of his intelligence, rise above the animal life of the senses, and not be content to show his skill iu handling words devoid of conceptual content and freighted only with material images. The substance of an object, so long as that object exists, is as such immutable.^ Peter's substance is that in virtue of which Peter exists purely and simply, that is to say, as Peter. So long as Peter exists, liis substance as such cannot change. And when Peter's substance does change (when Peter's body becomes a lifeless corpse) Peter exists no longer, he is dead. 1 Metaph., vii, i. '' No doubt when Peter grows the change affects Peter's substanei itself, which increases, but solely in respect of its quantity. It does not affect it as substance. 225
  • 232. AN INTRODUCTION TO PHILOSOPHY Moreover, in itself substance is invisible, imperceptible by the senses. For the senses do not apprehend being as such, but present to us directly only the changing and the moving. In a certain sense, to be sure, it is indeed the substance of Peter that my eyes see, as it was truly Jesus whom the disciples saw at Emmaus, but my eyes thus apprehend the substance only in fact and materially, not formally. In other words the object seen or touched is some-thing which while seen or touched is at the same time also a substance ; but it is not seen or touched as a substance. As a substance it is conceived, not seen or touched, and so far as it is seen or touched it is coloured or exerting resistance, not being and substance. In the language of philosophy substance is intelligible in itself (per se) and sensible only accidentally (per accidens). That therefore in things which possesses most importance for us escapes the direct grasp of our senses and imagination, and is a pure object of the intellect, since the intellect alone appre-hends being as such (sub ratione entis). Observe that, if from the standpoint of existence substance is in things the being which is the primary and immediate object of the intellect, on the other hand to discover not only that a particular object possesses a substance, but also in what that substance consists, or what is its nature, we are obhged to take our stand upon that which reveals this nature to our senses, namely the operations, phenomena, or accidents, of the substance. In this sense we know the substance by the accidents. 226
  • 233. ontology: substance and accident ACCroENT Consider now such things as the laughter, movement, sorrow, joy, colour, etc., which I perceive in Peter, and which make Peter exist in certain aspects. These things are capable of existence. But they obviously do not exist after the same fashion as substance. To exist they must belong to another being previously existing.^ They exist as something which belongs to a being or subject already in existence. In this sense we say that they exist in something other than them-selves.- Conclusion XVI.—An accident is a nature or essence whose property is to exist in something else. 1 Previously—if not in the order of time, at least in the order of nature. 2 The accident of which we are sp>eaking is the predicamental accident which is contrasted with the substance. The term accident, when it is contrasted with property and signifies a predicate not derived from the essence (the predicable accident) has another meaning. If we are thinking of the predicamental accident, or of the contrast between substance and accident (a contrast between real beings) , we may say that the attribute relates to an accident (the intellect in virtue of which a man is capable of laughter is an accident really distinct from the substance) . If, on the other hand, we are thinking of the predicable accident, that is to say, of the contrast between those unreal beings of logic (the predicables) , genus, species, specific difference, property, and accident, it denotes not an accident but a property, an attribute predicated of the subject, not as something which helps to constitute his specific essence, but as arising necessarily from it. Conversely, if we are considering the predicamental accident, we must say that the individuating characteristics (the possession of a particular temperament, or heredity) belong, radically at least, to the substantial, not to the accidental order. If, on the other hand, we are considering the predicable accident, we must say that these character-istics are accidents (attributes predicated of the subject, neither as helping to constitute the specific essence nor as derived from it). 227
  • 234. AN INTRODUCTION TO PHILOSOPHY It follows that though an accident partakes, indeed, of being, it does not exist as a being ; it is essentially of a being, ens entis, and capable of existence only as the complement or perfection of a being. Thus the word being is predicable of the accident only in a secondary and indirect sense, and whereas being in the primary sense of the term is from the standpoint of existence the subject of action, so that our intellect apprehends immediately and of itself the subject of action, the substance, that which exists in itself, we find it difficult to arrive at a clear understanding of the accident. To succeed we are obliged to elaborate our notion of being, to make it more pliable, plane it down, bind it to the real, in short, to apprehend the accident by analogy with the substance which is contrasted with it. The mere fact that the term accident is a substantive involves us in the danger of regarding the accident as a substance, a piece of substance or a reduced substance. The imagination intervenes, and we depict to ourselves accidents and phenomena as fragments of matter inlaid in the suppositum, like a casing of mosaic or marquetry. Those for whom the words accident or phenomenon evoke images of this kind miss the notion of accident altogether. They conceive in fact only pseudo-substances and are incapable of advancing a step further in philosophy. An original effort of the intellect elaborating the notion of being is here the sole remedy. It is obvious that things such as an act of thought or a movement of emotion cannot be confused with our substance, because they come and go, and change within us, whereas our substance never ceases or
  • 235. ontology: substance and accident changes, being as such immutable so long as we exist. Nevertheless these things are realities which affect us intrinsically. They are, therefore, really distinct from the substance in which they exist or, in technical language, inhere. There are thus contingent accidents (such that the subject can exist without them) real and really distinct from the substance. But if change, by showing that there are in a subject things which come and go, helps us to arrive at the notion of accident, it is very far from being a necessary attribute possessed by every accident. There are things without which a subject cannot exist, and which nevertheless are accidents, additional beings which complete the substance ; our understanding itself, for example, and our will are evidently something real in us. But they cannot be confused with our substance. For we possess a distinct notion of them wholly extrinsic to that of substance,^ which would be impossible if they were not essentially different from the latter.* Therefore the understanding and the 1 We distinctly conceive the understanding or intellect as a faculty of knowing whose object is being, the will as a faculty of conation whose object is the good, substance as a nature or essence whose specific property is self-subsistence. The three concepts fall wholly outside each other. 2 We are able to reason in this way because we are dealing with things proportionate to our intellect, which apprehends them by a proper and distinct concept (things which are, we say, known by their essence). In such cases, if two concepts are wholly external to each other, it is because the things they piesent to the mind really differ one from the other. Otherwise our intellect would deceive us. It is in this way that we prove that quantity or extension is an accident really distinct from corporeal substance, and that in every created thing the essence is really distinct from the existence. (On the latter point see John of St. Thomas, Cursus Phil., Phil. Nat., q. 7, a. 4.) When, however, the distinction is due to the standpoint from which the mind 229
  • 236. AN INTRODUCTION TO PHILOSOPHY will are real things in us distinct from our substance, consequently accidents (belonging as we shall see later to the category of quality) . There are, therefore, necessary accidents (without which the subject cannot exist) real and really distinct from the substance. The different schools which profess conflicting doctrines on the problem of substance may be repre-sented roughly by the following scheme : Philosophy of Aristotle and St. Thomas There are as many substances as there are individuals. In virtue of its substance each of these possesses primary being, but there are in each real accidents really distinct from the substance. Substantialists Phenomenalists There are no real acci- There is no substance ; dents really distinct from the accidents which are the substance, which is the apparent to the senses or sole reality. the consciousness (pheno- Descartes, Leibniz, and mena) are the sole reality, especially Spinoza. The English sensationalists — German pantheists of the neo-Kantian school. Philo-nineteenth century. sophy of pure becoming. Descartes denied the existence of real accidents really distinct from the substance. He identified corporeal substance with extension and the substance of the soul with the act of thought. He thus set views the same object we have indeed two distinct concepts, but not wholly external one to the other. For example, I distinguish in Peter his being a man and his being an animal, though in reality they are one and the same being. But the concept man, far from being external to the concept animal, imphes it on the contrary. 230
  • 237. ontology: substance and accident philosophy on a path which could only lead to pantheism (for if there are no accidents distinct from substance, every substance is its action—a perfection which belongs to God alone—and the concept of, substance becomes identical with that of absolute Being or God, with whom everything is thus confused), or if pantheism is to be avoided, to the denial of substance, which such a philosophy will do its utmost to disprove and to banish from the human intellect. Spinoza erected on the Cartesian foundation a monism or absolute pantheism from which Leibniz attempted in vain to escape by substituting for the single substance of Spinoza an infinite multitude of individual substances (monads), thus in effect replacing Spinoza's God by a boundless host of deities. Though they rejected the notion of substance, for which they substituted that of becoming or evolution, and regarded the thing-in-itself not as an object which imposes itself on the mind but as a background of the mind which produced the object, the German metaphysicians in the succession of Kant (Fichte, Schelling, Hegel) may be ranked among the exponents of pantheistic sub-stantialism since they posited a single principle which constitutes by its development the stuff and reality of all things. In the phenomenalist camp the English sensationalists and associationalists maintained that states of consciousness (sensations, emotions, ideas, etc.) are the sole reality accessible to us, and attempted to reduce the whole of psychology to the mutual association of these states of consciousness. The philosophers of pure becoming (Bergson, in particular, who thus joins hands with 231
  • 238. AN INTRODUCTION TO PHILOSOPHY " Heraclitus over a gulf of twenty-five centuries) deny the existence of anything permanent in things, and maintain that change without any subject of change is the sole reality. (In psychology these philosophers are opposed to the formxr group inasmuch as they substitute " a continuous stream of consciousness (W. James) for a mosaic or " host " of states of con-sciousness. But they are agreed in rejecting the notion of substance.) Kant (eighteenth century) substituted for the dis-tinction between substance and accidents in things (substance and accidents being equally objects of knowledge, the former by means of the second) ^ an opposition of two separate worlds, the world of things as they are in themselves {things-in-thernselves, noumena) and the world of phenomena manufactured by our mind. He regarded the thing-in-itself as wholly unknowable, though he affirmed its existence. This thing-in-itself, sought now in the subject of knowledge, became all in all to the German pantheists of the nineteenth century. Renouvier and the French neo- Kantians, on the contrary, taught that the thing-in-itself, the substance, is not merely unknowable but absolutely non-existent, and the concept of it chimerical. The various phenomenaHst philosophers just men-tioned failed to perceive that what they really deny is the accident, not the substance. What they under-stood hy phenomena is a mere pseudo-substance expressed by a concept ashamed of itself and self-contradictory, a substance pulverised, melted down, emptied of real subsistence, something which is not accident, but 1 See above, p. 226. 232
  • 239. ontology: substance and accident being of a beings the pure complement of being, and inconceivable except as correlated with substance. Since they have never really understood what is meant by substance, and posit, under the appellation of phenomena, pseudo-substances, they quite naturally refuse to admit another substance behind these pseudo-substances of their imagination. THE INDIVIDUALITY OF SUBSTANCE The being primarily apprehended by the intellect from the standpoint of existence (the substance) is something individual. The intellect indeed appre-hends it as individual, for it apprehends the being of things only by turning to the sensations and images which reveal things to us under the conditions of their existence and in their individuality. Moreover, that alone can really exist which is by nature a com-pletely circumscribed and self-contained unit, or an individual, [a) Our intellect, it is true, can have no direct knowledge of this substance in its individuaUty ; it simply knows, by turning to the images from which it derives its ideas, that this substance is individual, it does not know in what its individuahty consists ; Peter's substance is directly revealed to it only by a universal idea. Peter's substance thus perceived, in abstraction from his individuality, is simply Peter's nature apprehended in the characters which constitute his essence, strictly speaking. And since we say of man that he moves, laughs, possesses understanding and will, etc., as we say it (primarily and in the first place) of Peter or Paul—since in consequence the property of existing beneath the accidents, which strictly belongs 233
  • 240. AN INTRODUCTION TO PHILOSOPHY to the subject and his individual nature, is thence transferred to the nature of the subject stripped of its individuaHty by abstraction—we shall also give the name of substance, though in a secondary sense (substantia secunda), to Peter's nature, as apprehended in abstraction from his individuality, that is to say, the universal essence man or humanity. On the other hand, by substance in the primary sense {substantia prima), we shall understand the individual substance.^ It is now clear that when we consider the being primarily apprehended by the intellect in material things we stress either the individual being or the universal being according as we consider this being primarily apprehended in relation to existence or in relation to intelligibiUty. In relation to intelligibility the being primarily apprehended in things by the intellect is the essence, strictly speaking, which in itself is not individual and exists in the mind in a condition of universaHty ; and it is only in an improper sense that the term essence is used of the essence individuated by the individual matter (that is, of the individual nature). In relation to existence, on the contrary, the being primarily apprehended in things by the intellect is 1 In the Aristotelian and Scholastic vocabulary the term substantia prima, ovaia irpLbrrj, denotes (see above note) the individual nature of the subject of action without determining whether or no it is terminated by subsistence. Usually indeed it does in fact denote the terminated nature, or subject of action, the hoc aliquid. It does not, however, denote formally the subject of action taken as such and contrasted with the (non-terminated) nature. That function belongs to the terms suppositwn and persona {vw6(rra.<n$). Remark that the distinction between the subject of action and the nature (non-terminated by the subsistence) is in the main due to the Schoolmen. Aristotle himself did not make it explicitly. 234
  • 241. ontology: substance and accident the individual substance/ and it is in a secondary sense that the term substance is used of the nature stripped of its individuality by abstraction (that is, of the essence in the strict sense) * . Here we may call to mind what was said above ' about the individual nature. We see at once how we should classify the different concepts with which we have made acquaintance hitherto.* what { primarily as intelli- i the gible {essence in the J sub- strict sense) and in | ject j virtue of which it that which is : is exists essence in the-; - primarily as complete- n wide sense ly circumscribed {in-dividual nature) and in virtue of which it exists as such ^ <Aa<u)AicA primarily is: the pri-marysubject ofaction {sup-positum, person) J existence in a particular respect : accident absolutely : substantia ^ secunda ' substarue substantia prima act of being : 1 Taken precisely as that in virtue of which the subject possesses its first being, the substance, substantia prima, is the subject's individual nature. We said above (p. 214) that that in virtue of which a thing is susceptible of existence is the universal essence, the reason being that we were then considering precisely what is the ground on which a thing is susceptible of existence, in contrast with that which is merely a condition or state in which it must be in order to exist. Here, however, we are dealing with that in which the existence of the thing considered precisely as in the state necessary in order to exist is grounded, and this is not the universal essence, but the individual nature of the subject. 2 Hence in the order of pure spirits and there alone (where no distinc-tion exists between the individual nature and the essence, see above p. 2 1 1 , n. 3 ; p. 2 1 3, n. 2), the substance in the primary sense of the word is also the essence, strictly speaking. In the material order, on the other hand, substance in the primary sense of the word is the subject's individual nature, and it is onJy in a secondary sense that the essence, strictly speaking, is called substarue, ' See page 209. * The quod and the quo. We have already observed (see above, p. 200, note 1) that the essence taken in the concrete or as attributed 235
  • 242. AN INTRODUCTION TO PHILOSOPHY (b) Per se, a se, in se. We have defined substance as a thing whose property it is to exist by itself (per se) or in itself (in se). We must determine carefully the exact sense of these expressions. A thing is said to exist in itself {in se) when it does not exist as part of a whole previously existing, but to the thing {what a thing primarily is as intelligible) is not presented to the mind in its purity ; it is in fact presented to the mind together with the thing or subject which it determines. To possess it in its purity, it must be conceived separately, without the thing or subject it determines, as for instance when we speak of humanity or, to force language, of the being man, the entity man. In that case it must be defined as that in virtue of which a thing primarily is what it is as intelligible, or, in other words, that in virtue of which a thing is constituted in a particular degfee of being primarily intelligible. For this reason it will he better to substitute in our synopses for the term what the phrase that in virtue of which. So we finally get the following table : INTELLIGIBILITY BEING^ . that in virtueof which {id quo) Peter exists and is / in a particular respect : what he primar-J whiteness accident ily is as intelli-j gible {esserue in |^ absolutely : the strict sense) humanity . . substantia^ Peter exists and is^ secunda that which is : what he primar essence in thej ily is as complete-wide sense ly circumscribed (essence indivi-duated by mat ter, the individ-ual nature) that which primarily exists : suppositum, person {id quod) act of being : that in virtue of which {id quo) a thing is outside nothingness or its causes : existence Peter's indivi-^ dual nature substance ysubsiantia prima Peter EXISTENCE The distinction between the quod and the ^mo plays a part of the first importance in the metaphysical analysis of things. 236
  • 243. ontology: substance and accident itself constitutes the whole which exists. In this sense Peter exists in himself. A thing is said to exist by itselfor in virtue ofitself (^^r se), when it is brought into existence in virtue of itself, or of its own nature (by the causes on which it depends, if it is a created nature) . In this sense Peter exists j&^r se.^ Philosophy makes frequent use of this expression, per se. It always means in virtue of itself, in virtue of its own essence [per suam essentiam)—whether the quality under consideration forms part of the essence of the thing or necessarily results from it as its principle (in which case per se is opposed to per accidens) ^ or whether we merely wish to state that the attribute under con-sideration immediately pertains to the thing which does not receive it through the intermediary of anything other than its own essence (in which case per se is opposed to per aliud). It is in this sense that the subject of action exists per se, whereas the accident exists per aliud. But the expression per se does not mean in virtue of itself or of its own nature as the absolutely first principle or as the complete and ultimate explanation. This is something totally different, which is expressed by the phrase a se, of orfrom itself (as opposed to ab alio). That which is a se is evidently per se, but that which is per se is not ^ Existence per se or in se can, we have already seen (see pp. 2 1 8, 223) , be ascribed, as it is understood in a more or less strict sense, either to substance in general (that which exists per se and in se contains in itself whatever is necessary in order to receive existence and is not part of a previously existent whole) or exclusively to the subject of action {suppositum or person, which contains in itself everything necessary in order to receive existence, and exists in no respect as part of a whole). * For example, Peter is per se alive, endowed with intellect, and the faculty of laughter, the artist is per se one who fashions objects. But Peter is per accidens a sufferer from influenza or the heir to a large fortune, the artist per accideru celibate or married, etc.
  • 244. AN INTRODUCTION TO PHILOSOPHY by any means for that reason a se. That which exists a se or from itself, possessing in itself the entire explana-tion of its existence, is uncaused ; God alone is from himself, a se. Created substances on the contrary [created subjects of action) are caused ; they exist per se, in virtue of their essence ; they do not exist a se. In their own nature they possess everything necessary to receive existence, but not to possess an existence not received from without. They are sufficient by themselves to exist, in the strictly qualified sense that they do not exist as something which belongs to some-thing else, but, absolutely speaking, they are by no means a sufficient ground of their own existence. That which is a se cannot cease to exist ; that which is per se without being a se can lose its existence. The distinction between that which exists a se and that which exists per se is perfectly clear. Nevertheless certain philosophers have lost sight of it, notably Spinoza, who ascribes aseity [a-se-itas) to every substance. (From which it follows immediately that there is only one substance, and that everything is God—monism and pantheism). When indeed Spinoza defines substance as that which is in itself and is conceived by itself he really means, as the context shows,^ that which to be and to be conceived needs absolutely nothing except itself Descartes had already defined substance ambiguously as res quae ita exsistit ut nulla alia re indigeat ad exsistendum, a thing which exists in such a fashion that it has need of nothing else in order to exist. ' 1 Cf. Ethics, i, 7. " Principia Philosophiae, i, 51. 238
  • 245. VII ONTOLOGY : ACT AND POTENTIALITY When we studied being first from the standpoint of intelligibility, then fi-om that of existence, we saw that the object primarily apprehended by the intellect, being in the primary sense of the term, was in the former case what we call essence, in the latter what we call substance. We must now consider the being of things (under-standing the term being in its most general and indeterminate sense) from the standpoint of action, in reference to the manner in which things behave in reality, or, if you prefer, in reference to what they do. This new standpoint acquaints us with a third primary sense of the term being. (a) What is the first truth which the intellect grasps as soon as it has formed the notion of being ? It is sufficient to consider the notion to see at once that what is, is (principle of identity), or again that what is, cannot not be at the same time and in the same relation (principle of non-contradiction). That is to say, that everything is what it is, that it is not what it is not, and that it is everything that it is. We will now consider what things do, what is their natural behaviour, what is the primary fact of experi-ence grasped by the senses and consciousness. Things change. The arrow flies, the animal runs, what was 239
  • 246. . AN INTRODUCTION TO PHILOSOPHY cold becomes hot under the action of fire, food becomes flesh, what was Uving dies, and every spring that which had no being comes into existence. Although, like all our primary notions, it is very difficult to explain scientifically, everybody knows by experience in what this great fact of change or motion consists. ¥/e may say that wherever there is change there is a transition (from one being to another, or from one mode of being to another). And for transition to exist, there must be something which undergoes it, something which is the subject of change, for example a subject which ceases to be in a particular place or a particular thing {terminus a quo, the arrow pressed to the bow, thefood, the seed), to be in another place or another thing {terminus ad quern, the arrow in the target, flesh, the mature plant) There is no change without a subject whi:h is changed, and which must be some particular thing before the change is effected ^—in other words being is prior to change. Those, indeed, who maintain that change is prior to being, and that there is change without a subject which is changed and which is some particular thing before the change is effected, deny the principle of identity and fall into absurdity. For when they take up this position, they must either continue to accept the notion of being, in which case to affirm that there is change without a subject of change, or that change is prior to being, is to affirm that what has no being 1 For example, in a substantial change the first matter, which is not a being, but potential being, is the subject which is changed and consti-tutes a particular body, or particular being, by its union with a particular substantial form, before it constitutes some other body by its union with another substantial form. 240
  • 247. ontology: act and potentiality changes, which is manifestly absurd ; or they must reject the notion of being as illusory and argue that instead of conceiving being we must conceive change, in which case they must reject as false, together with the notion of being, the principle of identity which is bound up with it, and maintain that thought is essentially deceptive, which is equally absurd. We are, therefore, absolutely obliged to hold that being is prior to change, and that there is no change without a subject which is changed, and which is some particular thing before it changes ; or, in the language of philosophy, that there is no motion without a subject which is moved. (b) We will now turn away from experience and every sensible representation and attempt to consider change with our intellect, that is to say, in terms of being, the formal object of the latter. We shall inquire how or in what respect the starting-point of change can thus become the goal. You will answer, perhaps, that it is according as it is this or that, in respect of what it is, that the starting-point becomes the goal. But the starting-point is nothing but what it is, and is already everything that it is, and therefore in this respect is incapable of becoming, for it already is. You may then say that it is according as it is not this or that, in respect of what it is not, that the starting-point becomes the goal. But in respect of what it is not, the thing is nothing whatever, is pure nothing, and therefore cannot be the source of the product of change. It is incapable of becoming, for it simply is not. Hence the starting-point of change cannot become its goal—either in respect of what it is or in respect 241
  • 248. AN INTRODUCTION TO PHILOSOPHY of what it is not. In other words, the new being which is the product of change can be derived neither from the being which akeady exists, nor from a nothing, which has no existence whatever. Is change, therefore, impossible, as Parmenides maintained ? And are we obliged with him to deny the evidence of our senses, which witnesses to the fact of change ? {c) No. But we are obliged to develop and explore our idea of being. Evidently in the analysis we have just made something has been left out. The starting-point of change is no doubt already everything which it is, but it is not yet all which it can be ; it is not yet that particular thing it is destined to become, but it possesses the means to be it, it can be it. Therefore between being and not being there is the power of being. It is neither in respect of what it is, nor in respect of what it is not, but in respect of what it can be that the starting-point of change becomes its goal. The arrow is here {on the bow, for instance) and from the standpoint of being pure and simple, it is nowhere else ; but it can be there {at the goal, for example), and possesses the means to be there. Bread is bread and nothing but bread, and not flesh, so far as it is, in the sense of being pure and simple, that is to say, of being completely realised ; but it can cease to be bread and become flesh. There is in it that which enables it to undergo the change under the action of certain determinate causes. POTENCY OR POTENTIALITY Things therefore are not confined and held fast by what they are and what they are not. Even while 242
  • 249. ontology: act and potentiality they are here and not there, this and not that, they possess the power to be there and no longer here, that and no longer this. But so long as they are here or are this, that power which they possess remains mere power and is not manifested. This power in them is as such something real. Consider a man asleep. He neither sees nor speaks nor walks. But he is not therefore bUnd, paralysed, or dumb. He is really capable of seeing, speaking, and walking. While he does not speak he retains the power to speak, he has it in him ; whereas he cannot without a violation of nature become a tree or a bird. Or again take a billiard ball at rest. It is immobile (not moving). But it is not therefore immovable. It is really capable of motion. While it does not move, it retains the power of being moved, it has it in itself; whereas it has no natural power of passing through a wall. The power of being is not being in the full and primary sense of the term ; but power of being without as yet being is not sheer non-entity. Power of being taken precisely as such is irre-ducible either to nonentity or to being pure and simple. It is something different from either, something sui generis for which philosophy must find a place. Pre-cisely so far as things can be something they are not, they, after an inferior fashion, are. We have thus found something which does not deserve to be called being, on which that title can be bestowed only in a secondary and improper sense, as an alms, so to speak, but which nevertheless is real. It is what philosophers term potency or potentiality. In using the term potency we must be on our guard against ambiguity. This potency is not that of which 243
  • 250. AN INTRODUCTION TO PHILOSOPHY we think when we say that a being is potent. This potency is not an active power ; power to effect something or to work, at least when understood as active, is the absolute contrary of the power or potency with which we are now concerned, being not potency but act. The potency of which we are speaking is entirely passive, nothing more than a real power of being or becoming. Wax is in potency to receive the impress of the seal, water in potency to become ice or vapour. The active powers (for instance, the faculties of the soul) are also rightly termed potencies or potenti-alities^ but only in so far as they are not, or are capable of not being, actually operative, or so far as they are simple capacities of action or operation. ACT Since, however, power of being, though not nothing, is not being in the full and primary sense of the term, we must find a name for being in the full and primary sense of the term as distinguished from potentiality. Philosophers call it act. Here also we must beware of ambiguity. We are not concerned, at least not primarily and chiefly, with an act in the ordinary sense of the word, with doing or action. Action or operation is indeed an act, being in act, but it is what is termed the secondary act [actus operationis) . Action presupposes being. And the primary act is the act of being [actiis exsistentiae) , more-over of being a particular thing {actus essentiae) . For example, a body is luminous in act, even when it is not illuminating anything else. Clay, once modelled, is a statue in act, water at 32° Fahrenheit is ice in act, 244
  • 251. ontology: act and potentiality and the moment anything effectively is one thing or another and especially the moment anything exists, it is in act. Act may therefore be defined as being in the strict sense of the term, taken in the fullness thus signified, or again the finished, the determinate, or the perfect as such. Potentiality, on the other hand, is the determin-able, the perfectible, that which is capable of being finished, as such ; not a being but a real power of being. We must take care not to attempt to think with our imagination these concepts of act and potentiaUty. They can be thought by the intellect alone. Least of all must we conceive of potentiaUty as some sort of being in act which we imagine as more or less hazy, indefinite, inactive, and hidden in the object. Poten-tiaUty in itself is absolutely incapable of being repre-sented. It is not a spring or an organ hidden in the thing, nor a character prefigured in it after the fashion of an imaginary statue outUned beforehand by the veins of marble within the block, nor yet an act thwarted or rendered abortive, like an effort or pressure over-come by the resistance of an obstacle. It is absolutely nothing done or in process of doing, absolutely nothing in act. In itself it cannot be conceived (for in that case it would necessarily be conceived as something determinate). It can be conceived only by means of the act (the particular thing) with which it is corre-lated, as the simple power of being that particular thing. Conclusion XVII.—Being, considered in relation to the fullness and perfection which the term 245
  • 252. AN INTRODUCTION TO PHILOSOPHY signifies, is divided into being in the strict sense or act, and power of being or potentiality. We are now in a position to understand change. The product of change arises neither out of being in act nor out of nothing, but from potential being. In other words, the action of the efficient cause draws, educes, from the potentiality of the subject the deter-mination, the form, which was wanting in the starting-point of the change and characterises its goal, as when the action of fire educes from the potentiality of water (the water is cold, but can be hot) the determina-tion (a specific intensity of heat) which characterises it as the result of the change. The change is the transition from potentiality to act, or, more accur-ately, according to a definition to which we must return later, it is the act of a thing in potentiality taken precisely in respect of its potentiality : actus exsistentis in potentia prout in potentia. ACT AND POTENTIALITY IN THINGS From what has been said it follows that all change-able things, in whatever respect they are subject to change, are compounded of potentiality and act. God alone, since he is absolutely unchangeable, is devoid of any potentiality. Since he is subsistent Being itself or the Fullness of being, he is incapable of becoming ; there is no perfection which he does not possess or rather is not already ; he is pure act. The being of all other things, on the contrary, is too poor and too weak to reaUse simultaneously everything they are capable of being. For every one of them there is really open a vast range of possibilities, 246
  • 253. ontology: act and potentiality of which they can never rezilise more than a few, and that by changing. Here we may observe that the obscure and mysterious concept of first matter whose acquaintance we made when we studied the notion of individual nature is that of a pure potentiality in the order of substance, which can be any and every body and by itself is none. It is the purely potential principle which in union with an actual principle (a substantial form) constitutes a particular corporeal substance, and is the subject of substantial changes. Potentiality and act divide between them the totality of created being, both in the order of substance and in the order of accidents. In other words, they are, like being itself, transcendental objects of thought which exceed or transcend every limitation of class or category, and include all created things. The substance of bodies is compounded of potentiality {first matter) and act [substantial form) . The substance of incorporeal things [pure spirits) is not composite ; it is in respect of that which constitutes its nature or essence wholly act. But it is not therefore pure act (in the case of pure created spirits), for this substance {substantial essence) is itself potentiality in relation to that which is the ultimate act of everything real {actualitas omnis formae) or existence : pure spirits do not derive their existence from themselves, a se ; they can not be. On the other hand, we may remark that every accident {whiteness, strength, virtue, etc.) is an act {accidental form) which determines the subject and is itself sometimes in potentiality in respect of further determinations. The intellect, for example, is an 247
  • 254. AN INTRODUCTION TO PHILOSOPHY accident (an accidental form) whose subject is the soul, and it is in potentiality in respect of a particular act of thought. It is clear that all the notions with which we have become acquainted hitherto can be classified in the following fashion : ' uncreated being : God, ens a se primarily as in- .pure ac creat ed being that which is essence < in the wide sense what telligible {ess the ence in the strict sub- ^ sense) and in vir- I tnai ject t^e of which it exists V primarily as com-pletely circum-scribed {individ-ual nature) and in virtue of which it exists as such that which primarily is : the primary subject of action {supposit-um, person) j act of being : existence in a particu-lar respect : absolutely : substantia secunda substantia prima accident {ac- cidentalform) substance composed (in the corporeal order) of po-ten tiali ty and act, and itself actual-ised by the ultimate act of every-thing real beii^ com' pound oi pote tialit and c Aft-er what has been said, it is sufficient to consider the notions of potentiality and act to see immediately the truth of the following Axioms : (i) Potentiality cannot exist in the pure state, apart from any act. This is evident. For, since existence is an act, potentiahty can only exist in beings which are in some other respect in act.^ 1 Hence first matter cannot exist separately unactualised by some particular substantial form. Similarly essence is in relation to the act of existence a potentiality really distinct from existence but actual in virtue of existence. 248
  • 255. ontology: act and potentiality (ii) Nothing is educed from potentiality to act except by some being in act. It is plainly impossible that that which is in potentiality, that which is capable of having a determination or a perfection but does not have it, should give to itself what it lacks, so far as it does not possess it, that is to say, so far as it is in potentiality. (iii) Act is prior to potentiality. A consequence of the preceding axiom. ^ (iv) Potentiality is essentially relative to act and is for the sake of the act [potentia dicitur ad actum). It is indeed only in relation to the act that the potentiality can be conceived (only in relation to being white that we can conceive the power of being white) ; and it is also only for the determination or perfection that the determinable and the perfectible as such are. (v) Act and potentiality belong to the same order ; that is to say, both must be in the order of substance, or both in the order of accident. For it is evident that every act which at the same time completes and specifically determines a potentiaHty must belong to the same order as that potentiality. The activity of thought, for example, belongs to the order of accident like the faculty itself from which it proceeds and which is in potentiality to that activity. (vi) Everything acts according to its nature in act. 1 This axiom is the metaphysical explanation of the truth previously affirmed : being is prior to becoming or change. Absolutely speaking this is true. In the order of material causality on the other hand potentiality is prior to act, becoming prior to being, the seed prior to the tree. But the seed itself presupposes the tree which produced it and at the beginning of the entire process the actuality of the First Cause.
  • 256. AN INTRODUCTION TO PHILOSOPHY Since activity is an act {actus operationis) which is brought into being by the subject from which it proceeds it presupposes (as laid down by Axiom ii) that the latter is in act to the extent to which it produces that activity. The same truth differently enunciated is expressed by the dictum action or operation manifests being {operatic sequitur esse). (vii) The combination of two beings in act cannot produce something which is one of itself We call one of itself {unum per se), as opposed to one by accident^ a thing which constitutes a single being and not a conjunction of beings, in other words a thing which is one in virtue of the nature by which it exists. For example, a living organism is a unit of itself, whereas a machine or a house is an accidental unit.^ This distinction once understood, it is plain that two beings in act, and as such constituting two beings, can never by their combination constitute anything except a conjunction of beings, that is to say, an accidental unit.* Once more on this question of act and potentiality we find philosophers divided into three great schools. The school of Aristotle and St. Thomas teaches the distinction between potentiality and act, the priority 1 No doubt, if you destroy the unity of the machine or the house, you destroy the machine or the house, but you do not destroy the natures (or substances) of which it is composed {iron, steel, bricks, etc.). Destroy, on the other hand, the unity of an organism, you destroy its very nature {substance). 2 This axiom plays an important part in natural philosophy and particularly in psychology. For instance, the Cartesian conception which regards the soul and the body as independent of each other, two complete substances, is unable to explain the substantial unity of the human being, because two complete substances are two beings in act. 250
  • 257. . ontology: act and potentiality of act to potentiality, the reality of motion and becoming, but the priority of being to motion. It also shows that between God (the pure act) and all things besides (compounds of potentiality and act) there is an absolute and infinite difference. Exaggerated intellectualism (Parmenides, Spinoza, Hegel) refuses to admit the notion of potentiality, because by itself it is obscure. If, however, everything which is, is wholly act or pure act, either motion must be unreal (Parmenides) or contraries identical (Hegel) and creatures must possess the same nature as God (pantheism) Anti-intellectualism (Herachtus, Bergson) equally rejects the distinction between potentiality and act, but because the notion of being is in the opinion of these philosophers illusory. If, however, being is denied in favour of becoming or pure change, pure act can no longer exist ; and however the exponents of this school may struggle to escape the logic of their position, God must possess the same nature as things (pantheism). Moderate Intellectualism [School of Aristotle and St. Thomas) Potentiality and act in things. God or the pure act is absolutely distinct from created things. Exaggerated Intellectualism Anti-intellectualism No potentiality in things. Neither act nor being. Everything is absorbed Everything is absorbed in either in pure being, or in change or pure becoming, the contradiction which and God is identical or constitutes becoming, and continuous with things, created things are identified with God. Material and formal, virtual and formal (or actual), implicit and explicit, in accomplished act, in express act. 251
  • 258. AN INTRODUCTION TO PHILOSOPHY In connection with the notions of potentiality and act philosophers use certain expressions whose meaning must now be explained. We have just learned the meaning of the two correlatives in potentiality and in act. The marble before it has been carved is a statue in potentiality ; as soon as the sculptor has given it the form he intends, it is a statue in act. {a) Closely related to these expressions in potentiality and in act are the expressions so frequently met in philosophy material and formal. They have been borrowed from natural philosophy (cosmology) which proves that every corporeal substance is compounded of two principles, Jirst matter (pure potentiality in the order of substance) and substantial form or first act.^ The terms material and formal have passed from natural philosophy into all branches of philosophy, to designate, by analogy, on the one hand whatever, in itself indeterminate and potential, plays the part of a subject which receives a determination, on the other hand whatever possesses of itself a determining, actuali-sing, and specificatory function, or again whatever is taken as possessing a particular character, in a particular aspect. It is in this sense that, as we have already seen,^ we distinguish between the material and formal objects of a virtue, science, or faculty. Hence arises in particular the distinction between the material and formal statement. We speak materially when we do not take the things of which we speak precisely as possessing the characters denoted by the words we use ; we speak formally when in the 1 See above, pp. 166-168. ^ See above, p. 106. 252
  • 259. ontology: substance and accident . things of which we speak we consider not so much the subject which possesses these characters as the characters themselves, with the sharp contour and clear-cut line they describe in it. This distinction is extremely important. Formal statement should, in-deed, be the constant aim of philosophy ; and, on the other hand, many propositions are true formaliter loquendo which are false materialiter loquendo, and vice versa. For examples the following propositions are true understood formally, but false if understood materially : Everything which is, is good (so far as it is). The common good is always preferable to the good of the individual (provided the common good is under-stood formally, in which case the union of the soul with God, that is with the transcendent common good of all creatures, is to be preferred to everything else) Superiors ought always to be obeyed. (So far as they are superiors, and command nothing opposed to the orders of a superior of higher rank.) There are men who are natural slaves. (If slave is understood formally, as meaning destined for manual or servile work.) Every virtue is stable. (If we consider solely its quality of virtue.) Knowledge is infallible (so far as it is knowledge). Conversely, the following propositions are true only in a certain context, and if understood materially.^ * It would be of interest to show how philosophy, since it has given up the technical language of Scholasticism, has increasingly tended to use terms in a material rather than a formal sense. Hence a number of badly stated problems, and a host of misunderstandings both between
  • 260. AN INTRODUCTION TO PHILOSOPHY This picture is the Adoration of the Magi. This book is the philosophy of Pythagoras. Speech has been given to man to conceal his thoughts. Philosophy is proud. The British Constitution is good because it is illogical. {b) We must be careful to distinguish in potentiality from virtual, for they signify entirely different things.* A thing is said to be virtual, or to exist virtually, when it is contained in another thing of superior rank, not in its being or proper determination (its formality) but under another being or determination (another formahty), so that it is truly there according to the virtue or degree of perfection which belongs to it, but not formally or actually. In this case, the being in which it is found is not in potentiality in respect of it, but on the contrary in act after a higher fashion. Its superiority is, so to speak, an obstacle which prevents the thing which it virtually contains from being present with its proper and inferior determina-tion (formality). Thus the perfections of all corporeal objects exist modern philosophers themselves, and still more between the modern philosophers and the ancients, with their formal terminology. It may also be observed that certain philosophic terms understood in a material sense, have acquired a meaning totally different from their original significance. Take, for example, the term object. For the ancients the object meant what is placed be/ore the mind or presented to it, considered formally as such. Hence imaginary beings, the chimaera for example, were said to exist objectively or as objects present to the mind, but not really or as things existing outside the mind. The modems, on the contrary, understand by an object the thing itself or the subject which is presented to the mind, and to exist objectively is therefore the same as to exist really or outside the mind. 1 Observe, however, that the expression potentially, in potentia, is sometimes used improperly in the sense of virtually.
  • 261. ontology: act and potentiality . virtually in God, conclusions are virtually contained in the premisses, partial lives exist virtually in the life of an organism. {c) We must now call attention to the fact that the contrast between implicit and explicit is not to be confused with that between virtual and formal {actual) A thing contained implicitly in another may be there formally or actually, not virtually : but it is present in a confused fashion, wrapped up and hidden as a flower lies hidden and folded in the bud. For example, in the truth Peter is a man there is implicitly contained the further truth Peter is a rational animal. {d) Finally a thing given formally and explicitly may be in act in two different ways. Take, for instance, a man who is running as fast as he can to escape his enemies. If I say he is fleeing I mention what he is doing in express act {in actu signato) (what he is doing as stressed or expressed, so to speak, by his intention). If, however, I say that he is accelerating the rate of his breathing I mention what he is doing only in accomplished act {in actu exercito). Similarly, a man who reads Ronsard, Lamartine, or Victor Hugo, to count the number of times they use the words aimer or cherir^ reads the poets, to be sure, and reads them formaDy and expUcitly, but that is not what he does in intention. On the contrary, when we consider his object in reading the poets, we must say that he is preparing an essay of stylometric literary criticism. We may say that he reads the poets in effect or in accomplished act, but that expressly or in express act he is preparing the essay in question. Again, when we repeat lilia agri non laborant neque nent, thinking solely of the meaning of the sentence, 855
  • 262. AN INTRODUCTION TO PHILOSOPHY what we then know in express act is the lilies of the field, which are present to our mind as neither toiling nor spinning. But at the same time we know in accomplished act the nominative plural, lilia^ which in turn we shall know in express act, if we return to the sentence in question and submit it to grammatical analysis. That is to say, the phrase in express act {in actu signato) is used of things to which intellect or will are directed, when they are the object of a concept of the intellect or an intention of the will specially formed for them, and are thus presented to the mind or brought into being under the actual heading or on the actual ground expressed by their name. When on the contrary they are presented to the mind or brought into being on occasion of something else and without being intended in themselves, we say that they are present only in accomplished act {in actu exercito). 256
  • 263. VIII THEODICY (natural THEOLOGy) . Metaphysics studies being as such, but for that very reason is obHged to study the cause of being. That is why its highest branch, the coping-stone so to speak of the entire metaphysical edifice, is con-cerned wdth him who is subsistent Being itself. This branch of metaphysics is called natural theology (the science of God in so far as he is accessible to natural reason, or, from another point of view, so far as he is the cause of things and author of the natural order) Since Leibniz it has also been known by the less appropriate name theodicy. Leibniz in his Theodicy (1710) undertook to defend Divine Providence against the attacks of the sceptics (particularly Bayle). The term theodicy (etymologi-cally : justification of God) has been used since that time to denote the branch of philosophy whose object is God. But the name is objectionable on two grounds : first because the providence of God has no need to be justified by philosophers ; secondly because provi-dence and the problem of evil are neither the sole nor the most important questions of which natural theology has to treat. The primary questions discussed by natural theo-logy are obviously those which concern God's existence itself. For the existence of God is not in fact, as Male- 257
  • 264. AN INTRODUCTION TO PHILOSOPHY branche and the ontologists believed, evident to us immediately and prior to any discursive activity of the mind ; it is in virtue of that intellectual operation which is the activity most profoundly distinctive of man, namely ratiocination, that it becomes evident to us, and to attain that certainty reasoning must proceed not from the mere idea or notion of perfect being (the ontological argument of St. Anselm and Descartes) but from facts whose existence is established beyond dispute. St. Thomas, resuming the entire tradition of the past, shows by five different arguments how the conclusion God exists is imposed with absolute necessity on the human reason. There exist in the world motion or change ; beings and events newly brought into being ; things which are and are capable of not being ; things graded in degrees of perfec-tion, whose perfection, which consists in being, is more or less limited, obscured, mingled with imperfection ; irrational natures disposed towards an object or end, as is proved, not only by the complex system of the universe or the structure of Uving organisms, but even by the simple aptitude of every agent to produce its specific operation. To account for these various facts we are compelled—for under pain of absurdity we are obliged to stop at an ultimate explanation of existence —to admit a cause which moves without being moved, causes without being caused, cannot lack existence, contains in its purity the perfection of which things partake in greater or less degree, possesses an intellect which is the final ground of all natures and the first principle of all things. Such a cause we term God ; it is pure act, deriving its existence from itself {a se) . In other words, being itself is its nature 258
  • 265. THEODICY (natural THEOLOGY) or essence, it is subsistent Being itself, he who is. This conclusion, which for the philosopher involves the most sublime truths of metaphysics, is reached very simply by common sense, for it is in truth the most fundamental natural operation of the human under-standing, so that it can be denied only by denying reason itself and its first principles (the laws of identity or non-contradiction, sufficient reason, causality) ; and as the history of philosophy shows only too plainly, the mind has no other choice than between the alter-natives : " the true God or radical irrationality." ^ It is also the province of theodicy to show with what manner of knowledge we know God and in due course to study his nature and perfections, in particular the unity, simplicity, and immutability which are immedi-ately deducible from the perfection of underived existence {aseity), the fundamental character of pure act, and which prove most clearly that he is abso-lutely and essentially distinct from the world ; his relations with the world, his knowledge, his activity as Creator and Mover ; and finally the problems involved in the divine foreknowledge of contingent events, particularly man's free acts, and those arising out of the existence of evil in the universe. The school of Aristotle and St. Thomas teaches that God is known by the natural reason analogically, so that we perceive the divine perfections {being, unity, goodness, wisdom, love, etc.) in the mirror of creatures, without asserting any unity of nature, common measure, or proportion, or mixture or confusion of any kind between God and created things. This ^ Garrigou-Lagrange, Dieu. son existence, sa nature, Paris, 3rd ed., 1920.
  • 266. . AN INTRODUCTION TO PHILOSOPHY teaching is opposed to two contrary errors, the error of the agnostics, who maintain that the Divine Being is beyond the ken of our intellect and God unknowable by the reason (sceptics ; phenomenalists ; positivists like Comte and Spencer, the Kantian school as a whole), and the error of the pantheists, who confuse the Divine Being with the being of created things (Parmenides, Heraclitus, the Stoics, Spinoza, the German metaphysicians after Lessing and Kant ; the modernists and immanentists) Philosophy of Aristotle and St. Thomas God is known by analogy and is absolutely distinct from creatures. Pantheism Agnosticism God confused with crea- God unknowable, tures. 260
  • 267. IX . THE PHILOSOPHY OF ART ETHICS The aim of the practical sciences is to know, not for the sake of knowing, but to procure by some action the good of man (other than the pure act of knowing truth). But the good of man can be understood in two different senses ; either of this or that particular good or of the good which is in itself alone the good of man and which, as we say, determines the signifi-cance of human life. the philosophy of art Of the different practical sciences which are con-cerned with the good of man from the first standpoint (that of particular goods, and not of the absolute good of human hfe), none, as we have pointed out already,^ is a philosophy. For none of these proposes to regulate human action in reference to the supreme cause in the practical order, that is (for in the practical order the aim or end pursued fulfils the function of cause or principle) in reference to the last end (the absolute good of man) Nor are these practical sciences sciences in the strict sense, for they do not proceed by way of demon-stration, drawing conclusions from their premisses. 1 See p. 149. 261
  • 268. AN INTRODUCTION TO PHILOSOPHY They are rather arts than sciences, and belong im-mediately to the wide category of art, not to that of science. The essential character of art taken in its complete extension is to instruct us how to make something, so that it is constructed, formed, or arranged, as it ought to be, and thus to secure the perfection or goodness, not of the maker, but of the object itself which he makes. Art therefore belongs to the practical order in the sense that it instructs us how to make something, considering not the use we should make of our free will but the manner in which the work as such and in itself should be executed. We may thus say that art is concerned with what is to be ina.dc, factibile, 7Tot.Y]T6v. This formal character of making is fulfilled primarily in the material objects produced or fashioned by man {the factibile in the strict sense). But in a wider sense it is to be found also in works of a purely spiritual nature. In this case it goes beyond the sphere of practice as such, so far as practice is opposed to speculation and signifies an activity other than that of pure knowing. It is in this sense that there is making in the purely speculative order (a form of argument, a proposition, are works, but works of the speculative reason), and there are arts, logic, for example, which are speculative arts. In order, however, to establish a general theory of art and making we must have recourse to the highest and most universal concepts and principles of human knowledge. Such a theory therefore belongs to the domain of philosophy. The province of philosophy thus defined is indeed 262
  • 269. THE PHILOSOPHY OF ART. ETHICS practical, since it is concerned with making, and its object is to order from above the branches of practical instruction. Nevertheless, since it is in the strict sense a science, it cannot be essentially practical, but remains essentially speculative in virtue of its object and method of procedure ; moreover, it is extremely remote from actual practice. Indeed, not only has it no concern with the application of the rules of art to a particular work to be accomphshed, but further it formulates rules which are far too general to be capable of such immediate apphcation and to be correctly termed rules of art in the strict sense ; it is therefore practical only in an improper sense and very imperfectly. The individual arts alone (branches of study essentially practical) possess rules sufficiently detailed to be immediately applicable to a particular work, and their apphcation belongs solely to them. Further, with the exception of the fine arts, whose object, beauty, being itself universal and immaterial, enables philosophy to perform effectively though from a remote height her office of supreme arbiter, the arts since they are devoid of any um'versal character, except the fact of being arts, of which philosophers can take cognizance, escape her jurisdiction almost completely. If we would accurately describe this branch of philosophy, we should term it the philosophy of making, but we shall call it simply the philosophy of art. ^ Here ^ The term aesthetics, which has now become current, would be doubly incorrect here. Modern writers understand by the word the theory of beauty and art, as though the philosophy of art were the place in which to treat questions concerning beauty considered in itself (such questions belong to ontology) , and as though art were confined to the fine arts (a mistake which vitiates the entire theory of art). Moreover, 263
  • 270. AN INTRODUCTION TO PHILOSOPHY we must first inquire what is the nature of art, if it is indeed, as St. Thomas teaches, a virtue of the practical intellect, and how it is distinguished on the one hand from the speculative virtues {understanding offirst principles^ knowledge, wisdom) , on the other from the moral virtues, prudence in particular ; how art is to be subdivided and the different arts classified ; and finally what are the first principles and distinctive conditions—though solely of the highest and most general order—of those arts which have beauty for their object (the fine arts) and which thus occupy a superior rank among the arts. ETHICS The practical science which aims at procuring man's unqualified good, his absolute good, is that of morals or ethics. Since its distinctive object is not the the word aesthetics is derived etymologically from sensibility {aiaddvofxai. = feel), whereas art, and beauty also, are matters of the intellect, quite as much as of feeling. Scholastic textbooks do not usually devote a separate treatise to the philosophy of art, and either study its problems in psychology alone, or, the better to explain the concept of prudence, in ethics. It would be necessary to classify the philosophy of art, like ethics itself, under natural philosophy, if we kept to the single standpoint of the specification of the sciences by their formal object. But if we adopt the wider standpoint of the end to which the sciences are ordered, we must distinguish practical from speculative philosophy, and it is equally necessary to distinguish, in practical philosophy itself, the philosophy of making and the philosophy of doing {cf. the author's Art and Scholasticism). This treatment presents the double advantage of corresponding to a very marked trend of modem thought, which tends to devote a special treatise (aesthetics) to the theory of art, and of returning to one of Aristotle's fundamental classifications : Tracra bidvoia fj irpaKTLKTf) i) iroirjTLKr) 7J OeuprjTiKTj {Metaph., ii, I, 1025 b 25). Cf. Top., vi, 6, 145 a 15, and viii, i, 157 a 10 ; Metaph., vi, i ; Nic. Eth., vi, 2, 1 139 a 27. Hamelin {op. cit., pp. 81 sgq.) makes a convincing defence against Zeller of Aristotle's real opinion on this matter. 264
  • 271. . THE PHILOSOPHY OF ART. ETHICS perfection of the works produced and fashioned by man but the good and perfection of the agent himself, or the use he freely makes of his faculties, it is in the strict sense the science of action, the science of human acts (in technical language, of the agibile or Trpaxxov, that is to say, of the free use, so far as it is free, of our faculties) Ethics is as practical as any true science in the strict sense can be, for it teaches not only the most general rules ofremote application, but also the particular rules applicable to the particular action to be performed. But at the same time this science has in view, not some particular secondary end, but the last end (the absolute good) of man, the supreme cause in the practical order. It is therefore a philosophy. It is without qualification the practical philosophy. Note.—Though ethics is as practical as any science in the strict sense can be, we must not therefore suppose that it is essentially practical (no science vere et proprie dicta is essentially practical), or that it is sufficient to make men behave rightly. It supplies, it is true, rules immediately applicable to particular cases, but it has no power to make us constantly apply them as we should in particular cases,^ in spite of the difficulties 1 On the contrary, the essentially practical sciences, that is to say the arts, themselves apply their rules to particular cases. These sciences are, strictly speaking, practical, but are not sciences in the strict sense but only improperly. There are thus many degrees of practicality. The philosophy of art (whose end is practical, and whose object an operabile, but to be known) provides no rules immediately applicable to particular cases. It is only improperly and very imperfectly practical. Ethics (whose end is practical, and whose object is also an operabile, but also to be known) does not apply, but provides rules immediately applicable to particular cases. It is as practical as a science in the 265
  • 272. — AN INTRODUCTION TO PHILOSOPHY arising from our passions and the complexity of material circumstances. It remains, therefore, essentially speculative in its final object {knowledge of human acts) and in its procedure (the deduction of truths from their premisses, not incitement to action) and is thus practical only in an improper sense. ^ If man is to do the right in the order of action moral science must be supplemented by the virtue of prudence^ which, if" we make use of it, makes us always judge correctly of the act we should perform, and will always that which we have thus judged to be right. On the other hand, ethics only supplies rules of human conduct in the natural order and in relation to man's last end as it would be, if it were a natural happiness. But since, in fact, man's last end is a supernatural good [God possessed not by the imperfect knowledge of human reason as such, but by the strict sense can be practical, but it is not strictly speaking or perfectly practical. The arts medicine, for example, or engineering—whose object is something to be done (not merely an operabile, but envisaged opera-biliter), provide rules immediately applicable and actually apply them to particular cases, but only by enabling us to judge of what is to be done, not by making us wiU to do it (for the artist can make a mistake and still be an artist (because he wills to make it). They are in the strict sense practical—but do not attain the highest degree of practicability. Finally prudence (whose object is also something to be done) applies to particular cases the rules of moral science and reason, not only teaching us to judge of the act to be performed, but making us employ as we ought our free activity (for the prudent man, as such, always wills the right). It is, strictly speaking, practical, and attains the highest degree of practicability. 1 Cf. St. Thomas, Sup. Boet. de Trin., q. 5, a. i, arf 3 : Scientia moralis, quamvis sit propter operationem, tamen ilia operaiio non est actus scientiae, sed actus virtutis, ut patet V Ethic. Unde non potest diet ars, sed rnagis in illis operationibus se habet virtus loco artis et idea veteres dijfinierunt virtutem esse artem bene recteque vivendi, ut dicit August. X, de Civ. Dei. 266
  • 273. THE PHILOSOPHY OF ART, ETHICS beatific and deifying vision of the Divine Essence), and since his actions must be regulated in reference to this supernatural end and so as to enable him to attain it, ethics or philosophic morality is evidently inadequate to teach him everything he needs to know in order to act rightly. It must be completed and elevated by the teachings of revelation. The epithet practical, applied to ethics, does not merely mean having as its aim an activity other than that of simply knowing (in this sense practical, whether used of art or morals, is opposed to speculative) ; but more strictly concerned with action and behaviour (the TrpaxTov, the distinctive sphere of moral science and the moral virtues as contrasted with the TtoiTjTov, the distinctive sphere of art). The fundamental question which practical philosophy must answer before any other is in what consists (from the standpoint of the natural order) the last end or absolute good of man. It must then study the actions by which man approaches or departs from his last end, examining first their nature and inner machinery, then what constitutes their moral character, that is to say, renders them morally good or bad. Ethics must, therefore, study in themselves the supreme rule of such actions (questions which treat of the eternal law and the natural law) and their immediate rule (questions relating to the conscience); it must also study the intrinsic principles from which those acts proceed, that is to say, the moral virtues and vices. But since ethics is a practical science it must not be content with these universal considerations ; it must proceed to the more particular determination of 267
  • 274. AN INTRODUCTION TO PHILOSOPHY ^ human acts and their rules. It is therefore obliged to study in great detail the rules which regulate man's conduct first so far as they concern the good of the agent himself and secondly so far as they concern the good of others (consequently the virtue o^justice). The latter inquiry introduces a number of most important questions pertaining to what is called natural right, and treating in the first place of man's obUgations to God (a question of natural religion), secondly of his obligations to his fellow-men. Here are discussed the questions which concern men as individuals [the rights of the individual, for example, the right of property), and those which concern them as members of a natural whole whose common good individuals must serve—the family and the poUtical society (the rights of society). Aristotle divided the science of morality, of human conduct [ethics in the wide sense) into three parts : the science of man's actions as an individual, ethics (in the stricter sense) ; the science of his actions as a member of the domestic society, economics ; the science of his actions as a member of the city (the civil society), politics.^ On the fundamental question of ethics—the question of man's last end—we find for the last time the schools of philosophy divided roughly into three groups. The school of Aristotle and St. Thomas teaches that the entire moral life depends on man's tendency to his sovereign good or happiness and that the object 1 I.e. of religion as it would be apart from the supernatural order to which man has in fact been raised. 2 See on this point Nic Eth., vi, 9, 11 42 a 9 ; Eud. Eth., i, 8, 12 18 b 1 3, and the two first chapters of the Politics. Cf. Hamelin, op. cit., p. 85. 268
  • 275. . THE PHILOSOPHY OF ART. ETHICS in which this happiness consists is God—whom, more-over, we ought to love, not for our own sake, but for himself (precisely because he is our last end, that is to say, that which is willed and loved for itself, not for the sake of anything beyond) The schools which find the end and rule of human conduct in pleasure {hedonism, Aristippus, Epicurus), utility {utilitarianism, Bentham, John Stuart Mill), the state (Hegel and sociologists), humanity (Auguste Comte), progress (Herbert Spencer), sympathy (the Scottish school), pity (Schopenhauer) or the production of the superman (Nietzsche), assign as man's last end some-thing created, and thereby degrade him below himself. The schools which claim that virtue (the Stoics, Spinoza) or duty (Kant) is self-sufficient, either because virtue is itself happiness, or because the pursuit of happiness is immoral, assign as man's last end man himself, and thereby, while seeming to deify man, really, Hke the schools last mentioned, degrade him below himself, for the greatness of man consists in the fact that his sole end is the uncreated Good. Thomist Philosophy [Ethics of Happiness or the Sovereign Good) Man is ordered to a last end other than himself, and this last end is God. Moral Systems which degrade Moral Systems which deify Man Man Man is ordered to a last Man is not ordered to end other than himself, any last end other than and this end is something himself, his own virtue created [hedonism, Epicure- being his last end (iS'toia'^m), anism, utilitarianism, etc.). or his goodness not de-pending on any good for which he has been made [Kantism). 269
  • 276. AN INTRODUCTION TO PHILOSOPHY Thus on every one of the great problems of philosophy the doctrine of Aristotle and St. Thomas, when compared with the doctrines of other philosophers, appears as an eminence between two contrary errors. This is an additional argument for its truth to be added to those enumerated earUer.^ The truth, indeed, is not to be found in a philosophy which keeps the mean between contrary errors by its mediocrity and by falling below them, being built up by borrowing from both, balancing one against another and mingUng them by arbitrary choices made without the Hght of a guiding principle {eclecticism) ; it must be sought in a philosophy which keeps the mean between contrary errors by its superiority, dominating both, so that they appear as fragments fallen and severed from its unity. For it is clear that, if this philosophy be true, it must reveal in full what error sees only in part and distorted by a bias, and thus must judge and secure, by its own principles, and in the hght of its own truth, v/hatever truth error contains though it cannot distinguish. 1 See pp. 99-101. 270
  • 277. CONCLUSION ; Philosophy is thus divided into three principal sections : logic, speculative philosophy, practical philosophy or, if we take account of the subdivisions of these three sections, into seven main sections : minor logic, major logic ; the philosophy of mathematics, the philosophy of nature, metaphysics ; the philosophy of art and ethics. This order is represented by the following table : I J I. Minor or formal logic : the rules of reasoning LOGIC 1^2. Major OT material logic : the matter of resisoning 3. The philosophy of mathematics : quantity . —„,.,,. i the material world J '"^^^ &/ 4. 1 he phuosobhy of nature : { , , , '^ man psychology (truth (epistemolog^') criticism being in general ontology being a se (natural theology) theodicy II SPECULATIVE PHILOSOPHY III r6. The philosophy of art : making PRACTICAL PHILOSOPHY [7. Ethics OT moral philosophy : action, conduct The division of philosophy into speculative and practical depends, not on the specific character of the various philosophic sciences, but on the end which they pursue. If that end is knowledge alone, the philosophy is speculative ; if the good of man, it is practical.^ If the philosophic sciences are classified from the standpoint of their specific character, ^ ethics, which 1 See St. Thomas, Sup. Boet. de Trin., q. 5, ad 4. ' This specification depends essentially on the degree of abstraction or degree of immateriality of the object studied. 271
  • 278. AN INTRODUCTION TO PHILOSOPHY treats of the moral virtues ^ and whose formal object is human action, and the philosophy of art, which treats of the practical intellectual virtues and whose formal object is human making, are divisions of the science of man, which itself belongs to natural philosophy (though it enters also into metaphysics). From this point of view we can recognise as specifically distinct philosophic sciences only logic, metaphysics, and the philosophy of nature, also the philosophy of mathematics, if this is not regarded as a subdivision of metaphysics or of the philosophy of nature. 1 Sic pertinet ad philosophiam [naturalem) , et est pars illius, quia agit dt anima ut est actus corporis, et consequenter de moralihus ejus, (John of St. Thomas, Cursus. phil., i, p. 732 ; Log., ii* q. 27, a. i.) 272
  • 282. Date Due Due Returned Due Returne- M' ^ '^' ' '" ^ ^^n n - - [jm 1 3 2^^ ; i j 1 I i V ~
  • 283. U'y.'yERSirY OF FLORIDA 3 12b5 OMMflM^fli 2