Part I: The Existential Method

THE EXISTENTIAL METHOD

It is perhaps to the great thinkers of the seventeenth and
eighteenth centuries that we are most indebted for what depth
and clarity there is in our metaphysics today. It is therefore
highly significant that many of these philosophers felt the selec-
tion of a fruitful method to be among the most important tasks
confronting them. "It were far better/' says Descartes, 4 "never to
think of investigating truth at all than to do so without a method.
... As well might a man burning with an unintelligent desire to find
treasure continuously roam the streets seeking to find something
that a passer-by might have chanced to drop." "I do not deny/'
he continues, "that sometimes in these wanderings" those who
philosophize in this manner "are lucky enough to find something
true . . . But I do not allow that this argues greater industry on
their part, but only greater luck." The beginning of metaphysical
wisdom, for Descartes as well as for many of his contemporaries
and successors, comes with the choice of a correct method. To
succeed, they hold, one must proceed along the proper path; an
advance in some other direction, with some other method, is
really no advance at all. Indeed, as Bacon puts it, "the lame . . .
in the path outstrip the swift who wander from it, and it is clear
that the very skill and swiftness of him who runs not in the right
direction must increase his aberration/' 5

Bacon's own contribution to the selection of a proper method
is chiefly a word of caution. We must avoid all hasty generaliza-
tions; only after prolonged and intimate acquaintance with
particulars through sense-experience and experiment may we
permit ourselves gradually to consider universals of wider and

wider significance. 6 Among the English philosophers of the period,
Bacon is undoubtedly the better known. Thomas Hobbes of
Malmesbury is however a more acute thinker whose excellent
style fittingly indicates the clarity and profundity of his thought.
Hobbes too felt the need to rebuild our metaphysics upon the
basis of a new method. He emphasizes the importance of a pre-
cise terminology. Like many thinkers as far back as Leonardo da
Vinci and possibly further, he feels that metaphysicians may
learn much from a consideration of the method used so success-
fully in mathematics. Leonardo had written: 7 "There is no
certainty where one can neither apply any of the mathematical
sciences nor any of those which are based on the mathematical
sciences." And Hobbes, selecting one feature for emulation in
metaphysics, writes: "A man that seeketh precise truth hath need
to remember what every name he uses stands for, and to place it
accordingly . . . And therefore in geometry, which is the only
science that it hath pleased God hitherto to bestow on mankind,
men begin at settling the significations of their words." 8 Proposi-
tions explaining words that represent our fundamental concepts
are, Hobbes holds, of indubitable truth. With these as a basis, he
holds, we should in teaching philosophy demonstrate those things
"which immediately succeed to universal definitions"; 9 and so
on down to less general propositions, affirming nothing "which
hath not good coherence" 10 with the definitions previously set
forth.

Descartes' contributions to the methodology of metaphysics
are likewise traceable to a desire to emulate the successes of
^mathematics. "Archimedes, in order that he might draw the
^terrestrial globe out of its plane and transport it elsewhere, de-
manded only that one point should be fixed and immovable; in
the same way," writes Descartes, 11 "I shall have the right to con-
ceive high hopes if I am happy enough to discover one thing only
which is certain and indubitable." It is not sufficient, however, to
have a fundamental proposition which is free from all doubt.
We must at all times, Descartes insists, eschew vague thinking and
doubtful ideas. In following out the implications of our funda-
mental proposition, we must use scrupulous care to assure our-
selves that our ideas are at all stages "clear and distinct." To reach
our goal, we must make use of the deductive method so success-

fill in mathematics; and we must continually guard ourselves
against vague and indistinct ideas. Moreover, we must not dis-
cuss metaphysical problems in whatever sequence they happen to
come to our attention. On the contrary, we must pay careful
attention to the order in which various subjects are considered,
not attempting to resolve complex problems before we have the
answers to the simpler problems which logically precede them.
"Those long chains of reasoning/' says Descartes, 12 "simple and
easy as they are, of which geometricians make use in order to
arrive at the most difficult demonstrations, had caused me to
imagine that all those things which fall under the cognizance
of man might very likely be mutually related in the same fash-
ion; and that, provided only that we abstain from receiving any-
thing as true which is not so, and always retain the order which is
necessary in order to deduce the one conclusion from the other,
there can be nothing so remote that we can not reach to it, nor
so recondite that we can not discover it."

In the "Essay concerning Human Understanding," Locke,
like many of his predecessors, stresses the importance of a care-
fully examined terminology. "I must confess," he says, 18 "that
when I first began this discourse of the understanding, and a
good while after, I had not the least thought that any considera-
tion of words was at all necessary to it. But when, having passed
over the original and composition of our ideas, I began to ex-
amine the extent and certainty of our knowledge, I found it had
so near a connexion with words, that unless their force and man-
ner of signification were first well observed, there could be very
little said clearly and pertinently concerning knowledge." "I am
apt to imagine," he continues, "that, were the imperfections of
language . . . more thoroughly weighed, a great many of the con-
troversies that make such a noise in the world would of them-
selves cease; and the way to knowledge, and perhaps peace too,
lie a great deal opener than it does." "Some gross and confused
conceptions men indeed ordinarily have, to which they apply the
common words of their language; and such a loose use of their
words serves them well enough in their ordinary discourses or
affairs. But this is not sufficient for philosophical inquiries," Be-
sides stressing the importance of clarity in thought and language,
Locke calls our attention to the desirability of determining the

limits beyond which our minds can not engage in fruitful dis-
cussions. "If we can find out how far the understanding can ex-
tend its view, how far it has faculties to attain certainty, and in
what cases it can only judge and guess, we may learn to content
ourselves with what is attainable by us in this state." 14

^The need to determine the limits within which the human
understanding must operate is emphasized -mese strongly by
Immanuel Kant. Beyond the limits of possible experience, Kant
holds, no knowledge is possible. "I had to remove knowledge/ 9
he writes, 15 "in order to make room for belief," Yet in marking
such a frontier, Kant was al0 motivated by a desire to determine
a region within which there can be developed a metaphysics and
a science having absolute certainty) Within the limits of possible
experience we can develop a metaphysics that will not be proble-
matical but apodictic. We can develop such a metaphysics, Kant
holds, if we allow reason to "move forward with the principles of
her judgments according to fixed law" and allow her to "compel
nature to answer her questions." 16 These principles with which
the mind operates are not, to be sure, divorced from experience,
since they are discovered only through attending to the mind in
action. Yet, with them as a basis, we must make use of the deduc-
tive method that has already been so successfully employed in
mathematics and in physics.

These references to certain philosophers of the seventeenth and
eighteenth centuries constitute of course only a small portion of
the voluminous material on the subject of method. Incomplete as
they are, however, they recall to us certain suggestions that have
been made time and again, suggestions as to what is needed for
the development of a successful metaphysics. Time and again our
attention is called to the necessity of clear thinking and an un-
ambiguous terminology. One writer urges us to cling to clear
and distinct ideas, another insists on determinate ideas, and a
third advises determining the significations of our terms. In one
form or another we are told that a successful metaphysics can be
developed only if we know exactly what we are thinking about
and just what our terms represent. We also find ourselves urged
to confine our thinking to subjects with which the human in-
tellect is competent to cope. For it is felt that, unless we know
what kind of problem can be handled with a prospect of sue-

cessful solution, much effort will be wasted in unprofitable dis-
cussion. Finally, we meet repeatedly with the warning that we
must proceed slowly and cautiously. At each stage in the develop-
ment of our thought we must guard against the temptation to
jump to the consideration of problems for which we are not yet
sufficiently prepared.

Let us seek to adhere in this treatise to the methodological
prescriptions which we have just discussed. Let us endeavor, that
is to say, (1) to make our thinking and the terminology through
which we express ourselves clear and precise, (2) to take up the
philosophical problems with which we shall deal in an orderly
manner and (3) to limit our attention to those matters which are
within the limits of human knowledge.

First, then, how are we to make our thinking and the ter-
minology through which we express ourselves clear and pre-
cise? The two, it would appear, are so interrelated that clear
thinking is well-nigh impossible without a carefully chosen ter-
minology. It seems to be the fate of words that, like machines,
they are capable of doing only a certain amount of work be-
fore they are in need of repair and rehabilitation. In the course
of an extensive use, words acquire secondary significations and
collateral meanings. They come to refer to no definite and pre-
cise entity, but rather to a composite something composed of
various concepts not clearly distinguished from one another. If
then we are to restrict ourselves to words that have definite
significations, such words as have, in the course of an extensive
use, come to have vague and indefinite meanings must either be
banned or rehabilitated.

Consider, for example, the word "idea." If we use the word
"idea" without first asking ourselves what definite entity we are
using it to represent, we shall almost unavoidably be using this
word to represent now one and now another portion of a vaguely
demarcated field of more or less related entities. Such an un-
critical use of the word "idea" on the part of others will make it
well-nigh impossible for us to understand and to evaluate their
pronouncements. If an author who uses "idea" without explana-
tion puts before us an argument whose pretended conclusion is
that ideas are necessarily involved in our thinking, or that ideas
are the sole objects of our thought, we shall find ourselves un-

able to determine whether or not his argument is sound and
his conclusion true. For before a proposition may be accepted or
rejected, it must first be understood. And a proposition in which
the word "idea" has the vague meaning that this word commonly
has is so lacking in exact reference as to be almost unintelligible.

The situation which obtains with respect to the word "idea"
obtains also, we hold, with respect to the word "existence." The
word "existence" has been held to represent what is permanent
and independent of our thought; and it has also been held to
represent what is given in sense-perception and is inseparable
from our thought. In the course of an extensive use, the significa-
tions of the word "existence" have become so various, so ramified
and so vague that the word as it comes to us out of the vocab-
ulary of current usage seems to have hardly any meaning at
all. It follows then that we can not use this word as it is commonly
used without becoming involved in vagueness and obscurity. If
we are to make a determined effort to keep our metaphysics
free from vagueness and ambiguity, we must in our construc-
tive efforts avoid the use of the word "existence" unless we
explain it. How, moreover, are we to understand the writings
of others in which the word "existence" occurs? The realist who
is an epistemological monist tells us that ideas do not exist; the
atheist tells us that God does not exist; some behaviorists tell us
that consciousness does not exist. But if, when such assertions are
made, we are not able to understand the word "existence" as it is
used, we shall be unable to determine whether what is being con-
sidered with respect to ideas, God and consciousness is their
intelligibility, their perceptibility, their inclusion in a systematic
whole, or some vague combination of all of these characteristics.
We shall gather that something is being denied of ideas, God or
consciousness; but we shall be unable to determine precisely
what it is that is being denied of them.

When we meet with the sentence: "Ideas exist," we are fre-
quently unable to determine whether existence is being pre-
dicated of mental content or of universals. And, in view of the
various senses in which "existence" has been used, we are fre-
quently unable to determine whether what is being predicated
of ideas is membership in some organic whole or perceptibility
or freedom from dependence on any conscious subject. The

situation is similar when we meet with the sentence: "Conscious-
ness exists." On the one hand, we may be unable to determine
whether existence is being predicated of a certain sort of mental
activity or whether it is being predicated of the field of objects.
And, on the other hand, it may be one of several characteristics
that the author is attributing to the entity he calls "consciousness."
There is the sentence: "Ideas exist" (or do not exist) and the
sentence: "Consciousness exists" (or does not exist). But we also
meet with the sentences: "Evil exists" and "Electrons exist^ and
"Centaurs exist." Existence or non-existence may be predicated
of anything. If, then, the signification of "existence" is left vague
and indeterminate, we have on our hands, as it were, a general
and blanket ambiguity which overspreads the more limited
ambiguities arising from the indeterminate use of one or another
of such words as "evil" or "consciousness" or "idea." We have on
our hands this all-pervasive ambiguity, that is to say, unless either
we use "existence" more sparingly than it is used or implied in
ordinary speech, or unless we select for this word a determinate
meaning.

It may be said, however, that the use of "existence" is by no
means so widespread as we have suggested. It may be said that
common speech uses "existence" but sparingly and that we can
well forego any detailed consideration of the meaning of this
term. Is not the term "existence" after all a scholastic and aca-
demic one and the question whether an entity "exists" an
artificial one? In the ordinary business of life, it is said, we are
not confronted with the problem whether an alleged entity exists
but only with the practical problem: what entities are we con-
fronted by to which we must give consideration? 17 Yet when we
ask what entities are we confronted by that deserve consideration,
we are asking a question which might in common speech be ex-
pressed as: "What entities are real?" And to ask what entities
are real is to ask whether this or that apparent, alleged, subsistent
entity is really existent or merely illusory and specious. 18 Ques-
tions involving "existence" seem thus to be not merely artificial
and academic, but to be deeply imbedded in our practical life
and in our customary conversation. Indeed when a sentence used
in our ordinary discourse does not explicitly contain the term
"existence," it may frequently be replaced by a sentence synony-

mous with it in which some grammatical form of this term occurs,
a sentence synonymous with it in the sense that we would ordi-
narily take the two sentences to have the same meaning. 19 The
"Some men are bald" of common speech is synonymous in this sense
with: "Some bald men exist." The "Some men are not patriotic"
of common speech is synonymous in this sense with: "Some un-
patriotic men exist." And since these are typical particular cate-
gorical propositions, it would seem that all propositions of this
form occurring in common speech are synonymous with exist-
ential propositions. It would seem, that is to say, that no particular
categorical proposition of common speech is free from vague-
ness so long as "existence" has but an indeterminate meaning.
With respect to this class of propositions, at any rate, it would
seem that ordinary discourse is tainted by vagueness and points
up the need for a renewed consideration of the meaning of
"existence."

The existential import of universal categorical propositions
used in common speech is not so obvious. Yet if "All men are
mortal" is not synonymous with "Mortal men exist," it would
seem that, keeping upon the level of ordinary discourse, a con-
siderable part of what is expressed in "All men are mortal" may
likewise be expressed in the sentence: "Immortal men do not
exist." Similarly, the "No stone is alive" of common speech seems
to be synonymous with "Living stones do not exist/' 20 Thus those
categorical propositions of common speech that are universal
seem, like those that are particular, to be not wholly free from
vagueness so long as "existence" is ambiguous. To the extent to
which common speech is made up of categorical propositions, it
would seem that even when "existence" does not occur explicitly,
it may be said to occur implicitly, resulting in a vagueness and
inaccuracy that can only be remedied by a careful determination
of the meaning of this term.

It may be argued that common speech is not a reliable guide
for the metaphysician in search of terminological exactitude.
Though it may be agreed that common speech is thoroughly in-
fected with a reference to "existence," it may be maintained that
this fact points to the desirability, not of re-examining the mean-
ing of "existence," but rather of developing a terminology in which
the word "existence" has no place. In the development of such a

terminology, modern mathematics, it may be felt, points out the
way for us to follow. For, it may be held, the modern mathemati-
cian makes no legitimate and essential use of "existence." If per-
chance he speaks of the existence of certain roots, he is making
an unfortunate and inappropriate use of the word. Generally
speaking, he does not begin his task, it is held, by predicating
existence of a certain space or of certain numbers. On the con-
trary, he takes this space and these numbers as subsistents, as
postulated entities. And he proceeds to develop their implications
while remaining entirely within the realm of subsistents. The
mathematician, on this view, is not concerned whether, for ex-
ample, Euclidean space exists or not. It is his task merely to point
out that Euclidean space determines the sum of the interior angles
of a plane triangle to a certain particular total.

So, it may be felt, we can develop a metaphysics in which the
term "existence" has no place. The metaphysician too, it may be
held, can begin with entities which are merely presented as sub-
sistents. And he too can limit himself to developing the implications
obtaining among these subsistents. His results, that is to say, may
all take the form: "A implies B." Does A exist? Does B exist? Such
questions, he may say, do not concern him as a metaphysician.
Rather, he may hold, it is for practical experience and common
usage to determine which entities are to be called "existent"; and
it is for the theologian to determine which entities are worthy of
being called "real."

Let us consider however the results that may be arrived at in a
metaphysics of this type. We conclude, let us suppose, that the sub-
sistent A implies the subsistent B. We assert: "A implies B"; and we
do not assert that A implies the absence of B, do not assert: "A im-
plies non-B." Yet when we have before us the two propositions:
"A implies B" and "A implies non-B," on what basis can the
metaphysician reject the latter and assert the former? Must he not
hold that A and B are really linked together in a way in which
A and non-B are not? Must he not be tacitly assuming that some
such entity as is generally called "reality" is so constituted as to
require the connection between A and B and to reject that be-
tween A and non-B? For if we make no such tacit assumption, if,
on the contrary, we constantly remind ourselves that we are dealing
with all subsistents, we must realize that the A that implies non-B

10

is a subsistent as well as the A that implies B. Without some
limitation based upon some distinction between the real and the
unreal, Euclidean space will be a subsistent and the Euclidean
space which involves 180 as the sum of the interior angles of a
plane triangle will be a subsistent. However, the Euclidean space
which involves a total of 90 for such a sum will be a subsistent
also. If we are merely discussing subsistents, in short, we may be
justified in stating: "A implies B." But we would be equally
justified in stating: "A implies non-B." We have no greater
justification for making the one statement than for making the
other; for all positive subsistential statements are on the same
footing.

The metaphysician who would avoid "existence" holds at times
that he is dealing only with what, for his purposes, may be mere
subsistents. And he holds at times that he is dealing only with
what, so far as he is concerned, may be mere postulates. It may
not be inappropriate, consequently, to point out two senses in
which the term "postulate" is used. In one sense an entity is
postulated when its existence is neither asserted nor denied, when
we seem to have it before us as a mere subsistent to be discussed.
In another sense a proposition, one which we should hold to be
explicitly or implicitly existential, will be said to be a postulate.
Such a proposition is a postulate in the sense that it functions as
a premise although unproved, although, that is to say, there are
no other propositions from which it has been deduced. In the
former sense God is a postulated entity in so far as God is regarded
merely as a subsistent. In the latter sense the proposition: "God
exists" may be regarded as a postulate; for this proposition may
be held to be one which is not deduced from other propositions
which are its premises.

The classic geometry brings before us the so-called postu-
late of parallels: through a given point there is only one
line parallel to a given line. The assertion here can hardly
be that there is only one such line that subsists. For every
thing that appears to be presented to us as an object subsists. And
unless a second parallel through the given point did at least appear
to be presented to us, non-Euclidean geometry would be incon-
ceivable and there would be no occasion for the postulate. The
so-called postulate of parallels must therefore be the existential

11

proposition: Through a given point there exists but one line
having certain characteristics. This proposition, it is obvious, is
a postulate in the second of the two senses we have distinguished
and not in the first. Since we are talking about an allegedly exist-
ing line, we are not holding this line before us merely as a subsis-
tent. One may, to be sure, use an existential proposition as a
postulate without accepting it. But to make use of an existen-
tial proposition is to concern one's self with 'existence/ It is
likely then that the metaphysician who would avoid the term
"existence/* and who takes the works of geometers as his guide,
has misread his mathematics. Generally speaking, mathematicians
put before us existential propositions of which they make use in
spite of the fact that these propositions are unproved. But they do
not put before us entities whose existential status is left entirely
out of consideration. They do not put before us the mere subsis-
tents to which the metaphysicians whose views we are considering
assign so important a role.

There is a further comment to be made on the doctrine that
metaphysics should avoid "existence" and should deal largely
with the relations obtaining among subsistents. As we have
seen, a considerable part of what is ordinarily meant by: "All
men are mortal' 1 may be expressed in the sentence: "Immortal
men do not exist/' By analogy, it would seem that much of
what is commonly expressed in "A implies B" might instead
be expressed in the sentence: "The A does not imply B does
not exist/ 1 If then a writer, believing that he is avoiding "exist-
ence" and that he is merely discussing subsistents, writes: "The
subsistent A implies the subsistent B," it would seem that he is
implicitly saying that a certain sort of A the A, namely, that
does not imply B- does not exist. It would seem, that is to say,
that he is referring to 'existence' after all.

"To the extent to which common speech is made up of cate-
gorical propositions/' we have seen, 21 it would seem that even
when "existence" does not occur explicitly, it may be said to
occur implicitly, resulting in a vagueness and inaccuracy that can
only be remedied by a careful determination of the meaning of
this term. And to the extent to which mathematical logicians fall
back upon implication and hypothetical propositions, a redetcr-
mination of the meaning of "existence" is, it would seem, likewise

12

indicated. It is true, that, when we assert: "If A is B, C is D," we
do not assert that A is B. But we are not justified in disregarding
the fact that we are asserting a connection between A being B and
C being D, a connection that in some sense we are asserting to
exist.

It would seem then that the ambiguities of "existence" as com-
monly used can not be avoided merely by the use of some alter-
native term, merely by concerning ourselves, for example, with
"implication" instead. For he who would develop a metaphysics
concerned merely with implications, must, if possible, describe
"implication" so that no reference to existence is involved; and
he must find a basis for rejecting: "A implies non-B" while he
asserts: "A implies B."

It appears then to be no easy task to develop a metaphysics from
which the term "existence" is excluded. Let us therefore acquiesce
in the continued use of "existence." Let us indeed bring into the
open the reference to existence that is so often implicit in our
assertions. And in the development of a metaphysics in which
"existence" has a prominent place, let us agree to make the effort
involved in a reconsideration of the meaning of this term. Indeed,
by continuing to use "existence," we shall be using a term exten-
sively employed in common parlance. And we shall be employing
a term which common parlance seems to regard as peculiarly
appropriate in metaphysics. For what, after all, is commonly
regarded as the proper field for metaphysical speculations? Is it
not commonly felt that the task of the metaphysician is to deter-
mine in a general way the nature of existence, the nature of real-
ity? And if this be the case, if, roughly speaking, the metaphy-
sician has the task of determining the general characteristics of
.existence so far as they may be determined without experiment,
surely it is inappropriate for him to avoid all mention of the
term "existence."

Words that in the course of an extensive use have "come to
have vague and indefinite meanings must," we have said, 22
"either be banned or rehabilitated." It has been our decision not
to avoid all mention of "existence." And so it remains for us
to set about rehabilitating this term. To assign "existence" a
definite signification is however to assign it a meaning which
does not coincide with the vague something to which "exist-

13

ence" commonly refers. A determinate signification can not be
interchangeable with an indeterminate signification. Our task
then will not be to arrive at some statement: "This is what 'exist-
ence* usually means"; but rather to arrive at some statement:
"This is what 'existence* means for us/'

Is it however permissible to assign a meaning to "existence"
as we might assign a meaning to "piety" or to "school"? It may
be agreed that I may assign "piety" whatever meaning I please
so long as I am consistent in my use of that word. But existence,
it may be held, is what it is. The word "existence," it may be held,
can be used to represent nothing else.

An objection of this sort seems to stem from the belief that
directly or indirectly we are aware of various entities, but not of
existence which somehow attaches itself to some of our objects
without being an object itself. If, however, existence characterized
certain objects without itself being an object, then the distinc-
tion between existence and non-existence would be unintelligible
to us. This however seems not to be the case. We do seem to be
aware of certain entities which in some sense of the word we
take to be existent and of certain entities which we take to be
non-existent. Directly or indirectly, therefore, existence must be
presented to us as a characteristic of certain objects. This charac-
teristic, some modification of it, or, indeed, any entity among
those of which we seem to be aware may, it would seem, be
represented by the word: "existence." "Existence," it follows,
may be used to represent a vague characteristic or a definite
characteristic among the entities of which we are somehow aware.
"Existence" may be given a definite meaning. And if "existence"
is to occur in our vocabulary at all, to express ourselves under-
standably we must give it a rather definite meaning.

The motive impelling us to redetermine the significations of
various words is the desire to establish for these words precise
and unambiguous meanings. If then we were to vary the senses
in which we use these words or were to shift from one signification
to another, our purpose would be thwarted and our redetermina-
tion of the meanings of these words would be in vain. Let us
bear such considerations in mind in redetermining the significa-
tion of "existence." Although we can not accept the suggestion
that we leave all concern with existence out of our terminological

14

discussions, there is a sense in which we can not play fast and
loose with "existence." When once the meaning of "existence"
has been even partially determined, all future use of that term
must agree with the signification previously chosen. We can not
continue to attach "existence" at random to whatever entities we
please. On the contrary, we are required to adhere in all strictness
to the meaning already selected. But before "existence" has had its
signification redetermined, existence is by no means a concept
that is sacred and untouchable. At such a stage it is not only
possible but highly desirable that we give "existence" a determi-
nate meaning.

Our initial discussion of method led us to three resolves, the
first of which was to make our thinking and the terminology
through which we express ourselves clear and precise. 28 This
clarity and precision we shall attempt to attain by giving precise
and determinate meanings to all important terms, the term "ex-
istence" being first in importance. A second conclusion to which
we were led by our discussion of method is that we must consider
metaphysical problems in their proper order, lest we attempt to
discuss matters for which we are not yet sufficiently prepared.
What, however, are these matters that we are called upon to dis-
cuss? The various questions which require resolution are for the
most part existential questions. We are called upon to decide, for
example, whether consciousness exists, whether a soul exists that
is able to outlive the death of the body, whether unperceived
entities exist, whether infinite collections exist, whether mental
content exists mediating between the subject and the object. The
resolution of each of these questions, it would appear, will be
affected by the decision we make as to the meaning of "existence."
For, the specific entities which exist and which together constitute
the world of existent entities will vary with the signification given
the term "existence." Not only may the determination of the
meaning of this term put before us the distinguishing character-
istics of existence as we are to use "existence"; it will also deter-
mine in large part the particular entities which exist. Only after
the signification of "existence" has been determined are we in a
position to resolve such questions as whether or not consciousness
exists, whether or not unperceived entities exist, whether or not
infinite collections exist. It behooves us, then, first to deter-

15

mine the signification of "existence"; and only after the meaning
of "existence" has been determined, to concern ourselves with
particular existential problems in the solution of which our deci-
sions as to the meaning of "existence" may be applied.

The writers whose discussions of method we have examined
have in the main emphasized three points. 24 They have urged
clear thinking and an accurate terminology; they have urged an
orderly procedure; and they have urged recognition of the limits
beyond which there can be no fruitful thinking. Clarity of thought
and accuracy in expression we shall attempt to attain through a
close regard for the significations of our important terms. Indeed
we shall seek a greater precision than has usually been attained
through a very careful attention to the signification of the almost
ubiquitous term: "existence." The order of procedure indicated
for us to follow is, first, the determination of the meaning of
"existence" and, second, the consideration of those existential
problems which this determined signification can aid us in solv-
ing. What remains to be asked is how we can avoid the considera-
tion of questions which in view of our equipment and resources
must be unanswerable.

As has already been pointed out, "the various questions which
require resolution are for the most part existential ques-
tions." 25 And so it would seem that when once "existence" has
been given a definite meaning that can readily be applied,
most questions put before us will be questions that we are pre-
pared to attack. An entity whose existence is in question may not
be clearly and unambiguously described. Or we may not be sup-
plied with all of the data necessary to determine whether or not a
given entity exists in our sense of "existence." But there will be
no existential questions in the face of which we shall be unable
to proceed, no entities to which the distinction between the real
and the unreal will not apply. When, however, "existence" has
no definite and unambiguous signification, then, to be sure, an
existential problem may well be unanswerable. 26 To determine
whether God exists, using "existence" in its usual indefinite sense,
that indeed may be beyond our powers. But when once the signi-
fication of "existence" has been determined, it is not unexperi-
enced entities that we shall avoid and not Kantian things-in-
themselves. Rather it is questions involving an indefinite and

16

unexplained "existence" that we shall neglect in order to avoid
the wasted effort that the consideration of an unanswerable ques-
tion involves.

In accordance with the procedure which we have outlined, the
determination of the signification of "existence" is to be the
foundation stone in our metaphysical structure. What then, we
ask ourselves, is the precise and definite entity which we should
use the term "existence" to represent? What is the clear and un-
ambiguous meaning which we should assign this most impor-
tant of terms? As we have already had occasion to observe, cur-
rent usage is, with respect to it, most indefinite. 27 So much so that,
when we assert that an entity exists, we may seem to be doing no
more than calling that entity to our hearer's attention. A hundred
real dollars, it has been said, contain not a penny more than a
hundred imaginary dollars. The assertion that the hundred
dollars exist, it may seem, tells us nothing about the hundred dol-
lars, joins no meaningful predicate to the subject term with which
it is linked.

Nevertheless, the term "existence," as ordinarily used, seems
to have some meaning. The assertion, for example, that God does
not exist is commonly regarded as quite different from the asser-
tion that God does exist, sufficiently different, in fact, to warrant
the most extreme measures. And if there is a difference, if, rather,
there is a difference of which we seem to be aware, that difference
must be between the object apparently presented to us that seems
to exist and the object apparently presented to us that seems not
to exist. Seeming to have as an object a hundred real dollars is
not identical with seeming to have as an object a hundred imagi-
nary dollars. What in the former case seems to be added to the
hundred dollars that is our object is not an additional quantity
of pennies but some vague quality of being important. It is to
be our task to substitute for this vague referend something more
precise that our term "existence" is to mean.

We are at liberty, of course, to determine upon one definite
and unambiguous meaning for our term "existence." Or we may
determine upon two or more distinct meanings, each of them
being definite and free from ambiguity. In the latter case,
for example, we may give "existence" a certain meaning when
"existence" is predicated of mathematical entities. And we may

17

give it a different meaning when it is predicated of characters
occurring in a novel. We may determine the signification of
"existence" so that one definite sense of this term is in question
when the existence of the number two is being considered; and
so that a different sense of this term is in question when the
existence of Hamlet or of Ivanhoe is being considered. We like-
wise are free to give "existence" and "reality" either the same or
different meanings. Ordinary usage is equivocal in this respect,
the terms often being used interchangeably, but sometimes not.

Common usage being indecisive, let us make the choice that
will make our task simplest and our procedure the most direct.
Let us agree to treat "existence" and "reality" as synonymous
terms. In this way, we shall be concentrating our attention upon
but a single task. Moreover, we shall find our language less mo-
notonous in that we shall be able to refer to the entity that exists
now by one of these terms and now by the other. Similarly let us
determine for our term "existence" but a single unambiguous
meaning. Let us agree to use "existence" in but one sense, no
matter what the context and no matter what the entities are whose
existence is being considered. By so doing, we shall be able to
concentrate our attention upon the determination of a single
definite and precise meaning. And we shall be spared the necessity
of explaining in each context just which sense of "existence" is
in question.

To be sure, we may commonly say of a lunatic that his million
dollars exist in his head. We may commonly say that Zeus exists
in Greek mythology but not in the physical world. And it may
not be altogether at variance with common usage to say that the
number two exists in the world of abstractions but not in the
world of concrete entities. 28

Yet in our ordinary speech we also recognize an existence that
is absolute existence. If we ask the man in the street whether the
lunatic's million dollars exist, he will answer immediately that
they do not exist. He will not ask us to specify which realm of
existence we are discussing. It appears then that when we com-
monly ask whether an entity exists, we are for the most part asking
whether it exists in the universe of real objects; existence that is
merely existence in thought or in the world of abstractions does
not concern us. And it is to be noticed that when we insist upon

18

taking into account various realms of existence, upon utilizing
various significations of "existence," the task of rendering the
meaning of "existence" precise has not been accomplished, but
has instead been replaced by a host of new and equally ardu-
ous tasks. We have now to ask what "existence" means when
it is predicated of physical entities, what when it is predicated of
mathematical entities, what when predicated of mental entities,
and what when predicated of the entities of science. Let us conse-
quently concentrate our attention upon the task of determining
a single signification. For if we do otherwise, we disperse our
attention and are likely to content ourselves with specious dis-
tinctions which do not make for real clarity but merely cover up
the difficulty. 29

We shall then select a definite signification which is to be the
signification of "existence," no matter what the context, and
which is likewise to be the signification of "reality." The propo-
sition or group of propositions with which we shall conclude
this part of our task will, let us suppose, be of the form: "An
existent is an entity which is such and such." Our proposition
obviously will not be one that we arrive at as a result of formal
argument and strict proof. It will, on the contrary, be a postulate,
an unproved assertion to be used as a premise in later discussion.
It is however one thing to postulate the Euclidean character of
perceptual space or the uniformity of nature; and it is another
thing to start with the premise: "An existent, in the sense in
which we use the term 'existence/ is an entity which is such and
such." In the former case the reader may feel that he is in pos-
session of some reason or of some experience which warrants
his rejection of the postulate. But in the case of "the existent is
the such and such," since we are merely presenting the meaning
which the term "existence" is to have in our writings, the reader
can have no reason for refusing us this terminological liberty.

We shall thus begin the construction of our metaphysical sys-
tem by attempting to assign to "existence" a precise and unam-
biguous meaning. The propositions in which this meaning is set
forth will be a postulate, a postulate, so to speak, which the reader
can have no reason for not granting. And with this postulate as
a basis, we shall, it is to be hoped, find ourselves in possession of
a premise from which we can determine the existence or non-
19

existence in our sense of the term "existencesof God, of con-
sciousness, and of unperceived entities.

When we come to consider particular existential problems, it
is desirable, we have agreed, that we take them up in the proper
order. In dealing with certain of these problems, to be sure, order
may be a matter of indifference. It may be, for example, that the
existence of individual substances can be considered as readily
after the existence of universals as before. However, we must be
on the watch for existential problems so related that the solu-
tion of one may reasonably be expected to aid us in the so-
lution of the other. Moreover, in dealing with the particular
existential problems which are subsequent to the determina-
tion of the meaning of "existence," order is not the sole con-
sideration to which our discussion of method commits us. It
is desirable that we assign a definite and unambiguous signifi-
cation, not only to the term "existence," but also to the other
important terms of which we are to make use. "Consciousness,"
"idea," "infinity," if these terms are to be used, they too must
represent definite entities if our thinking is to be clear, and if,
consequently, our metaphysical speculations are to result in
sound conclusions. When then we come to consider the exist-
ence or non-existence of consciousness, it is not sufficient that
we come to the task with an already determined definite signifi-
cation for "existence." We must now distinguish the various con-
cepts which the term "consciousness" has been used to represent.
We must bring out one or more definite and unambiguous mean-
ings which have been, or may be, assigned to this term. Only
then shall we find ourselves in a position to determine whether
consciousness in this sense, or in these senses, may be said to exist.
Having determined upon a definite meaning for "existence/*
we must bring into play whatever inventiveness and circumspec-
tion we are capable of in order to bring before us the entities
whose existence it is the task of the metaphysician to consider. We
must clarify the concepts thus brought before us so that in all
cases our thinking is clear, so that in all cases our important terms
have definite and unambiguous meanings. Finally, we must
bring the definite entities with which our analyses furnish us into
relation with our propositions determining the signification of
"existence." We must make use of our fundamental proposition

20

or group of propositions in determining the existence or non
existence in our sense of the term "existence" of these definite
entities.

A metaphysics which is developed in the manner which we have
outlined we shall take the liberty of calling an existential meta-
physics. And the method which we have outlined and determined
upon is, we shall say, the existential method as applied to the
solution of metaphysical problems. 30 A metaphysics that is exist-
ential will be based upon the realization that the term "existence"
is of fundamental importance. It will be based upon the realiza-
tion that this term needs a precise and unambiguous signification;
and upon the conviction that common usage furnishes us with
no signification of this sort. The metaphysician who makes use
of the existential method will consequently begin his constructive
labors by assigning to "existence" a definite, though to some ex-
tent an arbitrary, meaning. His first important propositions will
be those which, taken together, render explicit the significa-
tion that this term has for him. And these propositions, taken
together, will constitute the unfounded but unquestionable prem-
ise, the pou sto, of his metaphysical system. It is this existential
method which we shall attempt to apply in the present treatise.

We shall consequently determine upon a precise signification
which is to be the meaning that "existence" is to have in our
writings. What we are calling the "existential" method does not
however require the choice of the particular signification which
we shall select for "existence." The existential method does not
require us to replace the indefinite and general predicate in the
group of propositions which we may for the present summarize
as: "the existent is the such and such" with one particular and
unambiguous group of words rather than with another. Yet, how-
ever the predicate of this primary proposition is filled in, ex-
panded, or revised, the metaphysician who makes use of the
method which we are calling "existential" will regard the propo-
sitions in which the signification of "existence" is determined
the foundation stone of his metaphysical structure. He will utilize
this primary proposition as a premise from which he may partially
determine the existence or non-existence of various entities. The
content of the world of existents will vary, we have seen, with
the meaning that is chosen for the term: "existence." 81 Two

21

metaphysicians starting from different meanings may arrive at
different conclusions with respect to the existence or non-
existence of some particular entity. Since however they may both
be following the method which we are calling "existential/' it
follows that existential metaphysics does not involve any par-
ticular set of conclusions with respect to the content of the
world of reality. Existential metaphysics, in short, derives its name
from the existential method; and the system which is to be built
up in the following pages is but one of the ways in which that
method may be applied, is but one of the forms that an existential
metaphysics may take.

Descartes begins his "Meditations" by calling into question
practically all of our usual beliefs. He feels that in order to
develop a metaphysical structure that is firmly established, it
is first necessary to clear the ground. He resolves to "reject as
absolutely false everything as to which" he can "imagine the least
ground of doubt." 82 And so he concedes to the admirers of Mon-
taigne the invalidity of almost every proposition that has been
accepted as true. This task accomplished, Descartes undertakes to
find an indubitable proposition which will serve as a foun-
dation stone for a truly valid metaphysical structure. "Archi-
medes, in order that he might draw the terrestrial globe out of
its plane and transport it elsewhere, demanded only that one
point should be fixed and immovable; in the same way," says
Descartes, "I shall have the right to conceive high hopes if I am
happy enough to discover one thing only which is certain and
indubitable." 83 The proposition: "I exist as a being who is now
thinking" is for Descartes an indubitable truth of this sort. It is
a proposition which is shown to be true by the fact that its
denial is a self-contradiction. Not only, however, is this proposi-
tion indubitably true and in this sense clear; it also has, accord-
ing to Descartes, the second characteristic which is essential in a
first principle. "First, ... the principles must be very clear, and
. . . second" they must be such "that from them we may deduce
all other things." 34 Paying close attention to order, Descartes
proceeds, consequently, to deduce some of the implications of his
fundamental proposition. And so he arrives at the existence of
God, and, subsequently, at certain propositions "pertaining to cor-
poreal nature in so far as it is the object of pure mathematics." 88

22

Obviously, this procedure which Descartes employs has some
resemblance to that which we have determined upon. Just as
the Cartesian method begins by endeavoring to clear the ground,
so does the method which we are calling "existential." Whereas
Descartes holds that almost all pre-Cartesian assertions lack valid-
ity and a firm foundation, in a corresponding fashion it has been
our thesis that almost all previous assertions explicitly or im-
plicitly make use of a term which is vague and ambiguous. It is
our contention that in view of their overt or implied use of
"existence," these assertions, if not false, are vague and unin-
telligible. And, like Descartes, we too hold that they lack foun-
dation. For they make use of a term for which no precise sig-
nification has as yet been established. In the matter of the
foundation stone upon which the metaphysical structure is to
be based, here too there is a resemblance between the Carte-
sian method and that which we are calling "existential." In the
one method the structure is erected upon the "Cogito ergo sum,"
in the other upon a proposition or group of propositions in which
the determinate signification to be assigned "existence" is laid
down. There is a profound difference however in the grounds on
which these propositions are found valid. The fundamental
proposition of an existential metaphysics is in the nature of a
postulate; its validity lies neither in self-evidence nor yet in proof,
but rather in the liberty we have to develop a terminology which
is in some sense our own. Yet when the fundamental proposition
is once granted, an existential metaphysics develops in a manner
similar to that in which Descartes intended his metaphysics to
develop.

Let us however consider the possibility of arriving at a funda-
mental proposition in the Cartesian manner. Suppose I refuse to
accept the existence of all those entities whose existence is usually
granted. I am now doubting the existence of trees, of stones, of
men and of God. From this it follows, according to Descartes,
that I exist as an entity who is doubting these things. Such a con-
clusion follows, however, only because of the implicit use of
"existence" in the proposition which is made to serve as a prem-
ise. Just as, using the language of common parlance, "some men
are bald" appears to be equivalent to "some bald men exist," se
so the proposition: "I am doubting various things" appears to be

23

equivalent to the proposition: "I, as a doubter of various things,
exist/' It is this latter proposition which must then be regarded
as the foundation stone in the Cartesian system. And yet, on what
basis, we may ask, can the validity of this proposition be asserted?
Must we not say that the only justification this proposition can
have lies in the fact that in it the term "existence" is assigned a
signification in accordance with which "existence" denotes, among
other things, me the doubter? Descartes' fundamental proposition,
it would seem, turns out to be a sentence partially describing in
a denotative fashion the signification which "existence" has in his
writings.

Perhaps, however, we have misinterpreted Descartes. Perhaps
no reference to existence is to be read into the description of his
doubtings. Perhaps instead of asserting the existence of his doubt-
ing, he is merely refraining from attributing existence to the
various entities which appear to be his objects. Trees and men
and God, let us assume, are now merely subsistent entities. And
his doubting which also comes before him as an entity to be
considered, this too, let us suppose, is to be regarded as a sub-
sistent whose existence is neither asserted nor denied. But then the
absence of doubting in his mind seems also to come before him
as a subsistent. Yet in this situation, if we may so interpret Des-
cartes, he finds himself perforce considering the former object,
namely, the presence of doubting in his mind. He finds himself
in short considering two contradictory entities, the presence of
doubting and the absence of doubting, both of which, however,
are to be regarded merely as appearances, as subsistents. But surely
from this situation involving merely two subsistents, no conclu-
sion can be drawn with respect to reality. It is a matter of com-
mon agreement that we can not find a term in our conclusion
which does not occur in any of our premises. If then we are to
conclude that one of these mutually contradictory subsistents is
real, we must be tacitly assuming as a premise some proposition
which contains the term "real." We must be tacitly making use
as a premise of some such proposition as this: "If an entity insists
on coming before us when its contradictory comes before us,
then the former is a subsistent which is real." Again we find our-
selves brought back to a fundamental proposition in which there
is an assertion of existence. And here too, it appears, the validity

24

of our fundamental proposition must lie in the fact that it gives
existence a certain character, that in it the term "existence" is
being assigned a meaning.

An existential metaphysics, like the Cartesian philosophy, makes
use of a fundamental proposition from which subsequent truths
are deduced. With respect, however, to the justification of this
fundamental proposition, we find ourselves in accord, not so
much with Descartes, as with his English contemporary Hobbes.
"Primary propositions," writes Hobbes, 87 "are nothing but defini-
tions or parts of definitions, and these only are the principles of
demonstration, being truths constituted arbitrarily by the inven-
tors of speech, and therefore not to be demonstrated/'
(jDescartes and Hobbes were in a sense innovators who set op-
timistically to work to rebuild philosophy upon a new and firmer
basis. With the erudition and circumspection of Leibniz comes
a more sympathetic appreciation of the past. Formal logic and
the syllogism\ Leibniz holds, deserve a respectful place in our
philosophizing. Merely by developing the implications of certain
premises in strict logical form, we can, Leibniz holds, uncover the
self-contradictory character of certain propositions and of certain
notions. Thus 'swiftest motion/ he maintains, must be unreal
since logical analysis shows it to be self-contradictory. And the
eternal truths of mathematics and logic are known to be true
once it is shown that their contradictories involve self-contradic-
tion. (According to Leibniz, tfegn, mere logical analysis reveals to
us the non-existence of certain entities and the truth or falsity
of many propositions A There remain, however, many propositions
whose truth or falsity can not be determined by logical analysis.
These are the propositions with respect to which logical analysis
can uncover no self-contradiction either in them or in their con-
tradictories. If then we are to determine, for example, whether
there is ever a vacuum or whether, on the contrary, each place
contains some body, we need, Leibniz holds, some other tool in
addition to logical analysis, some other principle in addition to
the principle of contradiction. "This simple principle (the prin-
ciple of contradiction) is sufficient to demonstrate every part of
arithmetic and geometry" . . . But, Leibniz holds, 88 "in order to
proceed from mathematics to natural philosophy, another prin-
ciple is requisite."

25;

It is from a consideration of God's nature that Leibniz dis-
covers the second principle needed to distinguish reality from
unreality in those situations in which two contradictories are each
free from self-contradiction. God in the act of creation could not
have brought self-contradictory entities into existence. But in so
far as he was confronted by alternative systems of entities, each free
from internal contradiction, His nature, Leibniz holds, must have
impelled Him to bring into being that system and those en-
tities compatible with it that would result in the maximum of
reality. If we are confronted by two contradictory entities each
free from self-contradiction, we know, says Leibniz, that that one
must have been brought into existence which accords with God's
plan to bring into being the greatest possible number of com-
patible entities. We also know, he holds, that it would be incon-
sistent with God's nature for the act of creation to be in any
particular the exercise of an arbitrary and irrational choice. And
so if one of two contradictory propositions, each of which is free
from self-contradiction, points back to an irrational choice in
creation, we know that proposition to be false and its contra-
dictory true. It is these deductions from our knowledge of God
which, according to Leibniz, permit us to distinguish the real
from the unreal in certain cases in which logical analysis fails to
reveal any self-contradiction. A vacuum is not self-contradictory;
but since it does not accord with the fullness of being which fol-
lows from God's nature, it is unreal. A situation in which two
identically constituted substances are located at different places
is not self-contradictory; but since such a situation points back to
an irrational act in placing one here and one there rather than
vice versa, this situation too is unreal.

This distinction made by Leibniz between the principle of
contradiction and the principle of sufficient reason bears no re-
semblance to anything in Descartes' procedure. Yet here too there
is a resemblance to the existential method. The meaning of "exist-
ence" as developed in an existential metaphysics, may be regarded
as having two components. First, there is the vague and indeter-
minate signification of common usage. And, second, there is the
definite but uncommon signification into which the former is
transmuted through the terminological labors of the existen-
tial metaphysician. The former, the rough diamond furnished

26

by common usage, may be regarded as supplying us with the
principle of contradiction. And the more definite form added
by the existential metaphysician may be regarded as supplying
us with what may be called a principle of sufficient reason. Vague
and conflicting as are the significations generally attached to
"existence," it is generally agreed that the world of existent en-
tities contains no contradictions within itself, that the term ' 'exist-
ent" is not to be used to point to self-contradictory entities. This
characteristic of existence, however, which may be regarded as
implicit in the vague current meaning of "existence," does not
by itself furnish us with a complete and definite signification.
Whereas a law of contradiction may enable us to call certain self-
contradictory entities "unreal," we must make use of some second
principle if we are to be able more closely to delimit the real.
The proposition in which a definite but perhaps uncommon sig-
nification is assigned "existence" is, it follows, that element in an
existential metaphysics which is analogous to Leibniz's law of
sufficient reason. For it is this further, more precise element in
the signification of "existence" that must be brought into play if
we are to determine whether or not the term "existent" is prop-
erly to be applied to given entities which, without it, do not ap-
pear self-contradictory.

Our discussion of the "Cogito ergo sum" of Descartes has shown
us that the "Cogito" taken as the foundation stone of a metaphysi-
cal structure is in fact merely a proposition in which a signification
is being assigned "existence." 39 In short, the Cartesian method
turns out to be but a halting, partial, and unintended use of the
method which we are calling "existential." In a similar fashion
it is not difficult to show that Leibniz's principle of sufficient
reason is but an unfounded determination of the meaning of
"existence." What proof, for example, can be offered for the
proposition that God has chosen the maximum of existence? Does
not the validity of this proposition really lie in the fact that we
are, in laying down this proposition, giving "existence" a signifi-
cation in accordance with which it denotes the members of that
system which contains the maximum of compatible entities?

It turns out then that the validity of the law of sufficient rea-
son lies neither in self-evidence nor in proof. Like the "Cogito
ergo sum," and indeed like any proposition determining the

27

meaning of "existence," its validity, we hold, lies merely in
the freedom we have to develop a terminology which is in some
sense our own. The justification which Leibniz had given for
the law of sufficient reason was clearly unsatisfactory. And so
some of his immediate successors in Germany set themselves to
the task of establishing this law on what seemed to them a firmer
basis. These eighteenth-century philosophers whose erudition and
subtlety have not always been sufficiently appreciated, have left
us with arguments purporting to show that a denial of the law of
sufficient reason involves us in self-contradictions. Yet when Kant
begins his labors, the gap between the two principles is still un-
bridged. On the one hand there is the law of contradiction,
marking self-contradictory entities as unreal. And on the other
hand, there is a second and independent principle which must
be invoked, if we are not to accept all non-self-contradictory enti-
ties as real.

In the "Critique of Pure Reason" the distinction between these
two principles is crystallized in the distinction between analytic
judgments and synthetic judgments. "All analytic judgments,"
according to Kant, 40 "depend whoUy on the law of contradiction."
Synthetic judgments, whether a posteriori or a priori, agree, he
holds, in this: "that they can not possibly spring solely from the
principle of analysis, the law of contradiction." 41 "They require
a quite different principle. From whatever they may be deduced,
the deduction must, it is true, always be in accordance with the
principle of contradiction. For this principle must never be vio-
lated. But at the same time everything can not be deduced from
it." To be sure, the body of knowledge we may acquire solely
through the use of the law of contradiction is for Kant more
meagre than it is for Leibniz. 42 For Leibniz all mathematical
propositions derive their truth solely from the principle of con-
tradiction, whereas for Kant "seven plus five equals twelve" is
a synthetic proposition. 43 Nevertheless, in the writings of both
philosophers there is a distinction between two groups of truths;
and it is recognized that we need some principle other than that
of contradiction to give validity to what Kant calls our synthetic
judgments.

One of the most important judgments which Kant holds to be
synthetic is the judgment that all of our experience forms a uni-

28

fied whole. "Without . . . a unity which rests on a rule a priori
and subjects all phenomena to itself, no permanent and general
and therefore necessary unity of consciousness would be formed in
the manifold of our perceptions. Such perceptions would then
belong to no experience at all, they would be without an object,
a blind play of representations, less even than a dream." ** Kant
however seems determined that our perceptions shall not lack
objective reference, that they shall not be a blind play of repre-
sentations. And in order that they may be said to constitute
"knowledge" and that the entities to which they refer may be
said to be "real," Kant lays down the synthetic judgment upon
which, he holds, this consequence depends. The validity of the
proposition that our experience forms a unified whole seems thus
to be based merely upon the fact that this proposition enables us
to call the objects of our perceptions "real." This proposition,
which, in Kant's terminology, is not analytic, seems thus to be
merely an implicit determination of the content of reality and
hence of the meaning of the term "real." We advance beyond
the knowledge furnished us by the law of contradiction only by
adding a proposition which is in the nature of an explanation
further determining the signification of "reality."

The situation is very similar when we consider the synthetic
proposition advanced by Kant that each event has a cause. "If
we supposed that nothing precedes an event upon which such
event must follow according to rule, all succession of perception
would then exist in apprehension only, that is, subjectively . . .
I could not say of the object that it followed, because the follow-
ing in my apprehension only, without being determined by rule
in reference to what precedes, would not justify us in admitting
an objective following." 45 Kant however seems determined that
reality shall include objective and necessary sequences. He seems
to call such sequences "real" and to accept the causal law for
the sole reason that it justifies us in giving these sequences such a
designation. The proposition that each event has a cause seems thus
to be valid merely in the sense that it determines the sequences
we experience to be properly called "real." In laying down the
causal law. Kant is in effect determining the meaning of "ex-
istence" in such a way that this term will be applied to these
sequences. The validity which Kant finds for the causal law, that

29

is to say, is only the validity which attaches to a proposition de-
termining the meaning of a term. And so we add to the knowl-
edge furnished us by the law of contradiction by making use of a
proposition which implicitly determines somewhat further the
meaning of "existence." 46

The proposition that each event has a cause is not what Kant
terms analytic. For, analyze as much as we like, "we shall never
arrive from one object and its existence at the existence of an-
other/' 47 "There remained," Kant writes, "the possibility of
experience as that knowledge in which all objects must in the
end be capable of being given to us if their representation is to
have any objective reality for us." There remained, he should have
said, the promulgation of propositions determining the signifi-
cation of "reality" in such a way that our possible experience
would perforce be designated "real." "It was," quoting again
from Kant, "because people were ignorant of this method and
imagined that they could prove dogmatically synthetical propo-
sitions which the empirical use of the understanding follows as
its principles that so many and always unsuccessful attempts have
been made to prove the proposition of the 'sufficient reason/ "

In the foregoing discussion of Kant, we have been considering
the reality of possible experience and the validity of the synthetic
propositions which Kant holds apply to possible experience. Pos-
sible experience, however, Kant holds, is not the realm in which
lie all of the entities to which our thought is directed. Beyond
the "Herculean columns which nature herself has erected" lies
"a boundless ocean which, after deceiving us again and again,
makes us in the end cease all our laborious and tedious endeavors
as perfectly hopeless." 48 This is the realm of "rationalizing or
sophistical propositions which can neither hope for confirmation
nor need fear refutation from experience." 49 This is the realm
of vain, dogmatic metaphysics, and yet, to some extent also, of
justifiable faith. It was the denial of metaphysics, the denial of
knowledge o things-in-themselves that particularly impressed
Kant's early critics. 50 And Kant was subsequently much concerned
to refute the imputation that he had reduced everything to
illusion.

Without following Kant in his specific replies, let us consider
how such a criticism might well have been answered. "I confess

30

most humbly," Kant might have repeated, 61 that it "is entirely
beyond my power ... to extend human knowledge beyond the
limits of all possible experience." "My denial of a transcendent
metaphysics," he might have continued, "is based on the obvious
absurdity in attempting to go beyond experience with concepts
bound up with experience, and, more especially, on the various
absurdities into which, as I have shown in my Antinomies, an
attempt at transcendent metaphysics leads us. I also call your
attention," he might have continued, "to other sections of my
Dialectic in which I point out the invalidity of the principal
arguments of rational theology and of the major propositions with
which rational psychology is held to furnish us. If now you are
not going to content yourself with the remark that my negative
conclusions are displeasing to you, you must point out specific
errors in these passages of mine."

"Moreover," Kant might have reminded his critics, "I have
not contented myself with denying transcendent metaphysics.
Having shown that there is 'no rational psychology as a doctrine
furnishing any addition to our self-knowledge/ let me remark
that 'this refusal of our reason to give a satisfactory answer to
such curious questions which reach beyond the limits of this life'
should be taken 'as a hint to turn our self-knowledge away from
fruitless speculations to a fruitful practical use a use which'
... is 'directed always to objects of experience only/" 52 And,
he might have continued, "Before we venture beyond possible
experience, let us ask ourselves first whether we might not be
content with what possible experience contains." 53 "I suggest
therefore," he might have replied, "that you turn your attention
away from a transcendent metaphysics which I have shown to be
impossible to an immanent metaphysics, accepting my new point
of view that 'only in experience is there truth/ 54 I offer this sug-
gestion without misgivings," he might have said, "for what things
may be by themselves we know not, nor need we care to know,
because after all a thing can never come before me otherwise than
as a phenomenon." 55 "You may say," he might have added, "that
you are not interested in experience-for-us, that you are con-
cerned only about things in themselves. If, however, the argu-
ments of my Antinomies are sound, you must be convinced that
this hankering after transcendent metaphysics is but baying at

31

the moon. And I am hopeful that a careful study of my Analytic
will persuade you that the theses and problems of immanent meta-
physics which I there discuss will worthily replace in your atten-
tion the transcendent metaphysics which you must in any case
forego/'

Our doctrine that the correct method for metaphysics is to
develop the implications of propositions determining the signifi-
cation that the term "existence" has for us seems naturally to
evoke a criticism analogous to that which met Kant's * 'Critique
of Pure Reason/' "What we are interested in," our critic will tell
us, "is the nature of reality as it objectively is in itself, not the
nature of what you happen to choose to call 'real/ What we want
to know is whether or not God, consciousness and ideas are ob-
jectively real. It will not satisfy us to be told that you have de-
fined reality in such a way that in your terminology the word
'real* is properly to be linked with one or two of these entities
but not with the third. For all we care, you may tell us that
mermaids are real in the sense in which you choose to use the
word 'real/ and that, as you use this word, the King of England
is unreal/' "Our interest/' we shall be told, "lies in a realm
beyond mere terminology. Our concern is not with the word
'real' but with the world of reality itself which is independent
of any choice of words."

Just as this criticism is in some way analogous to that which
met the Critique of Pure Reason, so it points to a reply analogous
to the reply which, we have suggested, Kant might have made.
Just as Kant might have referred his critic to passages in which
he had in his opinion disproved the possibility of transcendent
metaphysics, so we may recall what has been said on the unintel-
ligibility of any discussion of reality which is divorced from a
consideration of the signification of the term "real/' 56 If what we
have said is sound, then must our critic realize what nonsense it
is to ask for a reality which is independent of any choice of words.
Moreover, we follow Kant further in not contenting ourselves
with negative conclusions. We invite our critic to engage with
us in a metaphysics which limits itself to the development of the
implications which may be drawn from propositions determining
the signification of our term "existence/' And we are hopeful
that a closer contact with such a metaphysics will show it to be

a richer and more enticing field than it may at first appear to be.
We are hopeful that, after our critic has been convinced of the
absurdity of baying at the moon, a closer acquaintance with a
metaphysics which applies the method which we call "existential"
will persuade him to shift his attention and his endeavors to this
more modest field. The inconclusiveness of a discussion of reality
which is divorced from a consideration of the signification of the
term "rear, this is a matter for argument and conviction. But
just as Kant could not by logic have forced his reader to become
interested in what is merely experience-for-us, so we can only
hope to evoke an interest in a metaphysics which is founded upon
an explanation of a term. Such a happy outcome, we are confident,
will result from a careful study of the theses and problems of an
existential metaphysics. And, to quote Descartes, 57 "it appears
to me that I can not do better than cause this to be established
by experience, that is to say, by inviting my readers to peruse this
book."

Summary

/In philosophy and indeed in most of our statements we are
-^implicitly, if not explicitly asserting or denying the existence
of some entity or other. The propositions through which we do
this can not be understood or evaluated unless the meaning of
our term "existence" is clear. Since "existence" has been used in
various senses, our meaning will not be clear unless we make it
so, unless we point out the specific sense in which we are using
this term.

The propositions in which we do point out how we are using
the term "existence" can not be overthrown by argument. Never-
theless, they are not trivial propositions. On the contrary, they
will serve as a major premise in a syllogism leading to the de-
termination of what exists and what does not exist in our sense
of "existence/)

Even this may seem trivial. But whether it seems so or no, it
is as far as any one can go. If the proposition "X exists" attempts
to make some assertion beyond "X exists in the sense in which I
am using the term 'existence/ " it is meaningless.

33

The program of this treatise will be to point out the meaning
our term "existence" has; to identify various entities whose
existence or non-existence customarily concerns philosophers (dis-
tinguishing these entities in certain cases from others with which
they may be confused) ; and then to determine whether or not
these entities exist in our sense of "existence."

TOWARDS DETERMINING THE MEANING
OF "EXISTENCE"

If a proposition is to be a definition, its subject-term and its
predicate-term must, let us agree, represent co-extensive entities.
If, for example, 'man* is to be defined as 'rational animal/ it
must be true that there is no man who is not a rational animal;
and it must be true that there is no rational animal who is not a
man.

Now our task is to determine the meaning of our term "exist-
ence," to define, if possible, the entity that our term "existence"
js to represent. What we seek is some proposition of the form:
''The existent is the such and such" or of the form: "To exist is
equivalent to being an A." And to accept as a definition a prop-
osition of the form: "To exist is to be an A," we must be willing
to accept both the proposition: "No entity exists which is not an
A" and the proposition: "There is no A which does not exist."

But what about: "There is no A which does not exist?" If
there is no A which does not exist, then all A's exist, and if "All
A's exist" is true, then there is at least one universal affirmative
^existential proposition which is true. Thus in order that our term
"existence" may be explained by means of a definition having
the form: "the existent is the such and such," there must be some
universal affirmative existential proposition which is true.

We have already had occasion to refer to certain existential

propositions which are extensively used or implied in ordinary

Discourse. 1 We have found that the categorical propositions of

common speech are to a considerable extent synonymous with

existential propositions similar in form to: "Some bald men

55

exist" or similar in form to: "Immortal men do not exist." Ot
the two existential propositions just stated, one, it is to be noted,
is a particular affirmative proposition and the other a universal
negative proposition. We have not found ordinary discourse
making use of, or implying, existential propositions which are
both universal and affirmative. We have not found ordinary dis-
course making use of that species of existential proposition of
which one instance must be true if our term "existence" is to be
explained by means of a definition having the form: "The exist-
ent is the such and such."

"All men exist" is a typical universal affirmative existential
proposition. But in what sense is it true that all men exist? All
real men, such as Socrates, Napoleon, you and I, do, let us agree,
exist. But if, in asserting that all men exist, we are asserting
merely that all existing men exist, our assertion conveys little
information. If the universal affirmative existential proposition:
"All A's exist" is synonymous with: "All existing A's exist,"
then the universal affirmative existential proposition is of little
use.

Let us see then what the situation is when our subject-term in-
tends to denote, not merely existing A's, but also A's which may
be alleged to exist. Let us suppose that, when we say "All men
exist," our subject-term intends to denote every individual, real
or fictitious, who may be alleged to be a man. The subject-term of
our existential proposition now seems to denote, not only So-
crates and Napoleon, but also Ivanhoe and the man whom I
imagine walking on my ceiling. But if our proposition is under-
stood in this sense, it is a proposition which, using "existence" in
any usual sense, is false.

We run into a similar difficulty whatever term we choose as
the subject of our universal affirmative existential proposition. If
we say that all spatial entities exist, intending to assert that all real
entities having spatial position exist, our proposition is not very
informative. And if, on the other hand, we are intending to assert
that all entities which may be alleged to have spatial position are
real, then we are apparently asserting the existence of the gods
on Mount Olympus and of the dragons who roam the woods*

When I assert that all A's exist, my predicament, to put it
briefly, is this. If I am discussing all conceivable, imaginable,

subsistent A's, my proposition, using "existence" in any usual
sense, is false. To be sure, since we may give "existence" any
meaning we please, "All subsistent A's exist" might be held to be
true. But if it is to be true that all subsistent A's exist, if it is to be
true that any A which I choose to imagine is an existent entity,
the world of existent entities must be regarded as a world that
can be populated at will. If, for example, all subsistent spatial
entities exist, I have merely to think of an entity as occurring
somewhere and, presto, it becomes real. Either then all uni-
versal affirmative existential propositions are either false or of
little value. Or, if we insist upon holding that there is some uni-
versal affirmative existential proposition which is both true and
useful as a definition, then we must be willing to use "existence"
in a sense from which it will follow that the world of existent
entities can be populated at wilL

Although "existence" as commonly used has a signification
which is extremely vague and inchoate, there are nevertheless two
or three propositions that may b laid down with respect to exist-
ence even before we refine upon the signification of this term.
"Existent," as commonly used, seems to be predicable only of
entities which are free from self-contradiction. 2 And "existence,"
as commonly used, seems to refer to a realm of entities which can
not be populated at will. Whereas we have agreed to redeter-
mine the signification of "existence," we also find it desirable to
retain whatever is definite and clear in the signification of this
term as it comes to us out of common speech. The rough diamond
with which ordinary discourse furnishes us is not to be cast aside;
it is to be treasured and cut and polished. If then "existence" as
commonly used seems to refer to a realm of entities which can
not be populated at will, let us agree to give our term "exist-
ence" a signification from which a similar consequence will
follow.

If we admit -universal affirmative existential propositions that
are both true and useful as definitions, the world of existent
entities will be one that can be populated at will. Since however
we have agreed to determine for our term "existence" a significa-
tion such that the world of existent entities will not be one that
can be populated at will, we must hold that there are no uni-
versal affirmative existential propositions that are both true and

37

useful as definitions. We must hold, that is to say, that, using
"existence" in the sense in which we are to use it, any proposi-
tion of the form: "All A's exist" is either false or of little value in
describing existence.

Our methodological discussions in the preceding chapter have
led us to determine to give to the term "existence" a signification
which is in some sense our own. We have supposed that we would
be able to assign a precise signification to "existence" by laying
down some proposition reading: "The existent is the such and
such." 3 We have supposed that we would be able to say that the
existent, in the sense in which we are to use the term "existence,"
has such and such a characteristic; and that the entity having this
characteristic exists in our sense of "existence." We have, in
short, anticipated being able to say that all entities that are such
and such, and that no entities that are not such and such, exist;
and we have supposed that such a statement would make clear
the signification we are assigning the term "existence." Since,
however, we have agreed that the world of existent entities, in
our sense of "existence," shall not be one that can be populated
at will, we can not lay down a truly universal proposition of the
form: "All subsistent entities having such and such a character-
istic exist." If we are to make use of a universal affirmative
existential proposition that is to be true at all, we must assert
merely that all existing entities having such and such a character-
istic exist. Yet, if our purpose is to make clear the signification
which we are assigning "existence," a proposition of this latter
form will be of little service.

It appears then that we can not very well explain our term
"existence" by stating that all entities having such and such a
characteristic exist in our sense of "existence." And so we are
left with but one-half of the statement which we had supposed
would explain our term "existence." We are left, that is to say,
with the proposition: "All existents have such and such a
characteristic," or with the proposition which follows from it, the
proposition: "No entity lacking such and such a characteristic
exists."

If we lay down the proposition: "No non-spatial entities exist,"
we give the reader considerable information as to the meaning
which we are assigning to our term "existence." We are inform-

38

ing him that "existence/ 1 in our sense of that term, is not a
characteristic of a non-spatial God, of ideas that are presented as
being in no place, or of universals regarded as not in their in-
stances. Thus propositions of the form: "No entities with such
and such a characteristic are real" are not to be disdained as
a means of conveying information as to the meaning which is
being assigned the term "real." If we say that no A's exist, the
reader is informed that each subsistent A is a non-existent entity.
Furthermore, the proposition which we thus put before the
reader has what may be called deductive power. There may sub-
sist X, Y and Z, entities whose existence is in question. But if X
and Z appear with the quality A, the non-existence of X and Z is
to be deduced directly from our initial proposition.

Whereas the proposition: "No subsisting such and such exists"
can, as we have just seen, be of much service to us, nevertheless
we can not be entirely satisfied with this proposition alone. If we
wish to explain the word "man," we can hardly content our-
selves with the proposition: "No finny creatures are men." The
reader is informed that to be a man is to be lacking fins; but he
does not have put before him other qualitites which belong to
man. The logical intension of 'man* is only partially revealed.
The logical extension of 'man* is less than that of 'non-finny
creature/ We come closer to our objective when we add the
proposition: "No invertebrates are men" or the proposition:
"No quadrupeds are men." Similarly, when "existence" is the
term to be explained. If we merely say that no subsisting A's
exist, we leave the intension of 'existence' too meagre and its
extension too large. But our failure is less marked when we add
the proposition: "No B's exist" and the proposition: "No C's
exist." In general, the more entities A, B, C ... we refer to in
this fashion in attempting to explain our word "existence," the
more fully we describe existence and the more numerous the
entities which are definitely marked out as non-existent.

With all this, however, we do not fully succeed in describing
the signification which we are assigning the term "existence."
Even when we say that no existent, as we use the term "existence,"
is either an A, a B, a C, or a D, our task has not been satisfac-
torily completed. For I may, it seems, imagine a man under my
chair; and I may imagine this man as being a sense-datum, in-

39

dependent of my thinking, causally related to other entities, and
so on. We can not rule out this man who is to be ruled out, since
we have agreed that the world of existents, in our sense of "exist-
ence," is not to be one that can be populated at will merely by
specifying some additional characteristic that an entity must lack
if it is to be an existent. No matter how comprehensive and
how varied the characteristics we make use of in our proposition:
"No existent is an A or a B or a C ..." we shall still fail to dis-
tinguish the subsistent non-A's, non-B's and non-C's which are
unreal, and which merely appear to be non-A's, non-B's, and non-
C's, from the subsistent non-A's, non-B's and non-C's which are
non-A's and non-B's and non-C's and which consequently are
real.

The proposition: "All existents are non-A's" or "No A's exist"
assigns certain entities to the realm of non-existence. But in order
that we may more fully describe the signification which we are
assigning the term "existence," we need some proposition of
another type. We can not complete our task by using only nega-
tive existential propositions. We have seen moreover that uni-
versal affirmative existential propositions can be of little service.
And so we are forced to make use of singular or particular ex-
istential propositions* We can not fully explain the signification
which we are assigning "existence" merely by laying down the
proposition: "No A's exist." We can not make use of the addi-
tional proposition: "All X's exist." And so we must supplement
our proposition: "No A's or B's or C's exist" with the proposi-
tion: "Some X's exist" or "Xj. and X 2 exist" and possibly with the
proposition: "Some Y's do not exist" or '% and Y 2 are non-exist-
ents."

It appears then that the task of explaining "existence" will not
be so simple as we had supposed. We shall be able to tell the
reader that the subsistents that are real are neither A's nor B's nor
C's nor D's. The more characteristics we make use of in this fash-
ion, the more fully will we be describing the signification which
we are assigning "existence." At the same time by making use of
more and more such characteristics, we increase the deductive
power of our explanation of "existence" with respect to sub-
sequent metaphysical discussions. For with each additional charac-
teristic, we may be assumed definitely to be assigning additional

40

entities to the realm of non-existence. To make our explanation
still more complete, however, we shall also have to make use of
propositions having the form: "Xi and X 2 exist*' and of proposi-
tions having the form: "Yi and Y 2 do not exist/' We shall have
to state that this particular entity and that particular entity are
to be called "existent" in our sense of "existence" and that this
particular entity and that particular entity are to be called "non-
existent" in our sense of "existence." In short, our explanation of
the term "existence" will have to fall into two parts. On the one
hand, we shall be making use of universal negative existential
propositions, marking out classes of entities that are unreal and
characteristics which definitely determine their possessors to be
non-existent. And on the other hand we shall be making use of
singular or particular existential propositions, pointing out defi-
nite entities to be included in the denotation of "existence" and
definite entities to be excluded from the denotation of "exist-
ence."

We shall thus attempt to explain our term "existence" through
the combined use of some such propositions as: "No non-spatial
entities exist," "The King of England exists" and "The immortal
Barbarossa does not exist." But it is necessary to point out some of
the results that propositions of these three types will, and some of
the results that they will not, accomplish. Let me suppose a sub-
sistent King of England alleged to be non-spatial. Since my sub-
sistent appears with the characteristic of non-spatiality, it will
follow, it may be said, that the King of England does not exist.
In determining non-spatial subsistents to be unreal, I rule out of
existence, it may be held, not merely unreal subsistents, but
along with them certain subsistents which are real. It may seem
that I have only in thought to give an existent the characteristic
of non-spatiality and, presto, it becomes unreal. Let us however
consider the singular negative proposition: "The immortal Bar-
barossa does not exist." From this proposition we can not con-
clude that there was no Barbarossa at all. We must, it would ap-
pear, distinguish between two different subsistents on the one
hand, Barbarossa with the qualities assigned him by the historian;
on the other hand, Barbarossa with the qualities assigned him by
legend. "The immortal Barbarossa does not exist" marks out
one clearly described and readily identified subsistent as unreal.

41

It is not to be understood as carrying over into the realm of the
non-existent other subsistent Barbarossas, among them the sub-
sistent Barbarossa discussed by the historian. Similarly with the
King of England. We must distinguish between the King of Eng-
land thought of as residing in Buckingham Palace and the King
of England thought of as non-spatial. "No non-spatial subsistents
are real" marks out the latter as unreal. But it leaves the King o
England residing in Buckingham Palace untouched. "No sub-
sistent A's are real" marks out as non-existent all entities appear-
ing with the quality A. But there may be some similar subsistent
appearing without the quality A which is real. In short our singu-
lar affirmative existential propositions and our singular nega-
tive existential propositions determine the existential status of
only those definitely described and readily identified subsistents
which are represented by the subject terms of our singular prop-
ositions.

The narrow limits within which our existential propositions
operate are also to be borne in mind when our propositions are
universal and negative. "No non-spatial subsistents are real"
disposes of subsistents appearing as non-spatial. But the world of
subsistents also, let us suppose, contains subsistents appearing as
extra-spatial and subsistents appearing as supra-spatial. It is as
fecund as the Hydra which Hercules had to encounter. Just as
Hercules struck off one head only to see two others appear, so we
assign one characteristic to the world of non-existence only to have
left confronting us other characteristics closely resembling what
we have just disposed of. When we dispose of non-spatial sub-
sistents, we dispose at the same time of extra-spatial subsist-
ents appearing as non-spatial. After elaborating a description of
extra-spatial subsistents, some of these subsistents no doubt appear
as non-spatial. But there is a residue which does not. Extra-spatial
subsistents, we may say, "resemble" or "are implied by" non-
spatial subsistents. But it is conceivable for them not to appear to
resemble, not to appear to be implied by, non-spatial subsistents.
We may eliminate whatever appears to resemble non-spatiality.
And by specifically eliminating, for example, supra-spatial sub-
sistents, we may dispose of some particular group of subsistents
whether they appear to resemble non-spatial entities or not. But
there is a residue of resembling or implied subsistents which no

42

negative existential proposition, either universal or singular, can
reach.

Not every subsistent is real, however, that a negative existential
proposition does not mark out as unreal. It is the entities repre-
sented by the subjects of our singular or particular affirmative
existential propositions that alone are definite members of the
world of existents. An extra-spatial subsistent that does not appear
as non-spatial is not unreal as a consequence of the proposition:
"No non-spatial subsistents are real." But it is not definitely
marked out as real unless it is enumerated among our Xi, X 2 , X s ,
. . . We can then determine not to enumerate among our exist-
ents any subsistent which appears as extra-spatial but not as non-
spatial. Having made use of the proposition: "No subsistent A's
are real/' we shall not list as real any subsistent which "ought" to
appear as resembling A or implied by A, but does not.

Whatever existential proposition we make use of in determi-
ning the signification of "existence," whether it be singular or uni-
versal, affirmative or negative, it determines the existential status
of those subsistents only which it definitely describes and identi-
fies. Unless we adopt this attitude with respect to negative exist-
ential propositions, the world of unreality has no obvious limits.
And unless we adopt this attitude with respect to singular affirma-
tive existential propositions, the world of reality can be populated
at will. If "The King of England exists" has the consequence
that the King of England thought of with whatever character-
istics we please exists, then the King of England who died at St.
Helena is real and the King of England who wrote the "Critique
of Pure Reason." We hold consequently that the subject of our
singular affirmative existential proposition is not the King of
England with whatever qualities he might be assigned. The
subject of our proposition is the King of England with his
qualities those which do in fact belong to him fully noted.
Or rather, since this is impossible, it is the King of England so
described as to leave no doubt as to which subsistent our term
"existence" is being used to denote. If then I am presented with
the King of England thought of with various characteristics, I
must distinguish between the various Kings of England presented
to me. The subsistent King of England who lives in Buckingham
Palace is represented by the subject of my affirmative existential

43

proposition. This King of England, consequently, exists. The non-
spatial King of England on the other hand, and the philosophical
King of England, are not represented by the subject of my affirma-
tive existential proposition. Consequently this proposition of
mine does not imply the existence of these merely imaginary
Kings of England.

There is a difference between the singular affirmative exist-
ential proposition and the universal affirmative existential prop-
osition. If we say: ''All subsisting non-A's exist," the world of
existent entities comes to be one that may be populated at will.
If we say: "X!, thought of with whatever characteristics we please,
exists," the world of existent entities is again one that may be pop-
ulated at will. But in order that the universal proposition may
be true, it must be emasculated to: "All existent non-A's exist/'
On the other hand, in order that the singular proposition may
be true, it need merely be reduced to: "Xi, described as such
and such a subsistent, appearing with this and that characteristic,
exists." The singular proposition, thus reduced, is not tautological.
We are not saying that the existing Xi exists. We are pointing
out an individual in such a manner that there is no doubt which
subsistent individual we are pointing to; and we are saying that
this subsistent individual is included in the denotation of "exist-
ence." The universal proposition, on the other hand, can not
fail to be tautological so long as it remains universal. If we are
to describe the existing non-A's without using the term "exist-
ence," our only recourse is to enumerate them, that is to say, to
replace our universal proposition with a collection of singular
propositions.

In order to describe the signification which we are assigning
"existence," it appears then that we are to lay down the universal
negative existential propositions: "No A's or B's or C's exist/'
and the singular or particular existential propositions: "X* and X 2
and X 8 exist" and '% and Y 2 and Y 8 do not exist." If we mention
various characteristics A, B, C, and point to a sufficient num-
ber of individuals Xi, X 2 , X 8 , . . . Y x , Y 2 , Y 8 , . . . the signification
of our term "existence" will, it is to be hoped, be clear. And we
shall, it is to be hoped, find ourselves in possession of a premise
that will be of service in the solution of particular existential
problems. In order to determine whether a given entity that

44

comes up for discussion is real or unreal, we shall have to apply
our propositions determining the signification of "existence/' We
shall first have to ask ourselves if the given entity subsists with
the characteristics A, B, or C. And if it appears that this entity is
presented to us as an A or as a B or as a C, then our task is com-
pleted. The given entity, presented to us in this manner, is un-
real. If, on the other hand, the given entity whose existence or
non-existence is to be determined does not subsist as an A or as a
B or as a C, then is our task not yet completed. We have still to
bring into play the singular or particular existential propositions
in which certain entities denoted by our term "existence" and
certain entities denoted by our term "non-existence" are pointed
out. If the entity under consideration is enumerated in the list
of entities which are specifically excluded from the denotation of
"existence," then, even though this entity lacks the characteristics
A, B and C, it is unreal. And, on the other hand, if, in addition to
lacking the characteristics A, B and C, it is listed among the
entities which are specifically included in the denotation of
"existence," then it is real.

We have agreed to lay down the universal negative existential
propositions: "No A's exist" and "No B's exist" and "No C's
exist." From these propositions, we have seen, it will follow that
any subsistent presented with characteristics A, B, or C is unreal.
There subsist, however, many subsistents lacking the character-
istics: A, B, C. Some of these entities will be enumerated in the
list which we are to draw up of entities specifically included in
the denotation of "existence." Others of them will be enumerated
in the list which we are to draw up of entities specifically ex-
cluded from the denotation of "existence." But no matter how
lengthy we make these two lists, many subsistents lacking charac-
teristics A, B and C will appear on neither list. Our propositions:
"Xi exists" and "X 2 exists" and "X 3 exists" and our propositions:
"Yi does not exist" and "Y 2 does not exist" and "Y 8 does not
exist" will by no means account for all of the subsistents appear-
ing without characteristics A, B and C. With respect to the enti-
ties thus unaccounted for, we can not determine from the sort
of explanation of "existence" that we have decided to give,
whether in our sense of "existence" they are existent or non-
existent. The sort of explanation of "existence" that we have

45

decided to give is thus not a complete definition.

Our interest in this treatise, it is to be remembered, is primarily
in the problems that are regarded as metaphysical. Were our
interest in some other field, our list of entities included in the
denotation of "existence" and our list of entities excluded from
the denotation of "existence" would both of them have to men-
tion entities that our lists will pass by. And were we attempting
in this treatise to deduce a complete system of knowledge and
not merely a system of metaphysics, our lists would have to be
much more encyclopedic, or, what is saying the same thing, our
singular affirmative existential propositions and our singular
negative existential propositions would have to be much more
numerous. Since, however, our interest in this treatise is primarily
in metaphysics, our lists will not have to mention the North Star
or the bee on yonder flower or the city of Bangkok. For we shall
not be called upon in this treatise to determine the existence or
non-existence of individual stars or bees or cities. We shall attempt
to draw up our lists so that our explanation of "existence" will
be available as a premise from which to deduce the existence or
non-existence of those entities whose ontological status is generally
regarded as a matter of concern to the metaphysician. If we suc-
ceed in doing this, then, for the limited subject-matter discussed
in this treatise, our explanation of "existence" will be the touch-
stone we require.

We have rejected the universal affirmative existential proposi-
tion: All men exist. We have agreed to make use of the singular
affirmative existential proposition: Xj exists or Socrates exists.
But what about the proposition: "The universal 'man' described
in such and such a manner, exists"? In asserting such a proposi-
tion, it is to be noted, we are not asserting that any entity that
is thought of as being a man exists. We are saying that the univer-
sal 'man/ considered as an idea in the mind of God, exists. Or
we are saying that the universal 'man/ considered as an entity
that is exemplified in certain individuals, such as Socrates and
Plato, exists. The proposition: "The universal 'man/ described
in such and such a manner, exists" does not, it seems, suffer from
the disabilities which affect the proposition: "All men exist."
For we are attributing 'existence' not to each real man nor to each
subsisting man, but to a certain subsistent that we describe and call

46

the universal: 'man/ We can not, we hold, make effective use o
the proposition: "All universals exist." But "The universal 'man/
described in such and such a manner, exists/' is a proposition
that may be both true and informative. The universal 'man* may
consequently be given a place on our list of entities denoted by
"existence" along with Socrates and Plato. So far as our present
discussion has carried us, our list may mention individual sub-
stances and individual qualities and individual relations. And it
may mention universal substances and universal qualities and
universal relations, whenever there is a suppositio individualis.

It is to be one of our tasks to draw up a list of entities, each of
which is denoted by our term "existence." And it is to be another
of our tasks to draw up a list of entities, each of which is excluded
from the denotation of our term "existence." For the drawing up
of these two lists we require no further discussion. A place is
reserved for these two lists at the end of the following chapter. 4
Taken together, they will, as we have said, partially describe the
signification we are assigning "existence."

When we partially determine the meaning of "existence" by
means of a singular existential proposition, we fix the existential
status of one particular entity. We do this, at least, provided the
subject-term of our singular existential proposition is so phrased
that there is no doubt as to which the entity is to which it refers.
When we partially determine the meaning of "existence" by
means of a universal negative existential proposition, we assign
to the realm of non-existents an entire class of entities. Here, too,
however, it is necessary that the subject-term of our proposition
be so phrased that there is not a complete uncertainty as to what
entities are apparently denoted by it. For if we say that all A's are
non-existents, and if the reader can not at all tell which entities
are presented as A's, then there are no entities that are definitely
being assigned to the realm of non-existents and our universal
negative existential proposition is not explaining, even partially,
our term: "existence." A universal negative existential proposition
asserts that no entities having such and such a characteristic exist.
It asserts that to exist is to be free from this or that characteristic.
Yet if this characteristic is vague and indefinite, if in learning that
existence is free from this characteristic we learn little about exist-
ence, then our universal negative existential proposition will

47

scarcely help one to understand our term "existence." It follows,
consequently, that our universal negative existential propositions
should be so chosen that they mark out fairly definite groups of
entities that are being assigned to the realm of non-existence.

Our task is to assign to the term "existence" a signification
more precise than that which this term ordinarily bears. The
"existence" of common speech is quite vague and ambiguous;
nevertheless, we have seen, it has, even as commonly used, some
meaning. To the extent to which the "existence" of common
speech has a precise signification, we have agreed that it will be
desirable to attach that signification to our term "existence." And
where the "existence" of common speech is vague, we want our
term "existence" to be more precise. If the "existence" of common
speech is precise in so far as it makes freedom from self-contradic-
tion a characteristic of existence, we want to explain our term
"existence" also so that all self-contradictory subsistents will fall
within the realm of the non-existent. We have agreed to explain
our term "existence" in part by means of universal negative exist-
ential propositions. Each such proposition, it is expected, will
assign to existence the property of being free from a certain char-
acteristic; and it will assign a group of subsistent entities to the
realm of the non-existent. We want to choose our universal nega-
tive existential propositions, consequently, in such a manner that
we do not assign to the realm of non-existence entities which
common speech definitely marks out as existent; and we do not
want to leave out of the realm of non-existence entities which
common speech definitely marks out as unreal.

We are at liberty to assign to our term "existence" any signifi-
cation we please. And so, as a partial explanation of the significa-
tion we are assigning "existence," we are as much at liberty to lay
down one universal negative existential proposition as we are to
lay down another. One universal negative existential proposition,
however, will assign to existence freedom from a richer, a more
definite, characteristic than another. One will assign to the realm
of the non-existent a more definite group of entities than another.
And one will assign to our term "existence" a signification more
in accord than another with the ordinary signification of "exist-
ence" in so far as that signification is precise. Whereas then any
universal negative existential proposition that is to be used in

48

assigning a signification to "existence*' is in the nature of a postu-
late without premises from which it can be deduced, one universal
negative existential proposition will enable us to carry out our
purpose more readily than another. Whereas there are no logical
grounds that force us to select one universal negative existential
proposition and to reject another, there are grounds of expediency
that permit us to prefer one universal negative existential propo-
sition to another. Thus we are left with certain criticisms that we
may bring, albeit no logical criticisms, against some of the univer-
sal negative existential propositions which may suggest themselves
to us as propositions to be used in partially describing the signi-
fication to be assigned to the term "existence."

For the remainder of this chapter then, let us call to mind
some of the universal negative existential propositions that might
be used in partially describing the meaning to be assigned
"existence." And, in view of the discussion of the preceding pages,
let us see which of these propositions it will, without more de-
tailed consideration later, be inexpedient to accept. In order to
obtain the material to which our considerations of expediency are
to be applied, let us review some of the philosophical writings
of the past. We must remember however that the philosophers
whom we are about to consider did not lay down universal nega-
tive existential propositions with the overt purpose of explaining
the term "existence." They may have mentioned "existence" only
casually; or they may have given assent to some universal affirma-
tive existential proposition. It is not our primary purpose at this
point to make an historical survey of the use of the term "exist-
ence" in the writings of various philosophers. Our task is to glance
through the history of philosophy in order to put before us
universal negative existential propositions from which to choose.

No question in Occidental philosophy, so far as we know, is
older than the question: What is it to be real? When the Milesians
found themselves confronted by a world of great variety and
ceaseless change, they asked themselves what the "nature" of
things is. "As Anaximandros and most of the physicists say,"
writes Aristotle, 5 the fundamental reality is something which "is
immortal and indestructible." And so we may elicit the doctrine
that only the permanent is real. This proposition, namely, that
whatever is impermanent is non-existent, is not to be extracted

49

merely from what has come down to us from the Milesians. From
Pannenides to Anaxagoras the real is that which persists un-
changed, unaffected by the lapse of time. There is disagreement
as to the number of such permanent entities and the qualities
that these entities possess, but among many Greek philosophers
there seems to be agreement that whatever is impermanent is
unreal. Indeed we find echoes of this doctrine as recent as Herbert
Spencer. "The most conspicuous contrast," writes Spencer, 6 "is
the contrast between that which perpetually changes and that
which does not change, between each ever-varying cluster of vivid
states and their unvarying nexus. This transcendent distinction
needs a name. I must use some mark to imply this duration as
distinguished from this transitoriness this permanence in the
midst of that which has no permanence. And the word 'existence/
as applied to the unknown nexus, has no other meaning. It ex-
presses nothing beyond this primordial fact in my experience."

Shall we partially describe the meaning which we are to
assign the term "existence" by means of the proposition: imper-
manent subsistents are unreal? If we take the term "permanence"
as it comes to us out of our everyday discourse, the typical sub-
sistents appearing as impermanent are such entities as flashes of
lightning. We choose, however, not to make use, unless there are
special considerations, of a universal negative existential proposi-
tion that will assign to the realm of non-existence entities which
common speech unhesitatingly marks out as existent. Surely,
there is no tendency in common speech to call mountains "real"
rather than sunsets, and Gothic cathedrals "real" rather than
soap bubbles. Common speech seems definitely to assign some
sunsets and some flashes of lightning to the realm of existent
entities. And so, unless permanence is used in some special sense,
the proposition: "Impermanent subsistents are unreal" would give
our term "reality" a meaning out of accord with common usage.

Fairly early in Greek thought the conviction developed that
the material things with which we commonly deal in our every-
day life are unimportant and unreal. Emphasis was shifted to
numbers, to forms, to universals, to ideals, and to scientific gen-
eralizations as the only realities. It is reason, the eyes of the mind,
that, it was said, puts us in touch with reality, not the senses which
are the eyes of the body. Among the Pythagoreans, then by Soc-

50

rates and by Plato, the world of intelligible entities was more
and more intensively explored, became richer and richer in con-
tent. And the conviction grew that whatever is merely mundane,
whatever is altogether a part of the spatial world, whatever is
given to us in sense perception only, is unworthy, unstable and
unreal. The Platonic dialogues are the great source of inspira-
tion for this identification of the real with the intelligible. There
we find in abundance passages in which the objects of the intel-
lect, the Ideas, are eulogized and called "real/' and in which
entities which are merely objects for the senses are called "un-
real." 7

With the intensification of religious interest and the spread
of Christianity, the conviction remains that only that is real
which is intelligible and not essentially sensible. The world of
intelligible entities is regarded somewhat differently. It is now
not so much the realm of secular generalizations and of moral
ideals that are independent of religious import as it is the realm
of spiritual truths, the realm of God, His Word, and His ideas.
The mind, says St. Augustine, 8 "is disabled by besotting and
inveterate vices not merely from delighting and abiding in, but
even from tolerating, His unchangeable light, until it has been
gradually healed, and renewed, and made capable of such felic-
ity." Man is naturally sinful; he usually is occupied with material
things, with the world of sense which is the world of illusion and
unreality. The world of sense, it is felt, has no existence per se.
It has only a shadowy and reflected importance in so far as it is
connected with, and derived from, the spiritual Word of God.
Material things "are known in one way by the angels in the
Word of God, in which are seen the eternally abiding causes and
reasons according to which these things are made; and in an-
other way in which these things are seen as they are in them-
selves. In the former way, they are known with a clearer knowl-
edge; in the latter they are known with a dimmer knowledge,
a knowledge rather of the bare works than of the design." 9 Scat-
tered through the Middle Ages we find marks of this other-world-
liness. That "in which there is any mutable element," says St.
Anselm, 10 "is not altogether what it is. ... And what has a past
existence which is no longer or a future existence which is not
yet, this does not properly and absolutely exist."

51

With the great scientific generalizations formulated in the six-
teenth and seventeenth centuries, the world of intelligible en-
tities finds new inhabitants. The world of intelligible entities
is still a world of spiritual truths. But the ideas of God are clear
and distinct ideas, truths of reason, in a word, mathematical for-
mulae. The world of mere sense is still unimportant and unreal.
Material things have no reality except in so far as they exemplify
mathematical formulae. And we have no real knowledge of mun-
dane things except in so far as we can subject them to number
and see their behavior as the fulfillment of some mathematical
law.

A tremendously important line of philosophers thus presents
us with a doctrine from which we may derive the proposition:
Subsistents appearing as merely sensible are unreal. We have
agreed not to make use, unless there are special considerations, of
a universal negative existential proposition that would assign to
our term "existence" a signification out of accord with common
usage where common usage is precise and definite. If now we
were partially to explain the meaning of our term "existence" by
means of the proposition: "All sensible subsistents are unreal,"
we should be assigning to the realm of the non-existent, not
merely sunsets and soap bubbles as these subsistents are com-
monly presented to us, but also ancient trees and Gothic cathe-
drals. If, then, we found the proposition: "Whatever subsists as
impermanent is unreal" unacceptable because of its divergence
from common usage, there is all the more reason for us to reject
the proposition which we are now considering.

However, "All sensible subsistents are unreal" is to be dis-
tinguished from "all merely sensible subsistents are unreal." Sun-
sets and soap bubbles and Gothic cathedrals may be subsistents
appearing as sensible; but they may not appear as merely sen-
sible. Consequently in assigning the merely sensible to the realm
of the non-existent, we may be leaving the door open for sunsets,
soap bubbles and cathedrals, appearing with the characteristics
with which they normally appear. Is there any respect, however,
in which a subsisting Gothic cathedral appears to be connected
with the eternal truths and some other subsisting sensible entity
not connected? The cathedral appears with the characteristic of
having been built in accordance with the formulae of physics;

52

its behavior exemplifies the law of gravitation. Yet, unless we are
told just what the eternal truths are and what sort of connection
with them is demanded, we have no basis upon which to dis-
tinguish the ontological status of a Gothic cathedral from that of
any other alleged sensible entity. Practically every sensible entity
appears connected, in some sense of the word "connection."
with the realm of intelligible truths. The proposition: "Merely
sensible subsistents are unreal" is ostensibly assigning to the
realm of the non-existent certain sensible subsistents. Yet with-
out a more detailed description of the intelligible and of the
nature of the connection that is demanded, none of the sensible
subsistents normally considered is indicated as falling within the
class of the merely sensible. A universal negative existential prop-
osition is effective in explaining the signification being assigned
"existence" in so far as it assigns a definite characteristic to 'exist-
ence' and in so far as it assigns entities to the realm of the non-
existent. It is hardly informative to be told that existence has the
characteristic of being somehow intelligible. And in assigning the
merely sensible to the realm of the non-existent, it turns out that
we are, in the absence of further propositions, leaving the realm
of the non-existent without any obvious inhabitants. It appears
then that: "All sensible subsistents are unreal" will not assign to
"existence" the sort of signification we seek to give it. And it
appears that: "All merely sensible subsistents are unreal" will not,
taken by itself, give "existence" a definite meaning.

At the beginning of Greek philosophy we meet with the doc-
trine that the impermanent is unreal. For many writers it is the
world of sense which is impermanent. And so we have arrived
at the doctrine that the sensible, or the merely sensible, is unreal.
Instead, however, of opposing to the merely sensible that which
is intelligible, there may be opposed to the merely sensible that
which is independent of sense-perception, that which persists
either unsensed or regardless of whether it is sensed or not. In-
dependence of sense perception has grown into independence of
any mental activity. We come thus to the doctrine known as
realism, the doctrine that whatever is merely or essentially mental
content is unreal, the doctrine that whatever is real is independ-
ent of any mind. A realism of this sort does not find very definite
expression in writings prior to the eighteenth century. It was

53

probably accepted by earlier writers. But the explicit statement
of it seems first to have been called forth by the exposition of
epistemological idealism. It has during the past century been
advocated by many eminent writers. And there is no doubt but
that the proposition: "Essentially mental subsistents are unreal"
establishes a partial signification for the term "existence" which
accords very well with current popular usage.

When we partially explain the signification being assigned
"existence" by means of the universal negative existential propo-
sition: "All essentially mental subsistents are unreal," we are
definitely assigning to the realm of the non-existent subsistents
appearing as dream objects, and we are definitely assigning to the
realm of the non-existent subsistents appearing as members of a
Berkeleian or Kantian world of experience. Moreover, we are
definitely assigning to the realm of the non-existent the ideas
which certain epistemological dualists hold are in all cases the
immediate objects of our consciousness. For these ideas, as con-
trasted with the ulterior realities to which they refer, are nor-
mally thought of as having no life outside of the conscious states
whose immediate objects they are. It follows then, if we may in-
dulge in a digression, that one can hardly be an epistemological
dualist proclaiming the existence of such ideas, if one is partially
to explain the signification of "existence" by means of the propo-
sition: "All essentially mental subsistents are unreal."

If, in partially explaining the signification which we are as-
signing to "existence," we make use of the proposition: "Whatever
is essentially mental is unreal," we shall not be running counter
to common usage. And we shall not be failing to give our term
"existence" any definite meaning at all. We have already com-
mitted ourselves however to the acceptance of the proposition:
"Self-contradictory subsistents are unreal." And we shall dis-
cover later that the entity that is in no sense an object of con-
sciousness is self-contradictory. 11 If then we may assume that our
later finding will be correct, the entity that is in no sense an
object of consciousness is an entity that we shall find presented
to us as self-contradictory. It is an entity, consequently, which our
propositions setting forth the meaning of "existence" definitely
will assign to the realm of the non-existent. If, then, in partially
determining the signification of "existence," we were to make

54

use of the proposition: "All essentially mental subsistents are un-
real/' we should find assigned to the world of non-existence both
the subsistent that is in no sense an object of consciousness and
the subsistent that is essentially mental. We should be placing
practically all subsistents among the unreals and should have
nothing for the term "existent" to denote.

Whereas we have found many writers holding that the merely
sensible is unimportant and unreal, there is a distinguished group
of philosophers who take what is, generally speaking, an opposite
point of view. "Reality and the evidence of sensation," 12 says
Diogenes Laertius in expounding the Epicurean philosophy,
"establish the certainty of the senses; for the impressions of the
sight and hearing are just as real, just as evident, as pain." It is
the entities with which we become acquainted through sense per-
ception which are for these writers most certainly known to be
real. Entities which are merely entities of thought are known less
directly, less surely. In becoming acquainted with them the mind
follows a more tortuous path and is more likely to be led astray.
"Let men please themselves as they will," says Francis Bacon, 13 "in
admiring and almost adoring the human mind, this is certain:
that as an uneven mirror distorts the rays of objects according to
its own figure and section, so the mind, when it receives impres-
sions of objects through the sense, can not be trusted to report
them truly, but in forming its notions mixes up its own nature
with the nature of things." And so Bacon arrives at the position:
"The evidence of the sense, helped and guarded by a certain
process of correction, I retain. But the mental operation which
follows the act of sense I for the most part reject." 14

This acceptance of the reality of entities given to us in sense
perception and this sceptical attitude towards entities not di-
rectly bound up with sense perception finds expression in many
passages in Locke, Berkeley and Hume. "The ideas of sense,"
says Berkeley 15 for example, "are allowed to have more reality
in them . . . than the creatures of the mind." A similar attitude
is frequently expressed by Kant. "What is real in external phe-
nomena," says Kant, 16 "is real in perception only, and can not
be given in any other way." "From such perceptions, whether by
mere play of fancy or by experience, knowledge of objects can
be produced, and here no doubt deceptive representations may

55

arise without truly corresponding objects". . . "In order to
escape from these false appearances, one has to follow the rule
that whatever is connected according to empirical laws with a
perception is real."

"The postulate concerning our knowledge of the reality of
things requires perception, therefore sensation and the con-
sciousness of it, not, indeed, immediately of the object itself, the
existence of which is to be known, but yet of a connection be-
tween it and some real perception according to the analogies of
experience which determine in general all real combinations in
experience. . . . But if we do not begin with experience or do not
proceed according to the laws of the empirical connection of
phenomena, we are only making a vain display as if we could guess
and discover the existence of anything/' 17

It is unnecessary to trace this doctrine, which may be called
"empiricism/' down to our own day. It is the doctrine with so
many recent exponents, the doctrine that entities given to us in
sense perception are real, that entities connected with the objects
of perception, objects of possible but not of actual experience, are
less directly and less surely known to be real, and that entities
not properly connected with sense experiences are unreal. In
view of our discussion of the universal affirmative existential
proposition, we are not interested in the proposition: "All objects
of possible experience are real." But the proposition: "Subsistents
appearing as not properly connected with sense experience are
unreal" is a proposition of which we are at liberty to make use
in partially explaining the signification to be attached to our
terms: "reality" and "existence."

A universal negative existential proposition, let us remind
ourselves again, will be effective in assigning a meaning to
"existence" to the extent to which it definitely assigns entities to
the realm of the non-existent. Which, then, are the entities that
appear as not properly connected with sense experience? Unless
the universal negative existential proposition with which we
are dealing is expanded and the nature of a proper connection
defined, there are no entities which will obviously fall within
the realm of the unreal. Universals generally appear as the arche-
types of the objects of sense experience. God appears with the
characteristic of being implied by the objects of sense experience.

56

Even dream objects when recognized as dream objects frequently
appear as caused by something in the world of sense-experience.
Almost all entities, in short, are subsistents which appear as hav-
ing some sort of connection with the objects of sense-experience.
We can give the world of the non-existent some definite content
and thus more effectively explain "existence" if we disregard the
notion of a proper connection. If we lay down the proposition:
"All subsistents not appearing as percepts are unreal/ 1 God, and
the law of gravitation, and the other side of the moon, are at once
marked out as subsistents that, as they usually appear, do not exist.
Such a proposition, however, would assign to the term "existence"
a signification out of accord both with common usage and with
philosophical precedent.

Let us consider, then, the possibility of limiting reality to
entities given to us as having a certain definite kind of con-
nection with sense experience. The entity that seems merely to
be implied by sense experience is not, we may say, properly con-
nected with it. The only entities that are properly connected with
the actual objects of sense experience, we may say, are those that
are possible objects of sense experience, those entities that would
be perceived if we were at a different place or had senses suffi-
ciently acute. We thus arrive at the universal negative existential
proposition: Whatever appears with the characteristic of being
non-spatial is unreal. And we may in a similar fashion arrive at
the proposition: All timeless subsistents are unreal.

At least as far back as Plato we meet with the doctrine that
whatever is real must have a date. Against the timeless Being
of Parmenides, the objection is raised that such an alleged being
is unreal because it is not in time as an entity must be if it is to
be real. 18 An entity that does not participate in time, it is held,
does not participate in being. When we come down to Hobbes,
we find a similar attitude clearly expressed with respect to spatial
position. "If the triangle exists nowhere at all," Hobbes writes, 19
"I do not understand how it can have any nature; for that which
exists nowhere does not exist." Sometimes it is required of a real
entity only that it have a date, sometimes only that it have spatial
position. But quite frequently the two requirements are joined.
Reality is regarded as something that is limited to those subsist-
ents appearing with both a date and a spatial position. As Crusius,

57

one of the philosophers who wrote shortly before Kant, puts it,
to give an entity that is merely thoughtthat is, an entity that in
his terminology is merely possible a date and a spatial position
is to give it existence. 20 "If a substance is to exist, it must exist
immediately in some place and at some time." 21 For Kant, space
and time are transcendentally ideal but empirically real. Every
external entity that is empirically real that is to say, real as a
phenomenon must be in time and in space. And all real phe-
nomena without exception must be in time. Only events, some
recent writers seem to hold, are real. And an event, it is indicated,
is an entity that has a date and a position in a four-dimensional
spatio-temporal continuum.

If we partially explain "existence" by means of the proposition:
"Whatever appears as lacking a date or a spatial position is un-
real," there are various subsistents that our proposition definitely
assigns to the realm of the non-existent. Such a proposition clas-
sifies as unreal mental processes and mental content presented as
occurring nowhere, universals and scientific generalizations ap-
pearing as eternal, God appearing as a supra-spatial Deity. More-
over with such a proposition we assign to existence the char-
acter of being free from utter non-spatiality and the character
of being free from utter timelessness. Thus it can not be objected
that the proposition which we are considering gives no meaning
to "existence." Nor does this proposition definitely assign to the
realm of non-existence entities which common usage unhesitat-
ingly calls "real." A preliminary and somewhat casual discussion,
in short, fails to eliminate from further consideration: "What-
ever appears as lacking a date or a spatial position is unreal." To
be sure, there are such questions as: date with respect to what?
and spatial position with respect to what? In order to determine
which subsistents are unreal because of their lack of spatio-tem-
poral characteristics, a further discussion of space, time, and of
time-space is indicated. If the signification of "existence" is to
be as precise as possible, the realm of non-existence must contain
more entities than merely those which appear as totally undated
and existence must have a more definite characteristic than free-
dom from utter timelessness. In order to make the meaning of
our term "existence" as precise as possible, we shall later mark
out as unreal all subsistents appearing a? undated, or as lacking

58

a spatial position, with respect to a certain type of entity. 22 And
we shall mark out as unreal subsistents appearing as not having
a certain kind of date and position with respect to such an entity.
But in view of this preliminary discussion and pending such
modifications as our search for precision may later lead us to
make, we may at this point agree in explaining "existence" to
make use of the proposition: "Whatever appears as lacking a date
or as having no spatial position is unreal."

It is frequently felt that existent entities are related to one
another in that each of them is in some place and each of them
at some time. It is felt that existent entities taken together form
a system of entities that is bound up with a system of places or
Space and with a system of dates or Time. The non-existent, it
may be felt, is what does not belong to this system, what does
not fit into this one Space and this one Time. With some writers,
however, membership in this one Space and this one Time does
not seem to be the outstanding determinant of membership in
the system of existent entities. To exist is to be a member of a
system of related entities; but membership in this system is not
primarily a matter of place and time. Existence is evidenced by
a wealth of relations of all sorts with various other entities. The
non-existent is that which subsists disjoined from most other en-
tities and unconnected with them.

It is the consideration of fables and dream objects that is likely
to lead us to distinguish the existent from the non-existent in
this fashion. "And I ought," says Descartes at the end of his
"Meditations," "to set aside all the doubts of these past few days
as hyperbolical and ridiculous, especially that very common un-
certainty respecting sleep, which I could not distinguish from the
waking state; for at present I find a very notable difference be-
tween the two, inasmuch as our memory can never connect our
dreams one with the other, or with the whole course of our lives,
as it unites events which happen to us while we are awake. And,
as a matter of fact, if some one, while I was awake, quite sud-
denly appeared to me and disappeared as fast as do the images
which I see in sleep, so that I could not know from whence the
form came nor whither it went, it would not be without reason
that I should deem it a spectre or a phantom formed by my
brain (and similar to those which I form in sleep) rather than a

59

real man/' Similarly, Christian Wolff 23 holds that "in a dream
while you look at some one, he suddenly changes into some one
else or he vanishes straight-way and no one comes back to take
his place." Things behave in a strange, haphazard, and unrea-
sonable manner. And it is this that distinguishes them from real
entities and marks them as dreams. There is thus suggested to
us another manner in which we might partially describe the sig-
nification o our term "existence." We can not make effective use
of a universal affirmative existential proposition. And so we may
pass by the proposition: Whatever has many points of contact
with our usual experience is real. But perhaps in partially ex-
plaining the signification of "existence" it will be well for us to
make use of the proposition: Whatever appears as out of accord
with our usual experience, as having few points of contact with
the entities of which we are normally aware, is unreal.

Our usual experience reveals to us stones that are mute. A
subsistent stone that talks of its own accord differs from most of
its fellow subsisting stones. It appears as something surprising and
unusual, as something that could not be predicted or accounted
for, as a phenomenon having few points of contact with our
normal experience. If then in explaining "existence" we were
to make use of the proposition: "Subsis tents having few points
of contact with our normal experience are unreal," it would be
the unusual and extraordinary phenomenon, the rara avis, as it
were, that we would be assigning to the realm of the non-exist-
ent. What, however, is usual, and what is unusual? Conversations
with the Virgin Mary were not at all unusual in the Middle Ages;
nor were witches unusual in the New England of Cotton Mather.
The universal negative existential proposition which we are con-
sidering would not definitely and unambiguously assign to the
realm of the non-existent either visions of the Virgin Mary or
women riding on broom-sticks. Again, substances that give off
emanations are unusual in our experience, though pieces of
radium that give off such emanations are not rare. Consequently,
if we start from the consideration of all substances rather than
from the consideration merely of radium, our proposition seems
to assign to the realm of non-existence all substances alleged to
give off emanations. In general, we may say that the instances
of any species are few in number, and that this species is rare,

60

if we start with a genus that is sufficiently extensive. Rarity in
short is a relative thing. And so if mere rarity implies unreality,
membership in the realm of non-existence becomes relative and
indeterminate.

The proposition: "Subsistents having few points of contact
with our normal experience are unreal" does not definitely and
unambiguously point out a limited group of entities as unreal.
Nearly every phenomena is usual, if we take into consideration
the experience of some special group of subjects. And nearly
every phenomenon is rare, when we consider it an instance of an
extremely extensive genus. If we describe more closely the notion
of having many points of contact with normal experience, per-
haps we can arrive at a proposition that will definitely and un-
ambiguously assign a limited group of entities to the realm of the
unreal. Perhaps this can be accomplished by identifying the
phenomenon that is unusual with the phenomenon whose be-
havior is unpredictable. Perhaps it can be accomplished by iden-
tifying the phenomenon that is unusual with the phenomenon
that is observed by but a single subject. If we partially explain
"existence" by limiting reality to entities presented as having
been perceived by more than one subject, we rule out of existence
the fall of the tree of which I am the sole observer. We mark out
as "non-existent/' in our sense of the word "existence," an entity
that common usage obviously calls "real." And if we say that the
unpredictable is unreal, we meet with a problem akin to that
into which we run when we limit reality, not to what is experi-
enced, but merely to objects of possible experience. When is an
entity that is not actually experienced an object of possible ex-
perience? We found this a question to be answered only through
the introduction of other concepts than that of experience,
through the introduction, for example, of the concepts of time
and place. So it is with the question: When is a phenomenon that
is not actually predicted one that might have been predicted?
If we limit reality to what is actually predicted, we mark out as
unreal many entities that common usage calls "real." And if we
limit reality merely to what might be predicted, we are forced
to examine other concepts if we would have our universal nega-
tive existential proposition one that definitely and unambiguously
assigns a limited group of entities to the realm of non-existence.

61

A phenomenon is out o accord with our usual experience
when it is rare, exceptional, and surprising. In a more special
sense, however, a phenomenon may be held to be out of accord
with our usual experience when it fails to conform with the vari-
ous scientific generalizations that are valid for the objects of our
normal experience. There are, it may be held, various laws which
all real phenomena obey. There are, it may be held, various truths
of reason which constitute the form of reality. A phenomenon is
real, it may be said, when it conforms with these intelligible laws,
when its behavior presents material for which these truths of
reason can furnish a supplementing form. And a phenomenon is
unreal, it may be held, when it appears inconsistent with these
intelligible truths. A phenomenon is out of accord with our
usual experience, has few points of contact with the system of
existent entities, it may be said, when it disobeys the laws which
constitute the form of reality.

Leibniz is an outstanding advocate of the doctrine that all
existent entities are intimately bound up with one another
through membership in a systematic network of relations. Each
monad, to be sure, is its own cause; but the monads, taken to-
gether, form an organic system in which each bit is essential.
The world of real entities is, he holds, a system of interrelated
compossible entities. An entity is real if it belongs in the system,
if it sustains the relations that all real entities do sustain towards
one another. And a phenomenon is unreal if it appears to us as
coming without antecedents and as going without consequents,
as a stranger that has no connection with the interrelated world
formed by most subsistents. In passages in which he alludes to
the difference between the real and the unreal, Leibniz suggests
some of the doctrines that we have just been discussing. An entity
is real and belongs in the system of interrelated entities only
if it harmonizes with our normal experiences. He uses such
phrases as "agreement with the whole course of life" 24 and al-
ludes to the phenomenon of "future things" being "in a certain
degree . . . foreseen from past things." 25 There also appears how-
ever the more special doctrine that real entities are those which
conform with certain truths of reason, with certain intelligible
laws. "The basis of the truth of contingent and singular things," 26
he writes, "is in the succession which causes these phenomena of

62

the senses to be rightly united as the intelligible truths demand/'
We thus elicit the universal negative existential proposition:
Subsistents appearing as inconsistent with this, or with that, in-
telligible law are unreal. If we accept as an intelligible law the
proposition that every event has a cause, then any subsistent
appearing as an uncaused event is, in accordance with the propo-
sition which we are considering, marked out forthwith as unreal.
Our proposition does not fail to assign a definite group of en-
tities to the realm of the non-existent. It marks out as unreal a
group of entities that will be definite in proportion as our intel-
ligible laws are expressed with precision; and it marks out as
unreal a group of subsistents that will vary with the particular
propositions that are accepted and laid down as intelligible laws.
Nor, if our intelligible laws are carefully chosen, does it appear
that the universal negative existential proposition which we are
considering will assign to the realm of the non-existent any en-
tities which common usage unhesitatingly calls "real." The prob-
lem we run into, however, is the problem: Which propositions
are to be regarded as together constituting the intelligible laws?
The proposition: "Every entity has a date and a spatial position"
may be regarded as an intelligible law. And the proposition:
"Every entity is self-consistent" may be regarded as an intelligible
law. To the extent to which these propositions constitute the
system of intelligible laws, we have already committed ourselves
to the acceptance of the proposition: "Whatever is inconsistent
with the intelligible laws is unreal." For this proposition now
reduces to the proposition: "Whatever appears as lacking date
or spatial position is unreal" and to the proposition: "Whatever
subsists as self-contradictory is unreal."

When, in partially describing the signification to be attached
to the term "existence," we choose to make use of the proposi-
tion: "Whatever subsists as inconsistent with the intelligible laws
is unreal," there is one consequence which ensues which I should
like to point out. The entity which subsists as inconsistent with
some intelligible law is by our proposition forthwith assigned to
the realm of the non-existent. If the proposition: "The quantity
of matter is always constant" is regarded as an intelligible law,
then any phenomenon involving an increase or a decrease in the
quantity of matter is forthwith marked out as an unreal and il-

63

lusory phenomenon. Our intelligible laws, consequently, turn
out to be immune to overthrow by what are known as negative
instances. For the negative instance, instead of weakening or
destroying the validity of the intelligible law, is itself immedi-
ately ruled out as an illusory and unreal phenomenon.

These remarks apply with especial force to Kant, in whose
writings the intelligible truths are developed in some detail. The
most important of what we may call the intelligible laws seem
for Kant to be the propositions discussed in the Analogies of
Experience. In order to be real, a phenomenon must be given to
us as consistent with the intelligible laws; and we are not left
entirely in the dark as to what these intelligible laws are. For one
thing, in order that a given phenomenon may be real, it must not
in its behavior contradict the proposition that the quantity of
substance is constant. For another thing, it must not contradict
the proposition that every event has a cause. And for still another
thing, it must not contradict the proposition that there is dynam-
ical interaction between contemporaneous entities. These three
propositions discussed in the Analogies of Experience constitute
for Kant a part, though not the whole, of what we may call the
intelligible laws. And if we have these propositions in mind when,
in partially explaining "existence/* we make use of the proposi-
tion: "Whatever subsists as inconsistent with the intelligible laws
is unreal/' then the phenomenon that appears, for example, as
uncaused is immediately marked out as a phenomenon that is un-
real. The proposition that every event has a cause comes to be a
proposition whose validity does not rest upon experience. It
comes to be a proposition which can not be over-thrown by any
experience; for any phenomenon seeming to contradict it that
might be presented to us would immediately be marked out as
illusory and unreal. The causal law, in a word, comes to be a
presupposition of experience. But it comes to be a presupposi-
tion of valid experience only in the sense that it is being taken as
one of the intelligible laws to which we refer when, in partially
explaining "existence/* we say that whatever is given to us as in-
consistent with the intelligible laws is unreal. We come thus, by a
somewhat different route, to a position that has already been ex-
pressed in the previous chapter. "In laying down the causal law,
Kant is implicitly determining the signification of 'existence/ "

64

And so it appears "that the validity which Kant finds for the
causal law ... is only the validity which attaches to a proposition
determining the meaning of a term." 27

We might go on to consider a number of universal negative
existential propositions that we have not yet discussed in detail.
With respect to each of them, we might ask whether it assigns a
definite group of entities to the realm of the unreal and attaches
to existence freedom from some clearly described characteristic.
With respect to each of them we might also ask whether it defi-
nitely assigns to the realm of the non-existent entities which com-
mon usage unhesitatingly calls "real." In short, we might bring
up for consideration universal negative existential propositions
ad nauseam. And with respect to each of them we might ask
whether it is the sort of proposition of which we can well make
use in partially describing what we are to call "existence." We
have however already met with some positive results. We have
agreed to make use of the proposition: "Self-contradictory sub-
sistents are unreal." And we have agreed to make use of the prop-
osition: "Subsistents appearing as lacking a date or as lacking a
spatial position are unreal." Perhaps then we can forego a more
extended survey of the writings of the past. Perhaps we can fill
out for ourselves the group of universal negative existential prop-
ositions of which we are to make use in partially explaining our
term "existence."

With respect to logical self-consistency, one universal negative
existential proposition is as suitable as another to the task of ex-
plaining "existence." Our selection of one universal negative
existential proposition in preference to another is a matter of
choice and not a matter of logical compulsion. We have stated
however the considerations on which our choice will be based. 28
And, on the basis of these considerations, there are certain prop-
ositions of which we have already agreed to make use.

Universal negative existential propositions, we have seen, can
not by themselves completely determine for the term "existence"
a meaning that will be sufficiently precise. We must in addi-
tion make use of individual affirmative existential propositions
and of individual negative existential propositions. However,
other things being equal, the greater the number of universal
negative existential propositions of which we make use, the more

65

precise does the meaning of our term "existence" become. We
have consequently the task of joining additional universal nega-
tive existential propositions to the proposition: "Self-contradictory
subsistents are unreal" and to the proposition: "Subsistents ap-
pearing as lacking all date or all position are unreal/'

Moreover, we have the task of assuring ourselves that the uni-
versal negative existential propositions of which we have already
agreed to make use are sufficiently unambiguous and clear.

In a general way, however, we are at this point ready to
enumerate the propositions which taken together will explain
the meaning which "existence" is to have in the constructive
parts of this treatise. We are ready to set ourselves to the task of
laying down a number of universal negative existential proposi-
tions, each as clear in its expression and as unambiguous in its
reference as possible; and to the task of supplementing these prop-
ositions with singular or particular existential propositions both
affirmative and negative, with lists, that is to say, both of some of
the entities that are included in, and of some of the entities that
are excluded from, the denotation of "existence" in our sense
of that term. We are ready in short to address ourselves in earnest
to the task of laying down the group of existential propositions,
which, taken together, are to occupy in our metaphysics a posi-
tion similar to that which Descartes intended for his: "Cogito
ergo sum."

Summary

We are at liberty to determine what meaning we are going to
attach to our term "existence." Still we want the meaning we
choose to conform with the common meaning of "existence" in
so far as the latter can be determined and applied in such a way
as to mark out definite groups of entities as existent and definite
groups as non-existent.

One feature of the common meaning of "existence" is the
'hardness' of facts, the imperviousness of reality to expansion or
contraction through mere thinking. In order that what we call
"existence" may have this characteristic, our propositions explain-

66

ing our term "existence" must be limited to universal negative
ones, supplemented by individual or particular propositions.

Which universal negative propositions shall we use in our
explanation? Various possibilities are considered from two angles:
(1) would they determine for our term "existence" a meaning
somewhat in accord with what "existence" commonly means-
realizing of course that the common meaning is hazy; and (2)
would they determine for our term "existence" a meaning from
which it will follow that certain entities, but not all entities, are
definitely marked out as unreal.

None of these considerations are binding. They merely incline
us to give our term "existence" one meaning rather than another.

HOW WE SHALL USE THE TERMS: EXISTENCE
AND REALITY

There is no entity that is not what we are calling a "subsistent."
The world of subsistents includes the man walking on my ceil-
ing, God, everything, goodness, greater-than, mathematics. It
includes everything that can be mentioned and everything that
can not be mentioned, my alleged objects, your alleged objects,
and entities alleged to be objects for no one. It is this unlimited
field, including all entities that may be held to be real and all
entities that may be held to be unreal, that forms our universe
of discourse and is to be dichotomized into the real and the un-
real. It is difficult to refer to this unlimited field of subsistents
without appearing to hold that its members exist. If we say that
the man walking on my ceiling is a subsistent, the use of the
word "is" may create the impression that this subsisting man
exists. And a similar impression may be created by the remark
that this man has the characteristic of walking on my ceiling.
This danger of misinterpretation can not be completely overcome.
We shall refer to the man on my ceiling as a subsistent or an ap-
pearance. And instead of saying that he has a certain character-
istic, we shall, for the most part, say that he appears with this
characteristic or is presented with this characteristic. But it is not
to be assumed that a subsistent which appears is an appearance of
something which is real. Nor is it at this point to be assumed that
to appear is to appear to some conscious subject. Entities appear-
ing with the characteristic of being objects for no one are sub-
sistents. They too will be called "appearances." They too are

68

included in the unlimited field which it is our task to dichoto-
mize into the real and the unreal.

Some of the subsistents that are to be called "real" and some
of the subsistents that we are to call "unreal" may be dismissed
from our attention until we come to the end of this chapter.
There we shall enumerate certain real, subsistents and certain
unreal subsistents. For we have planned to deal with certain mem-
bers of our universe of discourse individually. The existential
status of these subsistents will be determined, and our term
"existence" explained in so far as it applies to them, by means of
singular or particular existential propositions, some positive and
some negative.

In determining the meaning of our term "existence," we have
agreed to make use of such propositions as "Xi exists" and
"X 2 exists": and we have agreed to make use of such propositions
as "Yi does not exist" and "Y 2 does not exist." But we have also
agreed to make use of universal negative existential propositions,
of propositions of the type: "No A's exist." Indeed, proposition
for proposition, our universal negative existential propositions
will describe the signification we are assigning "existence" more
fully than will our singular existential propositions. And so it is
to the selection of certain universal negative existential propo-
sitions that we turn.

There is one such proposition that we have already agreed to
use, namely, "No self-contradictory subsistents are real." Self-
contradictory subsistents, however, are subsistents which appear
as self-contradictory. The King of England who resides in Buck-
ingham Palace is a self-contradictory subsistent in so far as I
think of him as self-contradictory. And a square circle is not a
self-contradictory subsistent when its alleged squareness and its
alleged circularity do not appear as mutually contradictory. The
proposition: "No self-contradictory subsistents are real" marks out
as unreal the King of England who appears as self-contradictory
and the square circle which appears as self-contradictory. But it
does not determine the existential status of either the subsisting
King of England who does not appear as self-contradictory or of
the subsisting square circle which does not appear as self-con-
tradictory. Whatever appears as self-contradictory is, however, we

69

say, unreal. Existence, as we describe it, is characterized by free-
dom from explicit self-contradiction.

There is the subsistent which appears as round and not-round.
And there is the subsistent which appears as real and unreal. If
we partially determine the signification of "existence" by means
of the singular proposition: "X a exists," then Xi, let us suppose,
appears as real. But Xi may have been presented as, and may
appear as, a self-contradictory or unreal subsistent. One might
choose to use "real" in such a way that entities enumerated as
real are in all cases real. One might hold, that is to say, that
singular affirmative existential propositions used in determining
the signification of "existence" are a court of final authority, that
unreal appearances and self-contradictory appearances are real if
they are enumerated as real. Let us however choose the opposite
path. Let us hold that self-contradictory appearances and unreal
appearances are unreal even though they are enumerated as real.
Or, rather, let us agree to enumerate as real no subsistents which
appear as unreal and no subsistents which appear as self-contra-
dictory. If we partially determine the signification of "existence"
by means of the proposition: "No A's exist," let us agree to limit
the entities represented by the subject-terms of our singular af-
firmative existential propositions to "Xi not appearing as an A,"
"X 2 not appearing as an A," etc. 1

Self-contradictory appearances are in all cases unreal. Whatever
appears as round and not-round, and hence as self-contradictory,
does not exist. But what about the subsistent which appears as
round and square? Round and not-round are explicitly A and
non-A, round and square less explicitly so. To be sure, each
square subsistent that I consider readily takes on the appearance
of not-roundness. As soon as the quality of being not-round is
suggested to me, I recognize this quality as being an additional
characteristic with which my subsistent is appearing. The subsist-
ent under discussion is a subsistent appearing as square which
enlarges itself to be a subsistent appearing as square and appear-
ing as not-round. With respect to subsistents which thus enlarge
themselves, we may say that there are implicit characteristics with
which they appear. The subsistent which I am considering ap-
pears explicitly, let us suppose, as round and as square; and it

70

appears implicitly as not-round and as self-contradictory. Such
subsistents which appear implicitly as self-contradictory are, let
us say, unreal. Whereas we have already assigned to the realm of
-non-existence the subsistent appearing explicitly as round and
as not-round and as self-contradictory, let us also assign to this
realm the subsistent appearing explicitly as round and as square
and only implicitly as self-contradictory. Let us, that is to say, lay
down the additional proposition: "Whatever appears implicitly
self-contradictory is unreal.' 1 Or, to put it another way, no sub-
sistent is real whose explicit and implicit appearances appear to
contradict one another.

This disposes of the round square subsistent which enlarges
itself to become a round, square, not-round, self-contradictory
subsistent. But, whereas the subsistent which we have been dis-
cussing grows from a subsistent appearing as square to a subsistent
appearing as not-round, the unlimited world of subsistents con-
tains round square subsistents which do not thus enlarge them-
selves. It is conceivable, for example, for some one to hold that
round squares are not self-contradictory. The subsistent which he
considers, or can be imagined to consider, does not grow. As we
have already in effect noticed, there is no universal negative
existential proposition that will eliminate from reality the round
square subsistents which neither explicitly nor implicitly appear
as self-contradictory. 2 But we can agree not to enumerate as
real any of these non-growing round squares which do not,
even implicitly, appear as self-contradictory. Our procedure must
be to trace the growth of a subsistent from round and square to
round, square, not-round and self-contradictory; and then to de-
termine to enumerate as real no other subsistent which appears as
round and as square. No A's, we may say, are real; and no sub-
sistents implicitly appearing as A's. We may trace some subsistent
S to the point where it appears as an A. We may, that is to say,
point out some S which implicitly appears as an A. But all S's are
unreal only in so far as we thereupon resolve to enumerate no
S's among the entities we call existents. Subsistents which im-
plicitly appear as self-contradictory are unreal. And when we 3
show one subsistent to appear as implicitly self-contradictory, all
subsistents differing from it merely in that they do not appear as

self-contradictory are likewise unreal. For we have resolved not
to enumerate such resembling subsistents among the entities we
call ''real/' We may point out some S which implicitly appears
as an A. Therefore, we may say, no S is real. But this will be but
a short-hand and condensed way of assuming that our resolve to
enumerate no S's among our existents will be carried out.

Some subsistent appearing as round and square appears with
explicit and implicit characteristics which appear to contradict
one another. So too with the Cretan who appears as truly asserting
that no Cretan ever speaks the truth. My subsistent is a Cretan
making a certain true assertion. As an outgrowth of my original
object, I am led to consider an alleged situation in which no
Cretan ever speaks the truth. I come to consider various alleged
mendacious Cretans, among them my Cretan informant. Indeed I
come to consider my Cretan informant in the act of falsely as-
serting that no Cretan ever speaks truly. My subsistent is a Cretan
appearing explicitly with the characteristic of having just made a
certain true assertion. And this subsistent has grown to be a
Cretan appearing with the characteristic of having just made a cer-
tain untrue assertion. It implicitly appears as self-contradictory. The
Cretan that I am discussing is consequently unreal. And all sub-
sisting Cretans who appear to be truly asserting that no Cretan
ever speaks the truth, all such Cretans, whether they appear as
self-contradictory or not, will be eliminated from our lists of
real entities.

Obviously veracious Cretan does not enlarge itself to become
mendacious Cretan and self-contradictory Cretan as readily as
round square enlarges itself to become round, not-round, self-
contradictory square. There are intermediate subsistents to be
presented and these intermediate subsistents may not spontane-
ously offer themselves for discussion. Veracious Cretan not appear-
ing as implicitly self-contradictory is a more common subsistent
than round square not appearing as implicitly self-contradictory.
But it is all subsistents, whether they be common or uncommon,
that are to be dichotomized into the real and the unreal. It is a
veracious Cretan asserting that no Cretan ever speaks the truth
who implicitly appears as self-contradictory. Remaining in the
unlimited universe of subsistents which is prior to the distinc-
tion between the real and the unreal, one may perhaps describe

72

this Cretan as developing into a self-contradictory Cretan. But it
'Veracious Cretan is implicitly self-contradictory Cretan" is to
apply to the veracious Cretan who does not appear self-contra-
dictory as well as to the Cretan who does so appear, it will not
suffice to trace the growth or enlargement of a given subsistent. 4
In this treatise the growth or development of certain subsistents
is traced and the implicit appearances of these subsistents re-
vealed. Similar subsistents which do not so develop the opinions,
we may say, of those who do not agree with the developments we
trace are disposed of through our determination not to list as
real any subsistents similar to those which, as we develop them,
are implicitly unreal.

We have so determined the signification of our term "exist-
ence" that all round, not-round subsistents are unreal. We have
so determined the signification of "existence" that round squares
are unreal. And we have so determined the signification of "exist-
ence" that any Cretan appearing as truly asserting that no Cretan
ever speaks the truth is also unreal. No entity will be called "real"
that, except for its development, is indistinguishable from a sub-
sistent which our discussion reveals to us as self-contradictory.

How is it now, we may ask, with respect to the subsistent ap-
pearing as in no sense an object of consciousness? The thesis that
only ideas exist is frequently regarded as the doctrine on which
modern idealism is founded. Many idealists assert that only ideas
exist, holding that entities not ideas and entities not objects of
consciousness are self-contradictory. Let us then examine this
alleged contradiction at this point. Let us see if 'entity appearing
as in no sense an object of consciousness* develops into 'entity
appearing as self-contradictory/

"How say you, Hylas," asks Philonous, 5 "can you see a thing
which is at the same time unseen?" "No," answers Hylas, "that
were a contradiction." "Is it not as great a contradiction to talk
of conceiving a thing which is unconceived?" "It is," admits
Hylas. And he continues: "As I was thinking of a tree in a solitary
place where no one was present to see it, methought that was to
conceive a tree as existing unperceived or unthought of: not con-
sidering that I myself conceived it all the while." As Berkeley 6
explains it, "the mind, taking no notice of itself, is deluded to
think it can and does conceive bodies existing unthought of, , . ,

73

though at the same time they are apprehended by ... itself."

During the present century as realism has renewed its vigor in
Great Britain and in America, the fundamental doctrines of
idealism have been re-examined. "No thinker to whom one may
appeal/' admits Perry, 7 "is able to mention a thing that is not idea
for the obvious and simple reason that in mentioning it he makes
it an idea." Consequently, we are unable to discover what things
are as unknown. "In order to discover if possible exactly how a
thing is modified by the cognitive relationship, I look for instances
of things out of this relationship in order that I may compare
them with instances of things in this relationship. But I can find
no such instances, because 'finding' is a variety of the very relation-
ship that I am trying to eliminate." 8 There is this barrier, which
Perry calls the "ego-centric predicament," which prevents me
from using ordinary methods to discover what difference know-
ing makes to objects. But this predicament, Perry holds, does
not justify me in concluding that knowing makes all the difference
between existing and not existing. "Every mentioned thing is an
idea . . . But what the idealist requires is a proposition to the
effect that everything is an idea or that only ideas exist." 9

"We can not be aware of an entity that is not in some sense an
object. Therefore the entity that is not in some sense an object
does not exist." Here there is an obvious non-sequitur. If this
were the best argument the idealist could put forth, Perry would
be justified in regarding the ego-centric predicament as a method-
ological difficulty without ontological implications. But the real
point of the idealist's proper argument is not that the entity that
is in no sense an object of consciousness is undiscoverable. His
real point is that this entity appears self-contradictory. Indeed,
in some sense, the entity that is in no sense an object of con-
sciousness can be discovered, can be mentioned, can be thought
of. For we seem in the present paragraph to be discussing, men-
tioning and considering: 'the entity that is in no sense an object
of consciousness/ If this entity were not a subsistent at all, we in-
deed could not conclude that this entity is non-existent. We
would be in the predicament of not being able to assert that this
entity exists or that it does not exist. But the idealist asserts that
it does not exist; and the realist asserts that it may exist. In mak-
ing such assertions they claim to be discussing the entity that is

74

in no sense an object of consciousness. Their assertions exemplify
the fact that the entity that is in no sense an object of conscious-
ness can to some degree be discussed and considered. 10

There seems to be given to me, as a subsistent whose ontologi-
cal status is to be discussed, 'the entity that is in no sense an ob-
ject of consciousness/ Indeed at the present moment it is this
subsistent that I am considering. This entity appears as in no sense
an object of consciousness. And yet, as soon as the characteristic
of being in some sense an object suggests itself, I recognize this
characteristic as an additional appearance of the subsistent that I
am considering. Implicitly my subsistent appears as an entity that
I am considering, as in some sense an object of consciousness. The
entity that is in no sense an object of consciousness explicitly ap-
pears with the characteristic of being in no' sense an object of
consciousness. And implicitly it appears with the characteristic
of being in some sense an object of consciousness and hence with
the characteristic of being self-contradictory. However, "what-
ever appears implicitly self-contradictory is unreal." 11 We hold,
therefore, that the subsistent which we have been considering,
the entity that appears as in no sense an object of consciousness,
is unreal. And we resolve to list as real no 'entity that is in no
sense an object of consciousness' even if it does not appear as
self-contradictory.

The subsistent which we have been considering develops, it
may be agreed, into a subsistent which appears as self-contra-
dictory. But, it may be held, there are subsistents which no one
considers. The subsistent which no one considers is, however,
the very subsistent whose development we have just traced. The
subsistent which no one considers is the entity which is in no
sense an object of consciousness. I may refer to each member of
the world of subsistents. And when I talk about all subsistents,
there is no subsistent that as I trace its development does not
take on the characteristic of being in some sense an object of con-
sciousness. It is no ego-centric predicament which makes non-ob-
jects unreal. If non-objects could not be discussed, they could
not be asserted to be unreal. Rather, neither their reality nor
their unreality could be discussed. But the very fact that the
realist holds that some of these non-objects may be real is evidence
that these non-objects are not outside what we call the world of

75

subsistents. Non-objects appear both as non-objects and as objects.
And they are unreal because self-contradictory subsistents are
unreal. The self-contradictory subsistent which appears self-con-
tradictory is unreal because of the universal negative existential
proposition which partially determines the signification of our
term "existence." And the self-contradictory subsistent which does
not appear self-contradictory, the entity in no sense an object of
consciousness, for example, which, as it was apparently presented
to Perry, does not develop the appearance of self-contradictoriness,
this entity is unreal because of our resolve not to list it as real.

Whatever appears implicitly contradictory is unreal. And, we
may add, whatever appears, explicitly or implicitly, as in no sense
an object is unreal. The world of reality is free of subsistents ap-
pearing as non-objects. It contains no entities precluded from
appearing as objects. It is by no means to be concluded however
that each real entity is an immediate datum or object for some
conscious subject. The proposition that no subsistents are real
which appear as in no sense objects does not imply the non-exist-
ence of indirect objects or of entities referred to but not im-
mediately given. For the entity that is in some fashion referred
to is not an entity that is in no sense an object. The entity that is
in some fashion referred to can develop the appearance of being
in some sense an object without developing the appearance of
self-contradictoriness. Likewise it is not to be concluded that each
real entity is definitely and fully presented. Perhaps no one knows
whether Descartes' great-great-grandfather was tall or short. Per-
haps Descartes' great-great-grandfather is a subsistent which ap-
pears with few characteristics. It is a subsistent, let us suppose,
which appears with the characteristic of being one of Descartes'
ancestors, but without name, nationality or size. Nevertheless this
subsistent can develop the appearance of being in some sense an
object without developing the appearance of self-contradictori-
ness. I may refer to each member of the unlimited world of sub-
sistents. But this is very different from cataloging and describing
each subsistent. Subsistents appearing as in no sense objects are
unreal. But, so far as we have yet seen, subsistents appearing with
vague and barren characteristics may or may not be real.

"I know that there are Chinamen, but I know no individual
Chinamen. ... I may be able to think the universe, but may

76

know little of its details. It is therefore evident," says Spaulding, 12
"that there are two kinds of knowing." There is the full, detailed
and explicit manner in which the pen with which I am writing
appears as a subsistent. There is the vague indefinite and unde-
tailed manner in which 'everything' appears. Indeed there are
shades of definiteness, of fulness of content, between and at either
end. A centaur is a subsistent which I consider when I seem to
think of an animal with the body of a horse and the head of a
man. The same subsistent appears more vaguely when I seem to
think of a certain fabulous creature; and still more vaguely when
I seem to think of a given subsistent. Whatever appears with the
characteristic of being in no sense an object of consciousness is
unreal. But up to this point we have not excluded from the world
of existents, as we are to use the term "existence/* either 'a
certain fabulous creature* or 'a given subsistent/

There is the subsistent which appears simply as a fabulous
creature. And there is the subsistent which is less vague, which
appears with more detailed characteristics, the subsistent which
appears as the centaur who attempted to carry off Dejanira, the
wife of Hercules. It is no doubt possible for these subsistents to
be distinguished from one another and to be regarded as two.
Nevertheless as there suggest themselves the characteristic of
having the head of a man, the characteristic of having the body of
a horse and the characteristic of having attempted to carry off
Dejanira, I recognize these characteristics as implicit appearances
of the 'certain fabulous creature* that I was already considering.
'A certain fabulous creature* has developed into 'the centaur who
attempted to carry off Dejanira' just as 'round square* may de-
velop into 'round, not-round, self-contradictory square/ 18 We
have, to be sure, distinguished the Barbarossa who appears to
have died in Asia Minor from the Barbarossa who appears to be
now asleep in a cave. 14 When I begin by considering a Barbarossa
who died in Asia Minor and then come to consider a Barbarossa
now asleep in a cave, I find that a characteristic of my former
subsistent has been wiped out; I find that my subsistent has not
developed but has, on the contrary, been displaced by another
subsistent. A Barbarossa dead in Asia Minor which develops into,
which implicitly appears as, Barbarossa now asleep in a cave is,
let us suppose, unreal. But a Barbarossa dead in Asia Minor which

does not so develop is, let us suppose, real, and is to be dis-
tinguished from Barbarossa appearing as asleep in a cave.

The fabulous creature which develops into the centaur who
attempted to carry off Dejanira is not unreal because of any lack
of definite characteristics. But what shall we say with respect to a
subsistent described as 'a fabulous creature* which does not
so develop? It too the universal negative existential propositions
thus far adopted do not determine to be unreal. For, whereas this
fabulous creature appears neither explicitly nor implicitly with
definite characteristics, it does not develop the characteristic of
being in no sense an object of consciousness and does not im-
plicitly appear as self-contradictory. There is the fabulous crea-
ture which implicitly appears as the centaur who attempted to
carry off Dejanira. And there is the fabulous creature which does
not have any explicit definite appearances. But creatures which do
not have any explicit definite appearances may again be divided.
There is the fabulous creature of this sort which has, or develops,
the characteristic of being definitely presented to no conscious
subject. And there is the fabulous creature which, whereas it
does not develop any definite characteristics as we continue to
consider it, develops the appearance of appearing with definite
characteristics to some one. When I think of paleontology, for
example, I think of nothing definite. And, since I know no
paleontology, as I continue to consider paleontology my
subsistent continues without definite appearances. But my sub-
sistent takes on the characteristic of appearing with more details
to paleontologists. On the other hand, as I consider the millionth
digit in the square root of two, not only does my subsistent not
take on the characteristic of being, let us say, an eight or a nine,
but, since it seems to me that no one will carry the square root of
two out to a million places, my subsistent takes on the character-
istic of appearing to no one as a definite number.

In holding self-contradictory subsistents to be unreal and in
holding subsistents appearing as non-objects to be unreal, we do
not mark those subsistents as unreal which appear with the
characteristic of being definite appearances for no one. We are at
liberty to determine the signification of ' Existence' ' in any man-
ner that we find convenient. But to permit those subsistents to be
real which appear to be definite appearances for no one is to make

78

no attempt to exclude from the world of reality those 'given en-
tities' and 'certain subsistents' which seem to be thoroughly use-
less. Let us then determine the signification of "existence" in such
a manner that it will follow that subsistents with merely vague
and undetailed appearances may in some cases be real. But let
us hold that subsistents appearing with the characteristic of ap-
pearing to no one in a detailed manner are unreal. Indeed, let
us rule out of existence, not merely those subsistents which ori-
ginally appear as detailed appearances for no one, but also those
subsistents which take on this characteristic when it suggests it-
self. Let us partially determine the signification of "existence/*
that is to say, by laying down the universal negative existential
proposition: "Subsistents explicitly or implicitly appearing as
definite appearances for no one are unreal." And let us resolve to
list as real no subsistent which, except for its development, is
indistinguishable from a subsistent which we find taking on the
characteristic of being a definite appearance for no one.

We have in the preceding chapter adopted the rule that "our
universal negative existential propositions should be so chosen
that they mark out fairly definite groups of entities that are being
assigned to the realm of non-existence." 15 In view of the fact,
however, that there are so many shades of vagueness with which a
subsistent may appear, the universal negative existential prop-
osition which we have just laid down does not seem entirely satis-
factory. Is the subsistent that is unreal the subsistent which appears
with the characteristic of appearing to no one with as many as
four details or is it the subsistent which appears with the charac-
teristic of appearing to no one with as many as forty-four details?
I believe that, without attempting at this point further to refine
the distinction between vague appearances and detailed appear-
ances, the universal negative existential proposition just laid down
will be found to mark out some subsistents as unreal and to give
some characteristic to 'reality/ We shall to some extent determine
what is vague and what detailed as various subsistents are con-
sidered in the course of this treatise. We shall, that is to say, point
out certain subsistents that appear with the characteristic of ap-
pearing to no one in a sufficiently detailed manner, subsistents
that, we shall hold, the proposition just laid down marks out as
unreal.

79

Appearing with the characteristic of appearing definitely to no
one is to be distinguished from appearing without the character-
istic of appearing definitely to some one. A subsistent is not un-
real because it appears without a given characteristic. A sub-
sistent is unreal when explicitly or implicitly it appears with the
characteristic of being in no sense an object, with the character-
istic, that is to say, of appearing to no one. And a subsistent is
unreal when explicitly or implicitly it appears with the character-
istic of appearing definitely to no one. But the subsistent which
does not have or develop such an appearance, or the appearance
of being self-contradictory, this subsistent, considering the impli-
cations that may be deduced from the universal negative exist-
ential propositions thus far adopted, need not be unreal. Likewise
the subsistents which do not resemble one that we find developing
the appearance of self-contradictoriness or the appearance of be-
ing no one's definite object, these too, so far as our present re-
solutions carry us, may be listed among existing entities.

The universal negative existential propositions that we have
thus far laid down in partially determining the meaning of our
term "existence" have in one way or another suggested them-
selves to us as a consequence of our interest in the self-contra-
dictory. In the previous chapter we agreed to take as one starting
point the proposition: "Self-contradictory subsistents are unreal."
We also agreed, however, to make use of the proposition: "What-
ever appears as lacking a date or as having no spatial position is
unreal." From this latter proposition it follows that subsistents
appearing as lacking any date or position are unreal along with
subsistents appearing as self-contradictory, subsistents appearing
as non-objects, and subsistents appearing as definite appearances
for no one. Let us also lay down the proposition that subsistents
developing the appearance of utter non-spatiality or the appear-
ance of utter non-temporality are non-existent. And let us re-
solve to list as real no subsistents, which, except for their develop-
ment, are indistinguishable from those which in this treatise we
find takiog on the appearance of utter non-temporality or the ap-
pearance of utter non-spatiality.

No subsistents are real that explicitly or implicitly appear as
lacking all spatial position. No subsistents are real that explicitly
or implicitly appear as utterly undated. What shall we say, how-

80

ever, with respect to the entity that explicitly or implicitly ap-
pears dated with respect to one entity but not with respect to
another? Cinderella left for the ball before she lost her slipper.
The loss of the slipper is presented as occurring after the depar-
ture for the ball. But it is presented, let us assume, as having
neither preceded nor followed the fall of Constantinople. The
fall of Constantinople, we likewise suppose, appears as having
neither preceded nor followed the loss of the slipper. But whereas
the fall of Constantinople appearing as not temporally related to
the loss of the slipper appears without the claim that the loss of
the slipper is nevertheless real, the loss of the slipper, appearing as
not temporally related to the fall of Constantinople, does, we sup-
pose, appear with the claim that the fall of Constantinople is
nevertheless real. The fall of Constantinople presented in this
fashion may, we hold, be real. The loss of the slipper we hold to
be unreal. We mark out as unreal that entity which explicitly or
implicitly appears as utterly undated, as undated with respect to
any entity. And we also mark out as unreal that entity which ex-
plicitly or implicitly appears as undated with respect to some other
entity while appearing explicitly or implicitly with the claim that
that other entity is nevertheless real.

It is one thing to appear with the characteristic of lacking any
date. It is another thing to appear without the characteristic of
having a date. As we use the term "existence," an entity is not
unreal in so far as it appears without a given characteristic. It is
unreal if it appears, explicitly or implicitly, with a given charac-
teristic, with, for example, the characteristic of having no date
with respect to any entity, or with the characteristic of having no
date with respect to some other entity and with the claim that
this other entity is real. The entity that appears without the
characteristic of having a date and without the characteristic of
having no date may be real just as may the entity that appears
with the characteristic of having a date. It is the entity that appears
with the characteristic of having no date that is unreal; and the
entity that appears with the characteristic of having no date with
respect to an entity that appears real.

Without forgetting that the subsistent may be real that appears
without the characteristic of having a date and without the
characteristic of having no date, let us consider a subsistent that

81

appears with the characteristic of having a date. A subsisting
Socrates, let us suppose, appears with the characteristic of having
a date with respect to Plato. And a phase of Socrates' life appears
with the characteristic of being present, rather than past or fu-
ture, with respect to a phase of Plato's life. The phase of So-
crates' life which appears with the characteristic of being present
with respect to a phase of Plato's life may appear with the charac-
teristic of having a spatial position with respect to that phase of
Plato's life or with the characteristic of having no spatial posi-
tion with respect to that phase of Plato's life; or it may appear
without either characteristic. But if it appears with the character-
istic of having no spatial position with respect to an entity which
appears real and with respect to which it appears to be present,
then, let us say, it is unreal. As we use the term "reality," if we
may be permitted to sum up the connections that have up to this
point been brought out between existence, time and space, a sub-
sistent is unreal if it appears with the characteristic of having no
position with respect to any entity or with the characteristic of
having no date with respect to any entity. Moreover it is unreal
if it appears with the characteristic of having no date with respect
to an entity that appears real; or if it appears with the character-
istic of having no position with respect to an entity which appears
real and with respect to which it appears present.

An entity is unreal if it appears both real and unreal and hence
as implicitly self-contradictory, or if it appears temporally un-
related to an entity that appears real. It would of course be mere
tautology to say that an entity is unreal if it is unreal. And it
would be circular to say that an entity is unreal if it appears
temporally unrelated to an entity that is real. But the world of
subsistence which we are attempting to dichotomize includes,
among other subsistents, some subsistents appearing as real and
some subsistents appearing as unreal. It is, I believe, not tautologi-
cal to eliminate those appearing as unreal; and not circular to
eliminate those appearing temporally unrelated to subsistents
appearing as real.

We are attempting to attach a signification to "existence" that
will definitely assign certain subsistents to the realm of unreality.
And we are attempting to attach a signification to "existence"
that will not assign to the realm of unreality fcubsistents which

82

common usage seems to be agreed in calling "real." These are
resolves which we have adopted, although it is not logical con-
siderations which have compelled us to adopt them. We are, I
believe, carrying out these resolves in marking out as unreal the
loss of Cinderella's slipper which appears undatable with respect
to the fall of Constantinople that appears real; and the castles
which some novelist may present to us as being present with re-
spect to allegedly real events, with respect, let us say, to the wars
of Charlemagne, and yet as lacking spatial position with respect
to them.

If, however, we were to mark out as unreal subsistents appear-
ing to lack position with respect to entities alleged to be real
and earlier, or if we were to mark out as unreal subsistents
appearing to lack position with respect to entities alleged to be
real and later, we might be assigning to the realm of unreality
certain subsistents which are commonly called "real." The phase
of Socrates' life in which he was about to drink the cup of hem-
lock appears real, let us suppose; and it appears earlier than my
present writing. I may consider however that at different times
the earth has different positions with respect to the sun and that
whereas, taking the earth as at rest, I am a certain distance from
the place where Socrates was, taking the sun as at rest I am a much
greater distance from the place where the hemlock drinking oc-
curred. I may consider, that is to say, that Socrates^ position may
be projected into the present in various ways and that it is only by
taking one of these present positions as the "same" as Socrates'
that I have position with respect to the hemlock drinking. I may
hold that I have position primarily only with respect to present
entities and that my position with respect to past entities is at the
best ambiguous and is a position at all only in the sense that it is
a position with respect to some present entity held to be in the
"same" place. To hold then that my present writing has no un-
ambiguous position and no direct position with respect to the
hemlock drinking which appears both real and past might be to
have my present writing appear as lacking position with respect to
an entity appearing as real and with respect to which my present
writing appears to be temporally related. And so, whereas we are,
logically speaking, as much at liberty to mark out as unreal the
subsistent appearing as lacking position with respect to an entity

83

that appears real as we are to mark out as unreal the subsistent
appearing as lacking date with respect to an entity that appears
real, we choose in this connection to mark out as unreal merely
that subsistent which appears as lacking position with respect to an
entity that appears real and with respect to which it also appears
present.

Certain subsistents, we say, are unreal that appear with the
characteristic of being temporally unrelated to certain other sub-
sistents. And certain subsistents, we say, are unreal that appear
with the characteristic of being spatially unrelated to certain still
other subsistents. The fall of Constantinople that appears tempo-
rally unrelated to the loss of Cinderella's slipper that appears real
is itself unreal. But the fall of Constantinople that appears tempo-
rally unrelated to the loss of Cinderella's slipper that appears un-
real, this is a different subsistent which, so far as we have yet seen,
may be an existent entity.

It is then certain subsistents appearing with the characteristic
of being temporally unrelated to certain other subsistents that
are unreal; and certain subsistents appearing with the character-
istic of being spatially unrelated to certain still other subsistents.
What is it however to appear temporally or spatially unrelated to
a given entity? The entity that appears as having several dates or
several positions with respect to a given entity does not, we hold,
appear spatially or temporally unrelated to that entity. If Julius
Caesar appears real and the universal 'man* appears both with the
past date with respect to Caesar that is commonly attributed to
Alexander the Great and the future date with respect to Caesar
that is commonly attributed to Napoleon, then the universal 'man'
is not appearing temporally unrelated to an entity that appears
real. To appear temporally unrelated to a given entity is not the
same as appearing with the characteristic of having several dates,
or with the characteristic of having no single date, with respect
to that entity. A universal may be real if it appears as having
several dates and not a single date with respect to a subsistent
that appears real; or if it appears as having several positions and
not a single position with respect to a subsistent that appears
real ajid with respect to which some of its instances appear to be
present. But the universal that appears to have no one date and
no several dates, no one position and no several positions, such a

84

universal is unreal in the sense in which we choose to use the term
"real."

A subsistent is unreal if it appears with the characteristic of
having no date with respect to any entity or with the character-
istic of having no date with respect to an entity that appears real.
Also a subsistent is unreal if it appears with the characteristic of
having no position with respect to any entity or with the character-
istic of having no position with respect to an entity that appears
real and with respect to which it appears present. What appears
nowhere appears with the characteristic of having no position with
respect to any entity. What appears everywhere appears, taken
distributively, with the characteristic of having many positions
and, taken collectively, with the characteristic of having one very
vague position with respect to any entity with respect to which
it appears present. The subsistent appearing to be everywhere,
taken distributively, may, it would seem, be real. The subsistent
appearing to be nowhere is, as we use "existence/* unreal. But
what shall we say with respect to the subsistent appearing to be
everywhere, taken collectively? Shall we say that the cosmos, Space,
Time, etc., appearing as each having a single indefinite date with
respect to each entity that appears real may themselves be real?
Or shall we mark out as unreal not only those entities appearing
as having no date but also those appearing as having only an in-
definite date?

A subsistent, we have seen, may appear with many or with few
characteristics. 16 There are various degrees of accuracy or of vague-
ness with which it may be described and with which it may appear.
Similarly there are degrees of accuracy, we may say, with which a
subsistent may appear dated. With respect to the death of Napo-
leon, the Roman republic, the life of Cicero and the delivery
of the first oration against Catiline all appear earlier. But the de-
livery of the first oration against Catiline appears with a more
definite date with respect to the death of Napoleon than does the
Roman republic. With respect to Napoleon's death, one subsistent
may appear much earlier, another slightly earlier. But one may
also appear as rather definitely dated, another as not so definitely
dated. In determining the signification of our term "existence,"
we choose to make no use of the distinction between the sub-
sistent appearing as much earlier and the subsistent appearing as

85

slightly earlier, or the distinction between the subsistent appear-
ing as earlier, the subsistent appearing as present and the subsist-
ent appearing as later. But in order to eliminate from the world of
reality subsistents that seem to be vague and unmanageable, 17 let
us mark out as unreal certain subsistents appearing with indefinite
dates. A subsistent is unreal, we have said, if it appears with the
characteristic of having no date with respect to any entity or with
the characteristic of having no date with respect to an entity that
appears real. A subsistent is also unreal, let us add, if it appears
with the characteristic of having only a very indefinite date with
respect to an entity that appears real. If, that is to say, the death
of Napoleon appears real and the Cosmos or the time continuum
as a whole appears with the characteristic of having only a very
indefinite date with respect to Napoleon's death, then, as we use
the term "existence," the subsisting Cosmos or the subsisting time
continuum, appearing in this fashion, is unreal.

Similarly with position. A subsistent is unreal, we have said,
if it appears with the characteristic of having no position with re-
spect to any entity or with the characteristic of having no position
with respect to an entity which appears real and with respect to
which it appears present. A subsistent is also unreal, let us add,
if it appears with the characteristic of having only a very indefi-
nite position with respect to an entity which appears real and with
respect to which it appears present. We may again take the Cosmos
as our example, or, better, that instantaneous phase of the Cos-
mos which may be alleged to have been the state of the Cosmos
when Napoleon died. If Napoleon dying at St. Helena appears
real and this state of the Cosmos appears both with the character-
istic of being present with respect to the dying Napoleon and with
the characteristic of having only a very indefinite position with
respect to him, then this state of the Cosmos is appearing with
characteristics which, as we use "existence," mark it as unreal.

An everlasting subsistent, taken collectively, is unreal in so far
as it appears, explicitly or implicitly, with the characteristic of
having only a very indefinite date with respect to an entity that
appears real. An instantaneous but unlimited Space, as dis-
tinguished from limited portions of it, is unreal in so far as it
appears, explicitly or implicitly, with the characteristic of having
only a very indefinite position with respect to an entity which

86

appears real and with respect to which it appears present. An ever-
lasting subsistent or an unlimited Space that appears without
these characteristics is not ruled out of existence by the universal
negative existential propositions which we have thus far adopted.
It is ruled out only in so far as we take up for consideration some
individual subsistent alleged to be everlasting or some individual
subsistent described as an unlimited Space, find it unreal in ac-
cordance witth the universal negative existential proposition
just accepted, and thereupon resolve to list no similar subsistents
among those we call "real." 18

Whac is it, however, to appear with the characteristic of having
only a very indefinite date? The time continuum taken as a
whole appears, we say, at least implicitly, with the characteristic
of having only a very indefinite date with respect to the death of
Napoleon that appears real. The delivery of the first oration
against Catiline appears with a rather definite date, the Roman
republic with a less definite date, with respect to the same entity.
But just how vaguely, it may be asked, must an entity be dated for
it to appear with the characteristic of having only a very indefi-
nite date? In discussing the proposition that whatever appears with
the characteristic of being a definite appearance for no one is un-
real, we made no attempt to mark out any clear line of separation
between the vague and the detailed, between definite appearances
and indefinite appearances. 19 Similarly at this point we shall not
attempt accurately to determine which dates are fairly definite
and which are so indefinite that subsistents appearing to have
them are unreal. The subsisting Cosmos that I am now consider-
ing appears with the characteristic of having only a very indefinite
date with respect to the death of Napoleon that appears real. The
Roman republic that I am now considering appears with the
characteristic of having a not very definite date with respect to
the death of Napoleon that appears real. But neither explicitly
nor implicitly does it appear with the characteristic of having
a date of such indefiniteness that our existential proposition
marks it out as unreal. In short, somewhere between the Cosmos
on the one hand and the Roman republic or the Middle Ages on
the other, there is a line to be drawn between the subsistent appear-
ing with a characteristic that marks it out as unreal and the sub-
sistent with a characteristic that does not mark it out as unreal.

87

Since however we are determining the meaning of "existence"
only in order that we may determine the ontological status of such
entities as are to be considered in this treatise, we shall not at-
tempt to place this line more accurately until occasion, if ever,
requires it.

Whatever explicitly or implicitly appears as self-contradictory
or as not an object or as a definite appearance for no one is un-
real. Whatever explicitly or implicitly appears as lacking any date
or as having no date with respect to an entity that appears real or
as having only a very indefinite date with respect to an entity that
appears real, that too is unreal. And so is the subsistent that ex-
plicitly or implicitly appears as lacking any position; the subsistent
that explicitly or implicitly appears as having no position with
respect to an entity which appears real and with respect to which
it appears present; and the subsistent that explicitly or implicitly
appears as having only a very indefinite position with respect to
an entity which appears real and with respect to which it appears
present. These are propositions which partially determine the
meaning being assigned our term "existence." Together they
assign to the realm of the non-existent many subsistents and they
attribute to 'existence' the characteristic of freedom from self-
contradiction, freedom from utter non-spatiality, freedom from
this, and freedom from that. Our studies in the preceding chapter
left us with the resolve to examine and to utilize in our proposi-
tions explaining "existence** the notions of self-contradiction, of
time, and of space. The propositions with which this paragraph
begins are the result.

We already know that the propositions thus far accepted will
not suffice to give our term "existence" a precise meaning. We
already know that in the end our universal negative existential
propositions will have to be supplemented by singular or particu-
lar existential propositions, both affirmative and negative. But
before we resort to singular existential propositions, let us at-
tempt to develop additional universal negative propositions. Leav-
ing self-contradiction and space and time behind, let us attempt to
3&ark out some additional subsisting entities as unreal. The un-
limited space which appears as having only an indefinite position
with respect to the dying Napoleon who appears real and with
respect to whom thi$ unlimited space appears present, the eternal

8S

verity which appears utterly timeless and the square circle which
appears self-contradictory, these subsistents are already marked
out as unreal. But before we resort to individual existential prop-
ositions, let us attempt to eliminate the phlogiston that does not
appear self-contradictory, the present King of France who does
not appear to lack position with respect to me, and the sleeping
Barbarossa who does not appear undated.

When I think of the King of England I seem to have a feeling
of acceptance or assent or belief. No feeling of hesitation or of
disbelief seems to intervene. But when I press my eyeball and seem
to see a second rose in the vase on my desk, or when I try to im-
agine a man walking upside down on my ceiling, I may become
aware of a feeling of hesitation, a feeling of dissent or rejection
or disbelief. The King of England that I am now considering
appears with the characteristic of being in some sense an object.
And it appears with the characteristic of being an object such that
the apparent awareness of it is generally accompanied by a feeling
of belief. The man on my ceiling that I am now considering also
appears with the characteristic of being in some sense an object.
But it appears with the characteristic of being an object such that
the apparent awareness of it is generally accompanied by a feel-
ing of disbelief. The subsisting man on my ceiling and the sub-
sisting second rose in the vase on my desk, unlike the King of Eng-
land whom I am considering, appear with the characteristic of
being generally discredited. They are therefore, let us say, unreal.
Let us lay down the universal negative existential proposition that
whatever explicitly or implicitly appears as generally discredited
is unreal. And when a subsistent, as we develop it, takes on the
characteristic of appearing generally discredited, let us resolve to
list as real no subsistent which, except for its development, is in-
distinguishable from it.

The man on my ceiling, the second rose in the vase on my desk,
phlogiston, and the sleeping Barbarossa, all of these subsistents,
as we develop them, implicitly appear with the characteristic of
being generally discredited. These subsistents are therefore un-
real. And no other subsisting men on my ceiling, phlogistons, or
sleeping Barbarossas will be listed among the entities we are to
enumerate as real. Some subsisting King of England does, we may
suppose, develop the appearance of being generally discredited,

89

and is likewise unreal. But since the subsisting King of England
which we are considering does not develop this appearance, this
subsisting King of England, and other subsisting Kings of England
which, like it, do not develop the appearance of being generally
discredited, may very well be real.

No subsistent is real which appears with the characteristic of
being generally discredited, with the characteristic of lacking all
position, or with any one of various other characteristics. Repre-
senting that which appears with the characteristic of being self-
contradictory by the letter A, that which appears with the char-
acteristic of being generally discredited by the letter J, and so
on, we may say that no subsisting A's or B's or ... or J's are real.
To exist is at the least to be free from A-ness and B-ness . . . and
J-ness. But the subsistents that do not appears as A's or B's or as
J's, the subsistents that neither explicitly nor implicitly appear
with the characteristic of being self-contradictory or with the
characteristic of being generally discredited are some of them
real and some of them unreal. To exist is not merely to be free
from A-ness, from B-ness, from . . . and from J-ness. To exist is in
addition to be enumerated as real in one of our individual affir-
mative existential propositions. Some of the subsistents which do
not appear as A's or B's or ... or J's we have agreed not so to
enumerate. We have agreed not to enumerate as real any sub-
sistent, which, except for its development, is indistinguishable
from one which, as we develop it, implicitly appears as an A or
a B or ... or a J. Since the subsisting phlogiston which we are
now considering appears with the characteristic of being generally
discredited, we resolve not to enumerate as real any subsisting
phlogiston. But 'the fiftieth President of the United States, a So-
cialist named Jones' appears neither self-contradictory nor gen-
erally discredited; and, considering this subsistent as an individ-
ual subsistent, we have no rule to guide us and to determine us
to list this subsistent as real rather than as unreal. It is not all
subsistents not appearing as self-contradictory, etc., which are
real; not even all subsistents not appearing to resemble one which,
as we develop it, appears as self-contradictory. Reality is limited
to those subsistents really free from self-contradictoriness. And
those entities that are really free from self-contradictoriness can
be further described only by enumerating some of them.

90

No subsistent is real which explicitly or implicitly appears as an
A, a B, or ... or a J. With respect to the subsistent which neither
explicitly nor implicitly appears as an A, a B, or ... or a J, it is
real if listed below as X if X 2 , or ... or X n , unreal if listed below
as YI, Y 2 , or ... or Y a . To exist is to appear free from A-ness, B-
ness, . . . J~ness and to be enumerated as an X. To be unreal is to
appear explicitly or implicitly as an A, a B, or ... or a J; or to be
enumerated as a Y. In so far as a subsistent does not appear as an
A or ... or a J and is not enumerated as an X or as a Y, its existen-
tial status is left undetermined and the significations of our terms
"existence" and "non-existence" are left with some vagueness. It
will be found however that our universal negative existential
propositions: "No subsistent appearing as an A exists," etc., taken
in conjunction with our existential propositions: "Xi, etc. exists,"
"Yi, etc. does not exist," determine with reasonable precision the
characteristics of 'existence* and 'non-existence* and will enable
us to determine the existential status of most of the subsistents
presented to us in the course of this treatise. When we have with
a similar precision determined what it is to be true, we shall, I
believe, be in a position to investigate various problems of con-
cern to the metaphysician with a well-founded hope of being able
to determine which of the entities discussed in these problems are
real, and with a well-founded hope of being able to determine
which of the propositions in which attitudes towards these prob-
lems may be expressed are true.

And so, before we turn from the distinction between the real
and the unreal to the distinction between the true and the false,
we have only to give the following recapitulation of the character-
istics for which A, B, etc. stand and the following lists of X's and
Y's.

A Self-contradictory.

B In no sense an object of consciousness.

C A definite appearance for no subject.

D Lacking all date.

E Having no date with respect to an entity that
appears real.

F Having only a very indefinite date with respect
to an entity that appears real.

G Lacking all position.

91

H Having no position with respect to an entity
which appears real and with respect to which it
appears present.

I Having only a very indefinite position with re-
spect to an entity which appears real and with
respect to which it appears present.
J Generally discredited.

APPENDIX

A List of Certain Subsistents A List of Certain Subsistents

which, appearing neither ex- which, even when they appear

plicitly nor implicitly as self- neither explicitly nor iniplic-

contradictory, undated, etc., are itly as self-contradictory, un-

real. dated, etc., are nevertheless

Xi unreal.

X 2 - Y x -

Xa- Y 2 -

X 4 - Y 8 -

(I ask the reader to assume that there have just been enumerated
each of the entities that will later be referred to as having been
listed in this appendix)

Summary

We explain our term "existence" fairly adequately through
singular existential propositions and the following universal prop-
ositions:

1. No entity is real which is presented as self-contradictory.

2. No entity is real which is presented as in no sense an object
of consciousness.

3. No entity is real which is presented as a definite appearance
for no subject.

4. No entity is real which is presented as lacking all date.

92

5. No entity is real which is presented as having no date with
respect to an entity that appears real.

6. No entity is real which is presented as having only a very
indefinite date with respect to an entity that appears real.

7. No entity is real which is presented as lacking all position.

8. No entity is real which is presented as having no position
with respect to an entity which appears real and contempo-
raneous with it.

9. No entity is real which is presented as having only a very
indefinite position with respect to an entity which appears
real and contemporaneous with it.

10. No entity is real which is presented as generally discredited.

Propositions 1, 4 and 7 seem to give our term "existence" a
meaning in accord with common usage. But they leave the exis-
tential status of various subsistents undetermined to a greater
extent than is desirable. By considering 1, 4 and 7 in turn, we
are led to choose to supplement them with 2 and then with 3,
with 5 and 6, and with 8 and 9. Proposition 10 is added in an
effort to enlarge the content of the world of non-existing entities
in our sense of "existence" and to reduce the reliance that has to
be placed on individual existential propositions.

The discussion of proposition 2 is probably of greatest general
interest. The position taken is that the entity in no sense an
object of consciousness appears with the characteristic of being
implicitly self-contradictory and hence is unreal.

TOWARDS DETERMINING THE MEANING OF "TRUTH"

At this point in our story the meaning of our term "existence"
has been more or less determined. At this point we have agreed
that certain entities YI, Y 2 , Y 8 , even when appearing neither ex-
plicitly nor implicitly as self-contradictory, as undated, etc., are
unreal. And we have agreed that certain entities Xi, X 2 , X 3 , when
appearing neither explicitly nor implicitly as self-contradictory or
as undated, etc., are real in the sense in which we are using the
term "reality." Now, among the entities which are real in our sense
of "reality/' among the entities Xi, X 2 , X 3 , are certain words. The
word "Socrates," occurring in the copy of Plato's "Republic" that
is in my library and appearing neither explicitly nor implicitly as
self-contradictory, etc., is a real entity. And the word "Ivanhoe"
appearing with the characteristic of being in my copy of Scott's
novel is likewise a real entity.

Each entity that can be discussed is a subsistent. Some of these
subsistents, as, for example, the words "Socrates" and "Ivanhoe"
to which we have just pointed, are real entities. And^omeVtf these
subsistents are entities which, in our sense of the term "existence,"
are unreal entities. Without stopping to enquire whether they are
real or unreal, let us note that within the world of subsistents there
appear the entities: 'Socrates, the Athenian philosopher' and 'Ivan-
hoe, the medieval knight/ Thus we seem to have before us the sub-
sistent 'Socrates, the Athenian philosopher' whose ontological
status we may for the present leave undetermined, and an instance
of the word "Socrates" which is real; the subsistent 'Ivanhoe, the
medieval knight', whose ontological status we may for the present
leave undetermined, and an instance of the word "Ivanhoe" which

94

is real. Obviously there is a certain connection or a certain pseudo-
connection between the real word "Socrates" and the subsistent:
'Socrates, the Athenian philosopher/ between the real word
"Ivanhoe" and the subsistent: Ivanhoe, the medieval knight/
To put it briefly, the word "Socrates" represents or intends to.
represent the Athenian philosopher and the word "Ivanhoe"
represents or intends to represent the medieval knight. It would
carry us too far afield to attempt at this stage in our exposition to
analyze what this representation or this intention to represent
consists in. 1 Let us note simply that certain words are real and that
by virtue of their being words they seem to intend, to point to, or
to represent, certain other subsistents which may or may not be
real.

The wc5rd "Socrates," occurring in my copy of Plato's "Repub-
lic" is real; and the word "Ivanhoe" occurring in my copy of
Scott's novel is real. In a similar fashion the words "man" and N
"large" appearing with the characteristic of occurring on this page
are each of them subsistents which are'real. Wherea^, however,
the words "Socrates" and "Ivanhoe" represent or intffod to repre-
sent Subsistents which, if real, are individual substances, "man"
and "large^ represent or intend to represent subsistents which,
if real, arean the one case a universal substance^and in the other
case a universal quality. Nonetheless, the instances of "man" and
"large" to which reference has just been made are words which
are real, words which are to be kept in view along with "Socrates"
and "Ivanhoe." Indeed, we may enlarge the domain of real en-
tities to which we are attending by pointing to the words: "walk-
ing quickly down the street" and to the words: "President of the
United States." Each of these word groups subsisting with the
characteristic of occurring on this page is real and each of them
represents or intends to represent a subsistent which if real is
a quality or substance outside of this page. There is then one
instance of the word "Socrates" which is real, one instance of
"Ivanhoe," one instance of "man/' one instance of "large," one
instance of "walking quickly down the street" and one instance of
"President of the United States." Without further ado we may
say at once that many words and word groups are real, and that
many sentences are real. We may agree, for example, that each
of the preceding sentences in your copy of this book, appearing

95

neither explicitly nor implicitly as undated, etc. is a real sen-
tence. And we may agree that each of these sentences contains
words, word groups and phrases which severally represent, or
intend to represent, subsistents which may or may not be real.

We are working in this chapter towards the determination of
the significations to be assigned the terms "truth" and "falsity."
And we have come to have before us for our consideration various
real sentences, as, for example, the preceding sentences in your
copy of this book, in order that we may apply the distinction be-
tween the true and the false somewhere within the realm of real
sentences. It may be well therefore at this point to note that the
adjectives "true" and "false" as they occur in common speech
are by no means exclusively associated with such entities as sent-
ences. We commonly speak of true sentences, true propositions,
true judgments, true pictures, true ideas, true beliefs and true
friends. And so we ask ourselves whether, when we attempt to
determine the signification of "truth" by applying the distinction
between the true and the false somewhere within the realm of real
sentences, we are maintaining the contact with ordinary usage
that we wish to maintain. In so far as truth is commonly predi-
cated of such entities as propositions and judgments, we need
not be disturbed. For our concern with words, terms and sentences
will guide our attention to propositions and to judgments and
will enable us to point out certain entities to be called true prop-
ositions, certain entities to be called true judgments and certain
entities to be called false propositions. But the signification of
"truth" which we are developing will not enable us to apply the
distinction between the true and the false to friends or to pictures,
to beliefs or to ideas.

What we commonly call a true friend is, I suppose, a devoted
friend, a real friend; what we commonly call a false friend an
apparent friend who is not a friend. The distinction between the
real on the one hand and the unreal on the other is, it appears,
involved in the distinction between the so-called true friend and
the so-called false friend. Let us not use "true" and "false" to
point to the very distinction to which the contrast between the
real and the unreal points. And so let us not determine the sig-
nification of our term "truth" in such a way that there will be
true friends and false friends.

96

Just as it is the distinction between the real and the unreal
rather than the distinction between the true and the false that,
we shall say, applies to friends, so it is the distinction between
knowledge and error rather than the distinction between the
true and the false that applies, in our terminology, to such psy-
chological or epistemological entities as may be called ideas,
opinions, or beliefs. In a later chapter we shall deal at some length
with the distinction between knowledge and error. 2 And so we
are not permanently neglecting this important distinction when
we leave beliefs, ideas and opinions out of consideration in con-
cerning ourselves with the notion of truth and with the distinction
between the true and the false.

The words, word groups and phrases that occur in sentences
represent, as we have seen, 3 or intend to represent, subsistent
entities other than themselves. And the truth or falsity of these
sentences depends, we shall hold, upon the ontological status of
these subsistent entities that are intended to be represented. There
is a sense then in which sentences look beyond themselves and in
which their truth or falsity depends upon their correspondence
with entities beyond themselves. What more natural, then, than
that pictures should be called true or false and that their truth
or falsity should be held to depend upon their correspondence or
lack of correspondence with the objects they intend to portray?
Despite the similarity between words and pictures, however, I
believe we are not violating the ordinary usage of words in dis-
tinguishing between words and pictures, and in making the
distinction between truth and falsity one which does not apply
to pictures but, rather, applies exclusively to words and their
derivatives, to sentences, propositions and judgments.

It is within the realm of real sentences that we shall first at-
tempt to apply. the distinction between the true and the false.
And yet it is not each real sentence that we shall hold is either
true or false. There is the real sentence: "Where are you going?"
and the real sentence: "Shut the door"; but "Where are you go-
ing?" is not true and "Shut the door" not false. It is: "You seem
to be going some place" that may be true, "I desire you to shut
the door" that may be false. The distinction between the true
and the false, in short, is to be applied only to real sentences that
are declarative, not to real sentences that are interrogations or

97

commands.

Just as "Shut the door" is neither true nor false, so it is, as we
shall use the terms "truth" and "falsity," with the sentence:
"Take as your alleged object a subsisting Socrates." A subsisting
Socrates is presented as a datum; but the mere presentation in-
volves no assertion to be concurred in or denied. Similarly with
the sentence: 'Socrates is (i.e., appears as) a subsistent." "Socrates
subsists" expresses no real assertion, adds nothing to the datum
that "Subsisting Socrates" seems to present. Nor have we arrived
at a real assertion when the subsisting Socrates as a subsistent is
said to appear with various characteristics. For "Socrates appears
or subsists as a Greek and as a philosopher" still merely presents
an alleged datum and expresses no attitude with respect to this
datum that can be concurred in or denied. It is, one might say,
synonymous with: "Let Socrates be a Greek philosopher." Only
declarative sentences, we have said, are true or false. But sentences
of the type: "X subsists" or "X does not subsist" or "X subsists
with characteristic A," although declarative in form, are rather to
be classed with interrogations and commands than with the
declarative sentences to which we shall apply the distinction be-
tween the true and the false.

The only sentences that we shall call true or false are declara-
tive sentences, declarative sentences which are real and which
contain words, word groups or phrases which severally represent
or intend to represent subsistent entities. Among these declara-
tive sentences which we have before us, however, there are some
which do not conform to the grammar of the language in which
they are expressed. The English sentence "Green is or" is ungram-
matical and so is "We am here." It is desirable that we put such
sentences aside in working towards the determination of the
signification of "truth"; for without such an elimination we have
the task of applying the distinction between the true and the
false to many sentences which are incomprehensible or ambigu-
ous. The rules of grammar are many and vary from language to
language. They are however rather definite and are fairly gen-
erally understood. With respect to any given sentence it is usu-
ally obvious that it does, or that it does not, conform to the
grammar of the language in which it is expressed. It is generally
agreed, for example, that each English declarative sentence must

98

have a verb and a subject. And so it is clear that a given sentence
which contains no subject is a sentence to which, in the sense in
which we are using the terms "truth" and "falsity," the distinc-
tion between the true and the false does not apply. It may like-
wise be said to be a rule of English grammar that the subject
must be a noun or pronoun. And so, if "Green is a color" is to
be held to be a sentence to which the distinction between the
true and the false applies, the word "green" as it occurs in this
sentence must be held to be a noun. Let us hold that in our
sentence "Green is a color" the word "green" is indeed a noun.
Let us hold that this instance of the word "green" represents a
substance whose important and outstanding quality is its green-
ness. Let us, consequently, agree to use "truth" and "falsity" in
such a manner that the distinction between the true and the false
applies to our sentence: "Green is a color." It is only with respect
to some few sentences"Green is a color" is one of them that
their conformity or lack of conformity to the rules of grammar
is disputable. And so it is only a few sentences and a few grammat-
ical rules that we need discuss in order to make clear which sen-
tences we are eliminating from further consideration in working
towards the determination of the signification of "truth."

The subject of a grammatical English declarative sentence must
be a noun or a pronoun. Our sentence: "Green is a color" is
grammatically correct in that "green" is in this instance a noun.
Our sentence: "White is always serviceable" is grammatically
correct in that "white" in this instance modifies some such noun
as "clothing" which has been elided. Not only, however, must the
subject of a grammatical English declarative sentence be a noun
or pronoun; with certain predicates, abstract nouns are ruled out
as possible subjects of grammatical English declarative sentences.
"Brightness is fire" is not grammatically correct. It is a sentence
to which, as we employ "truth," the distinction between truth and
falsity does not apply. There is, to be sure, the grammatically
correct sentence: "Brightness is cheerful" and the grammatically
correct sentence: "Charity is godliness." 4 But a sentence whose
subject-term is an abstract noun is never grammatically correct,
we hold, when this subject-term is copulated with a concrete noun
or when the predicate-term is a cognate verb. "Brightness is fire"
is, we hold, ungrammatical; and so is "Motion moves." "Bright-

99

ness is fire" and "Motion moves" are both sentences, we hold,
that lie outside the distinction between the true and the false.
They are sentences to be eliminated from our further considera-
tion along with "Green is or" and "We am here" in so far as we
are working towards the determination of the signification of
"truth."

At this point we have before us sentences which are real, sent-
ences containing words, word-groups and phrases which severally
represent, or intend to represent, subsistents, sentences which
are declarative, which do not merely predicate subsistence, and
which conform to the grammatical rules of the language in which
they are expressed. These sentences which we have before us are,
let us say, propositions. And so we may say that sentences which
are not propositions are neither true nor false; and we may say
that, with respect to sentences, it is within the realm of proposi-
tions that the distinction between the true and the false is to
be applied.

Among the propositions which we have before us, let us pick
out for special consideration those sentences of ours which are
singular affirmative existential propositions. There is, for exam-
ple, the proposition: "Socrates, the Athenian philosopher, exists"
and there is the proposition: "Ivanhoe, the medieval knight,
exists." It is with respect to propositions having this form that
we shall find it simplest to apply the distinction between the
true and the false and thus partially to explain our term "truth."
Our sentence: "Socrates exists" is a true proposition, we shall say,
if, and only if, in our sense of "existence," the entity exists which
the word "Socrates" as it occurs in this sentence intends to repre-
sent. And our sentence: "Ivanhoe exists" is a true proposition
as we use the term "truth" if, and only if, in our sense of "exist-
ence," the entity exists which the word "Ivanhoe" as it occurs in
this sentence intends to represent. Since Socrates, the Athenian
philosopher, appearing neither explicitly nor implicitly as self-
contradictory or as undated, etc., is real in the sense in which
we are using the term "reality," the real proposition: "Socrates
exists" which occurs on this page is true in the sense in which
we are using the term "truth." And since Ivanhoe the medieval
knight, even when he appears neither explicitly nor implicitly as
contradictory or as undated, is unreal, our sentence: "Ivanhoe

100

exists" is in our terminology an untrue or false proposition. We
have thus certain real propositions definitely marked out as true
in our sense of "truth" and certain real propositions definitely
marked out as false in our sense of "falsity." We have thus made
a beginning in determining the meaning of our term "truth."

It is a simple matter to go on to determine the truth or falsity
of our negative singular existential propositions. Our sentence:
"Socrates does not exist" is false, let us say, if the entity exists
that the word "Socrates" as used in this sentence intends to repre-
sent; true if this entity does not exist. We are in a position, it
follows, to determine the truth or falsity of any singular existen-
tial proposition of ours. If the individual exists that our word X
intends to represent, "X exists" is true and "X does not exist"
false. And if the individual that our word X intends to represent
does not exist, "X exists" is false and "X does not exist" true. In
their application to singular existential propositions of ours, the
significations which we are assigning to the terms "truth" and
"falsity" have thus been determined.

To the extent to which we have thus far determined the signi-
fications of "truth" and "falsity," we have done so by referring
back to the distinction between the real and the unreal. Roughly
speaking, we have made the distinction between the real and the
unreal prior to the distinction between the true and the false;
and we have explained "truth" in terms of "reality." There are
those however who would object to the treatment of reality and
truth in this order. Truth, according to Bertrand Russell, 5 is
prior to reality, not reality prior to truth. When we discuss reality,
we do so'by means of propositions. And our discussion of reality
has validity, it is held, only in so far as our propositions referring
to reality are true. "When I say: this paper exists, I must," says
Moore, 6 "require that this proposition be true." If I am to make
valid remarks about reality, I must, it is held, already know what
constitutes validity, I must already understand the term "truth."
Do we however avoid such objections when we begin with a dis-
cussion of truth and proceed thence to a discussion of reality? The
distinction between truth and falsity, after it has once been put
before us, applies to all propositions including those in which
"reality" is explained. Similarly, however, the distinction between
the real and the unreal, after it has once been put before us,

101

applies to all entities including the sentences in which the mean-
ing of "truth" is discussed. Unless these sentences are real,
they can neither be true nor determine for us the meaning
of "truth." A discussion of truth presupposes the reality of the
sentences in which truth is discussed just as a discussion of reality
presupposes the truth of the propositions in which reality is dis-
cusssed. In a sense, then, truth presupposes reality; and reality
presupposes truth. Wherever we begin we find ourselves in a circle
rather than at the beginning of a linear chain. Indeed this circle
is even narrower than we have yet indicated. Not only does truth
in a sense presuppose reality, and reality truth; but reality in a
sense presupposes reality and truth presupposes truth. Just as the
sentences are real in which we determine the meaning of "truth,"
so the sentences are real in which we determine the meaning
of "reality." And just as some of the propositions are true
in which we discuss reality, so some of the propositions are true
in which we discuss truth. In a sense we can not discuss reality
unless we make use of real sentences and we can not make valid
propositions referring to truth unless these propositions are them-
selves valid and true, unless, it may be said, we already know what
validity and truth are.

It would be absurd to hold that such observations prevent us
from ever properly discussing either truth or reality. When we
attend to a concept with the purpose of discussing, analyzing and
defining it, we are not always introducing a term which has no
relevance to anything that has gone before. Rather we clarify a
concept so that as a result of the discussion the application of the
concept will be clear both with respect to what has preceded and
with respect to what is to follow. The sentences in the first chap-
ter of your copy of this book are real, but we did not know them
to be real until we had determined the signification of "reality."
The propositions in which we determine the significations of
"truth" and "falsity" are true; but we do not know them to be
true until we shall have determined the signification of "truth."
Without knowing a given sentence to be real or true we can
gather from it the signification that is being assigned "truth" or
"reality." And so a valid discussion of either truth or reality takes
place through the medium of propositions which are true and
of entities which are real, although these propositions are not

102

revealed as true and these entities are not revealed as real until the
discussion has been completed. Obviously the distinction between
the real and the unreal applies to all entities and, limiting our
attention to propositions occurring in this treatise, the distinction
between the true and the false applies, we shall hold, 7 to all
propositions. If this is the case, then we can not discuss either
'truth' or 'reality* by means of propositions without making use
of entities to which these distinctions which are in the course of
being elucidated already apply. But we can, we hold, and in many
cases must, analyze and define concepts whose application is not
limited to what is to follow. With both 'truth' and 'reality/ this
is the case; and it is as much the case with the one as with the
other. In exposition, we hold, we are at liberty to begin with
either concept and then to proceed to the other. Our difficulties
are just as great, or, we should hold, just as unimportant, whether
we begin with reality and proceed to a discussion of truth or
whether we begin with truth and proceed to a discussion of
reality.

It has been our decision to begin with a discussion of "reality "
and to explain "truth" in terms of 'reality/ If the argument of the
preceding paragraph is sound, there is no logical reason to com-
pel us to alter this decision and to begin instead with a discus-
sion of "truth/' But, we may ask, are there not motives of ex-
pediency that may determine us to alter our decision? Before we
proceed to explain "truth" in terms of reality, will it not be well
for us to consider the possibility of explaining "reality" in terms
of 'truth' or at least of explaining "truth" without referring back
to a previous discussion of 'reality'? To explain "truth" in terms
of reality is not logically unsound, but it may be inexpedient.
And explaining "truth" without referring back to a previous dis-
cussion of reality, whereas it is not logically necessary, may make
for greater simplicity in exposition.

There are those, we have seen, 8 who hold that truth is prior to
reality. A proposition or judgment is true or false, it may be said,
not according as the entities intended to be represented by its
terms are real or unreal, but rather according as it has or lacks
intrinsic marks which directly determine it to be true. Certain
judgments, it may be said, come before our minds with an insist-
ence and a claim that forces us to recognize them as true; and

103

certain judgments come before our minds, it may be said, with a
weakness and a logical unattractiveness that forces us to reject
them as false. Thus "two and two are four," it may be said, is
true, not because of anything concerning the ontological status
of 'two' and 'four/ but because "two and two are four" has an
intrinsic vitality and claim which we are bound to recognize.
"The recognition of the claim of a judgment/' says Rickert, 9
"constitutes its truth." In no other way, he holds, is truth to be
defined. For, he continues, "truth can only be defined as the
peculiar value that judgments have." There is here an attempt to
discuss truth without reference to reality. And since we may
begin with either concept, since, moreover, we are at liberty to
assign to terms whatever significations we please, there is no logi-
cal objection that can be raised against this procedure. We may
introduce the term "truth" without referring to a previous dis-
cussion of "reality." And we may subsequently introduce the term
"reality" by saying that an entity is real when the judgment that it
is real has the validity, the claim upon us, that characterizes true
judgments. But whereas there are no logical objections that can
be raised against this procedure, we may question whether a pro-
cedure of this sort explains with any success either "truth" or
"reality." And we may question whether a procedure of this sort
assists us in any way in applying the distinction between the true
and the false to individual propositions and judgments. If we are
in doubt as to the truth of an instance of "Ivanhoe exists," it will
not help us to be told that "Ivanhoe exists" is true if it has a
claim upon us. For, we may ask with James, 10 "What do you mean
by 'claim' here?" But it will help us to be told that our sentence
"Ivanhoe exists" is true if Ivanhoe the medieval knight is a real
entity; and then to be referred back to the rather full discussion
of reality in chapter three.

Just as it may be said that a judgment is true if intrinsically
it has a claim upon us, so it may be said that a judgment is true
if intrinsically it is clear and distinct. "I am certain that I am a
thing which thinks," says Descartes; 11 "but do I not then likewise
know what is requisite to render me certain of a truth? Certainly
in this first knowledge there is nothing that assures me of its
truth, excepting the clear and distinct perception of that which
I state." This clear and distinct perception would not "assure

104

me that what I say is true, if it could ever happen that a thing
which I conceived so clearly and distinctly could be false; and
accordingly it seems to me that already I can establish as a general
rule that all things which I perceive very clearly and very distinctly
are true." In this passage, to be sure, Descartes is not holding
that only those judgments are true which are clear and distinct.
But just as it may be held that a true judgment is one which has
validity and a logical claim upon us, so it may be held that a
true judgment is one which is clear and distinct. With either
explanation of "truth/' however, we have little to guide us in
applying the distinction between the true and the false to indi-
vidual propositions and judgments. To make either explanation
serviceable, there would be required a rather complete account
in the one case of 'claim' or Validity* and in the other case of
'clear and distinct/ There would be required indeed something
of an enumeration of the propositions or judgments that have a
claim or are clear and distinct. And so we should explain "truth"
prior to "reality" only by putting something analogous to the
appendix to our third chapter into our explanation of "truth"
instead of into our explanation of "reality."

Let us then proceed in the direction in which we have started.
Let us work towards determining the meaning of "truth" by
continuing to refer back to our explanation of "reality." If
the individual exists that our term X intends to represent, then
our real sentence: "X exists/' let us continue to say, 12 is true
and our real sentence: "X does not exist" false. And if the indi-
vidual that our term X intends to represent does not exist, then
our real sentence: "X exists," let us continue to say, is false and
our real sentence: "X does not exist" true. We are proceeding
thus from reality to truth, from reality to truth in so far as truth
is a characteristic of the real sentences that we call propositions.

But although it may be acceptable to proceed from reality to
truth rather than vice versa, it may seem strange that we leap at
one bound from reality to that aspect of the notion of truth in
which truth is considered a characteristic of the sentences that we
call propositions. "Just one moment!", we may be told; "Truth is
primarily a characteristic of judgments. It has application to the
sentences that you call propositions, sentences occurring on this
page and on that page, only secondarily, only in so far as these

105

sentences represent or express or symbolize true judgments."
Indeed there are those who hold that the distinction between the
true and the false is never properly applied to sentences occurring
on this page and on that page, that it applies only to judgments
which are outside of the printed or spoken word. It is from this
point of view that Leibniz finds fault with Locke's discussion of
truth. "What I find least to my taste in your definition of truth,"
says Leibniz, 13 "is that you seek truth in words. Thus the same
sense expressed in Latin, German, English, French, will not be
the same truth. . . . We shall then have also literal truths which
may be distinguished as truths upon paper or parchment, of
ordinary black ink or of printer's ink." Is there however any
reductio ad absurdam in this conclusion? Sentences exist that are
on this page or on that page. Some of them are of ordinary black
ink and some of them are of printer's ink; some of them in Latin
and some in French. Among these real sentences which are here
and there, of various kinds of ink and in various languages, there
are some which, in the sense in which we are using the terms
"truth" and "falsity," are true and some which are false. We are
at liberty to determine the meaning of "truth" in such a manner
that the distinction between the true and the false applies to cer-
tain real sentences. And we are exercising this liberty in a manner
not altogether at variance with common usage when we call
certain sentences propositions and call some propositions true and
some false.

It is obvious however that certain sentences which are true in
our sense of "truth" have a common point of reference. There is
the sentence: "Socrates exists" which occurs on one page of my
copy of this book; and there is the sentence: "Socrates exists"
which occurs on a corresponding page in your copy of this book.
There is the sentence: "Socrates exists" which occurs on another
page of my copy of this book; there is the sentence: "Socrates
exists" which occurs in my manuscript; and there is the sentence;
"Socrates est" which occurs, let us suppose, in some Latin manu-
script. Each of these sentences is true and each of them, we may
suppose, refers to the same fact. Ought we not then seek truth
in this fact, in this common point of reference? In concentrating
our attention upon sentences made by ink or by pencil, we are
dealing, it would seem, with mere shadows, with entities whose

106

truth or falsity is merely a reflection of the truth or falsity of
some objective situation outside these sentences.

What however is the fact which several sentences, each reading:
"Socrates exists," have as their common point of reference? When
I write the word "Socrates," there is something in my mind. And
so the word "Socrates" is somehow related to some act of cogni-
tion or to some idea of mine. At the same time, however, the
word "Socrates" is somehow related, directly or indirectly, to an
objective subsistent which is alleged to be outside of me and
outside of the word "Socrates." As we have seen, 1 * the word
"Socrates" represents or intends to represent Socrates the Athenian
philosopher who scorned the Sophists and died in jail. In the
case of the word "Socrates" there is thus what we may roughly
contrast as a subjective reference and an objective reference.
When we turn from the word "Socrates" to the sentence: "Soc-
rates exists," there is, it would seem, a similar dual reference.
There is on the one hand an act of judgment, or an asserting,
taking place in my mind; or the copulation of mental ideas that
we may call a mental judgment. And, on the other hand, there
may be some objective fact, some situation involving Socrates
himself, to which the sentence: "Socrates exists" may be said to
refer. Now the former of these entities, the act of judgment taking
place in my mind, or the copulation of mental ideas that we may
call a mental judgment, belongs within the realm of psychological
or epistemological entities to which we have agreed to apply the
distinction between knowledge and error rather than the distinc-
tion between the true and the false. 15 It may be a common refer-
ence to some such mental judgment that links together an in-
stance of the sentence: "Socrates exists" and an instance of the
sentence "Socrates est." Nevertheless let us turn our attention
to the investigation of the possibility of these two sentences being
linked together, not by a common subjective reference, but by a
common objective reference, by a common reference, that is to
say, to some objective situation involving Socrates himself.

What, however, is the objective fact which we may call a judg-
ment and to which we may say that the sentence: "Socrates exists"
intends to refer? It is not the substance Socrates himself, for this
substance the simple word "Socrates" represents or intends to
represent. Nor, we shall say, is the objective fact which might be

107

called a judgment some non-temporal fact having its habitat in
a world of objective but disembodied entities. For we choose to
deal primarily with real entities; and, in the sense in which we
are using the term "reality," any entity that appears as utterly
non-spatial is unreal. 16 The entity related to the sentence "Soc-
rates exists'* that we shall call a judgment or fact is some situation
involving Socrates himself; and yet it is not the substance Socrates.
It is, let us say, the existence of Socrates; that is to say, existence
appearing as an alleged quality of the subsistent Socrates. When
I utter the word "Socrates" or the word "Ivanhoe," I am appar-
ently making no assertion. My expression intends to refer to
a subsistent which may or may not be real. But if I say "Socrates
exists/ 1 there seems to be something that I am asserting, namely,
the existence of Socrates. If then we call such entities as 'the
existence of Socrates' judgments, our use of the word "judgment"
will permit us to say that a judgment is something that may
be asserted. Let us then call the existence of Socrates a fact or
judgment; and, since Socrates exists, let us furthermore call it
a true judgment. In the 'existence of Socrates/ we hold, we have
an instance of a judgment which is an objective situation, a
situation to which various sentences each reading: "Socrates
exists" may be said to refer. Not only, however, is the existence
of Socrates an objective judgment to which various propositions
each reading: "Socrates exists" may be said to refer. It is likewise
a true judgment; and its truth may be thought of as determining
the truth of the propositions which refer to it. Truth may be
thought of, in short, as belonging primarily to the judgment:
'the existence of Socrates' and as belonging secondarily and by
reflection, as it were, to the proposition: "Socrates exists" which
occurs on this page and to the proposition: "Socrates est" which
occurs in some Latin manuscript.

Socrates the Athenian philosopher is a subsistent. Appearing
neither as self-contradictory nor as undated, etc., this subsistent
is real. Likewise the quality of being an Athenian is a subsistent,
a subsistent which, appearing as a quality of Socrates, is real.
Similarly with the quality of existence, appearing as a quality of
Socrates. The existence of Socrates is a real subsistent, or, what
is the same thing, the true judgment 'the existence of Socrates'
is a real entity. How is it however with respect to non-existence

108

subsisting as a quality of Socrates? If Socrates appears as unreal,
both this subsistent and its alleged quality of non-existence are
unreal. Even the Socrates that appears both as real and as unreal
is unreal; and the non-existence of Socrates alleged to inhere in
it unreal. For the subsistent which I am considering appears as
self-contradictory. 17 There is, we conclude, no real objective
situation different from, but analogous to, the existence of Soc-
rates to which the proposition: "Socrates does not exist" refers.
There is no real non-existence of Socrates that might be called a
judgment. And so, whereas we have been successful in identifying
a real objective situation that is a true judgment and to which
various true propositions reading: "Socrates exists" may be said
to refer, we have been unsuccessful in our search for another real
objective situation that might be called a judgment and to which
various false propositions reading: "Socrates does not exist" might
similarly be said to refer.

The judgment 'the existence of Socrates/ appearing neither as
self-contradictory nor as undated, etc., is real. And 'the non-
existence of Socrates' is unreal. What, however, about the reality
or unreality of 'the existence of Ivanhoe'? If Ivanhoe appears with
the characteristic of being generally discredited, Ivanhoe is unreal
and the qualities that are alleged to inhere in such an Ivanhoe
are unreal. It would seem that if my subsistent is an existing
Ivanhoe, I am apparently thinking about an Ivanhoe that, by
hypothesis, is real and about the real judgment: the existence of
Ivanhoe. But if I appear to be thinking about an Ivanhoe that
subsists both as real and as generally discredited, my subsistent
appears as implicitly self-contradictory and, in the sense in which
we are using the term "reality," is unreal. 18 If Ivanhoe appears
as generally discredited, this Ivanhoe is unreal and each of the
qualities inhering in this Ivanhoe is unreal. 'The existence of
Ivanhoe 1 is unreal; and 'the non-existence of Ivanhoe' is unreal.
Just as there is no real 'non-existence of Socrates' that might be
called a judgment, so there is no real 'existence of Ivanhoe' and
no real 'non-existence of Ivanhoe' that might serve as real judg-
ments.

It appears then that the only real objective judgment involved
in a singular existential proposition is that directly referred to
by a true affirmative singular existential proposition, namely,

109

existence appearing as a quality of some real entity. The proposi-
tion: "Socrates exists*' which appears on this page and the propo-
sition: "Socrates est" which appears in some Latin manuscript
both refer to a common judgment which is real and true. Both
of these propositions may be regarded as deriving their truth
from the truth of the judgment: the existence of Socrates. But
the 'existence of Ivanhoe' to which two sentences each reading:
"Ivanhoe exists" might be held to refer is not a real judgment
at all. And so there is no real objective judgment which these
* wo sentences have as their common reference, no real objective
judgment whose falsity determines the falsity of these two propo-
sitions, no fact to which these two false propositions are directly
related.

Our desire then to determine the truth or falsity of groups of
propositions by first determining the truth or falsity of objective
judgments to which they refer has been only partially carried out.
If various false propositions reading: "Socrates does not exist"
are to be regarded as having a common reference to a real objec-
tive situation, the reference which they may be regarded as hav-
ing in common is what we might call a contra-reference to the
true judgment: 'the existence of Socrates.' And even this sort of
common contra-reference is lacking as a common characteristic of
various true propositions each reading: "Ivanhoe does not exist."

It appears then that we can not describe truth and falsity merely
with respect to objective judgments or facts and expect the dis-
tinction between truth and falsity thus determined within the
domain of judgments to indicate to us where falsity ends and
wh^ere truth begins within the entire domain of propositions, or
even within the entire domain of singular categorical existential
propositions. The truth of our sentence: "X exists" and the falsity
of our sentence: "X does not exist" may be said to be corollaries
of the truth of the judgment: the existence of X. But the truth of
our sentence: "Y does not exist" and the falsity of our sentence:
"Y exists" are laid down as partial explanations of "truth" applied
directly to the domain of sentences or propositions.

At this point we have behind us the determination of the
signification of "truth" with respect to certain entities that we
call "judgments." And we have behind us the determination of the
significations of both "truth" and "falsity" with respect to singu-

110

lar categorical existential propositions o ours. How is it, however,
xvith respect to categorical existential propositions that are not
singular? How is it with respect to our sentences: "All men exist,"
"Some men exist/' "No men exist" and "Some men do not exist"?
The universal 'man* it will be remembered, 19 "may ... be given
a place on our list of entities denoted by 'existence* along with
Socrates and Plato." Just as the alleged individual Socrates may
be real and the alleged individual Ivanhoe unreal, so the alleged
universal 'man* may be real and the alleged universal 'centaur'
unreal. Just as we hold that, when the alleged individual X is
real, our proposition: "X exists" is true and our proposition: "X
does not exist" false, so let us hold that, when the alleged univer-
sal U is real, our proposition: "Some U's exist" is true and our
proposition: "No U's exist" false. And just as we hold that when
the alleged individual X is unreal, our proposition: "X exists"
is false and our proposition: "X does not exist" true, so let us
hold that, when the alleged universal U is unreal, our proposition:
"Some U's exist" is false and our proposition: "No U's exist"
true. If, then, the alleged universal 'centaur' is unreal in our
sense of "reality," our sentence: "Some centaurs exist" is false as
we explain our term "falsity" and our sentence: "No centaurs
exist" true as we explain our term "truth." And if the alleged
universal 'man' is real in our sense of "reality," our sentence:
"Some men exist" is true and our sentence: "No men exist" false.
There is, to be sure, the proposition: "Some men do not exist"
as well as the proposition: "Some men exist," the proposition:
"All men exist" as well as the proposition: "No men exist." As
has already been pointed out, however, "all men" as it occurs in
an existential proposition, is synonymous either with "All exist-
ing men" or with "All subsisting men." 20 But: "All subsisting
men exist" is, let us say, false. And I can think of no assertion
expressed in: "All existing men exist" that is not expressed in:
"Some men exist." As "All men exist" is synonymous either
with: "All existing men exist" or with: "All subsisting men exist,"
so: "Some centaurs do not exist" is, it would seem, synonymous
either with: "Some subsisting centaurs do not exist" or with:
"Some existing centaurs do not exist." But our sentence: "Some
subsisting centaurs do not exist" is, let us say, true. And if: "Some
existing centaurs do not exist" is to be considered at all, I can

111

think of no assertion expressed in it that is not expressed in: "No
centaurs exist." Since our sentences: "Some men exist" and: "No
centaurs exist" have both been determined to be true in our sense
of "truth," our sentence: "All existing men exist" is, we hold,
true; and our sentence: "Some existing centaurs do not exist"
true. And since our sentences: "No men exist" and: "Some cen-
taurs exist" have both been determined to be false in our sense
of "falsity," "Some existing men do not exist," which seems to be
synonymous with the former, is, we hold, false, and: "All existing
centaurs exist," which seems to be synonymous with the latter,
likewise false.

We may then formalize as follows our explanations of "truth"
and "falsity" with respect to such categorical existential proposi-
tions of ours as: "Some U's do not exist" and: "All U's exist."
If the alleged universal U is real, "All subsisting U's exist" is
false and "All existing U's exist" true, "Some subsisting U's do
not exist" true and "Some existing U's do not exist" false. And
if the alleged universal U is unreal, then "All subsisting U's exist"
and "All existing U's exist" are both false, "Some subsisting U's do
not exist" and "Some existing U's do not exist" both true.

A categorical existential proposition of ours may express an
assertion with respect to alleged existing entities or with respect
to alleged subsisting entities; it may be affirmative or negative; it
may be a singular proposition, a particular proposition or a uni-
versal proposition. In any case it is true or false according as the
individual or universal whose existence is asserted is real or un-
real; true or false according as the individual or universal whose
non-existence is asserted is unreal or real. It is thus some entity's
reality or unreality in our sense of "reality" that determines the
truth or falsity-as we explain "truth" and "falsity"-of each cate-
gorical existential proposition of ours.

But what about the categorical existential propositions of
others? Since "existence" as used by others may not have the
meaning we have assigned that term, the: "Socrates exists" of
some other writer may not express an assertion with respect to
Socrates which is identical with the assertion expressed in our:
"Socrates exists." Shall we say that his: "Socrates exists" is true if
Socrates exists in the sense in which he is using "existence," in a
sense of "existence" which is perhaps vague and indefinite? Or

112

shall we say that his: "Socrates exists" is true if Socrates exists
in the sense in which we have explained "existence"? The former
course leads to as many meanings o "truth" as there are meanings
of "reality." For, taking such a course, the: "Socrates exists" of
one writer would be true, if the Socrates presented complied with
one set of qualifications; the: "Socrates exists" of another writer
true, if the Socrates presented complied with another set of quali-
fications. No author's: "Socrates exists" is true, let us say, unless
Socrates exists in the sense in which we have explained "exist-
ence." But no author's: "Socrates exists" is true, let us also say,
if it is a statement that we should express in "Socrates subsists." 21
Since our sentence: "Socrates subsists" expresses no assertion 22
and is, we hold, neither true nor false, the: "Socrates exists" of
some other writer that is synonymous with it likewise expresses no
assertion and is likewise, let us hold, neither true nor false. The
proposition that is true is our: "Socrates exists." And the propo-
sition that is true is the proposition of some other writer that is
synonymous with it, whatever form it may take. The: "Socrates
exists" of some other author is true, let us say, if it is synonymous
with a proposition which, in the form in which it would be ex-
pressed by us, is true. The: "Socrates exists" of some other author
is false, let us say, if it is synonymous with a proposition which,
in the form in which it would be expressed by us, is false.
And the: "Socrates exists" of some other author which is synony-
mous with no proposition as it would be used by us is, let us say,
neither true nor false.

Our terms "truth" and "falsity" have been explained with re-
spect to categorical existential propositions of ours and with re-
spect to propositions of others that are synonymous with them.
Each such proposition is true or false according as some entity is
real or unreal. Indeed it is the reality or unreality of some entity
or of some entities using "reality" in our sense of that word that,
we hold, determines the truth or falsity of each sentence of ours
that is a proposition. For each real declarative sentence of ours
which does not merely predicate subsistence, which conforms to
the grammatical rules of the language in which it is expressed,
and which contains words, word-groups and phrases representing
or intending to represent subsistents, 23 each such sentence of ours
is, we hold, synonymous with one or more of our categorical exist-

113

ential propositions. The explanation of our terms "truth" and
"falsity" in their application to propositions of ours which are not
categorical existential propositions is thus to be accomplished
through the reduction of such propositions to the categorical
existential propositions of ours with which, we hold, they are
synonymous.

To say that proposition B as it occurs in this treatise is synony-
mous with our existential proposition A is to say that A and B
express similar mental attitudes of mine. Since A, being a categori-
cal existential proposition, is true or false according as some al-
leged entity is real or unreal in our sense of "reality," the reader
is enabled to determine the alleged entity upon whose reality the
truth or falsity of our proposition B depends. It would seem to
require only patience and circumspection to designate categorical
existential propositions of ours synonymous-for-me with each
proposition as it might be used by me; and thus to support the
assertion that each proposition as it might be used by me is
synonymous with one or more of our categorical existential prop-
ositions. Moreover, the designation of synonymous propositions
sufficient to enable our terms "truth" and "falsity" to be applied
to each of the propositions in this treatise will be a fairly adequate
explanation of our terms "truth" and "falsity."

Let us however not lose sight of those sentences outside of this
treatise that do not have the form of categorical existential propo-
sitions. There are sentences outside of this treatise which do not
express an assertion either of existence or of non-existence in our
sense of "existence" and which consequently are neither true nor
false as we use "truth" and "falsity." But there is the sentence A
of some other writer which has the form of a categorical exist-
ential proposition and which expresses an assertion of existence or
of non-existence in some other sense of "existence." And there is
that writer's sentence B which, whereas it does not have the form
of a categorical existential proposition, expresses a mental attitude
of its author's identical with that expressed by his proposition A.
From the point of view of its author, B, that is to say, is synony-
mous with A. From the point of view of its author, there is ex-
pressed in B an assertion of existence in the very sense of "exist-
ence" in which there is an assertion of existence expressed in A.
The proposition B occurring in this treatise, and the existential

114

proposition A of ours to which it will be reduced, they both, we
assume, express an assertion of existence in a sense of "existence"
different from his. But in choosing the categorical existential
proposition A of ours to which our proposition B is to be reduced,
let us not lose sight of the A of some other author with which that
author's B seems to be synonymous. If some other author's:
"There is no Socrates" seems to be synonymous with his: "So-
crates does not exist," then, even though it is another sense of
"existence" that is involved, let us say that our: "There is no So-
crates" and our: "Socrates does not exist" are synonymous with
one other, that our: "There is no Socrates" expresses an assertion
of existence identical with that expressed in our: "Socrates does
not exist," that our: "There is no Socrates" is true or false accord-
ing as Socrates is unreal or real in our sense of "reality." Let us in
short attempt to conform with general usage in reducing to cate-
gorical existential propositions those propositions of ours which
are not categorical existential propositions, even though in our
case "existence" is used in one sense and in the case of general
usage in some other sense.

There is moreover, let us suppose, the categorical existential
proposition A outside of this treatise which expresses an asser-
tion of existence or of non-existence in our sense of "existence."
And there is the proposition B outside of this treatise which does
not have the form of a categorical existential proposition, but
which likewise expresses an assertion of existence or of non-
existence in our sense of "existence." The sentence outside of
this treatise which expresses no assertion of existence or of non-
existence in our sense of "existence" is, we have said, 24 as we use
"truth" and "falsity," neither true nor false. But our terms "truth"
and "falsity" are to be applied to propositions in which there are
expressed assertions of existence or of non-existence in our sense
of "existence," whether these propositions be propositions of ours
or propositions of others. And yet in order that our terms "truth"
and "falsity" may be applied to these propositions of others that
do not have the form of categorical existential propositions, we
must determine the categorical existential propositions of ours
with which these propositions, as used by their authors, are
synonymous.

The reduction of propositions that are not existential in form

115

to the categorical existential propositions of ours with which they
are synonymous is thus to be considered from two points of view.
On the one hand, we have the task of explaining our terms "truth"
and "falsity" in their application to propositions, not categorical
existential propositions, as they occur in this treatise or as they
might be used by me. And on the other hand, we have the task
of explaining our terms "truth" and "falsity" in their application
to those propositions of others which are not categorical existen-
tial propositions but which may perhaps be synonymous with
categorical existential propositions of ours. With respect to the
former task we can speak with assurance. For, although we choose
to be guided by general usage in determining the existential
propositions with which a proposition of a given form as used by
us is to be synonymous, it is our usage that is being set forth, it
is what is synonymous for me that is being stated. With respect
to the latter task, however, we can not speak with assurance. For
even though it should be existence in our sense of "existence" that
is asserted in categorical existential propositions of others, one
writer's proposition that is not a categorical existential proposi-
tion may be synonymous with a certain categorical existential
proposition of ours, another writer's synonymous with another of
our categorical existential propositions, a third writer's synony-
mous with none of our categorical existential propositions at all.
We can but point out the existential proposition of ours with
which some such writer's proposition, not explicitly existential,
may be presumed to be synonymous, point out the entity or
entities whose existence or non-existence in our sense of "exist-
ence" may be presumed to determine the truth or falsity of his
proposition. And we can on occasion point out alternative cate-
gorical existential propositions of ours with which his proposition
may be synonymous, point out alternative entities whose existence
or non-existence in our sense of "existence" may determine the
truth or falsity of his proposition. The meaning of our terms
"truth" and "falsity" in their application to propositions of
various forms as they might be used by me can, in short, be
adequately set forth. But even where "existence" has the meaning
that it has in our writings, the application of our terms "truth"
and "falsity" to propositions of others will vary with the asser-
tions of existence or of non-existence that a proposition of a given

116

form is used to express.

There is our singular affirmative categorical existential prop-
osition: "This large house exists"; and there is the proposition:
"This house is large," which is synonymous with it. Similarly,
it would seem that, as generally used, and certainly as it occurs
in this treatise, "Socrates, the author of the Critique of Pure
Reason, exists" is synonymous with: "Socrates is the author of
the Critique of Pure Reason." So with: "The man Socrates
exists" and: "Socrates is a man." And so with: "A prince named
Orion who had seven daughters and lived at some past date
exists" and: "Once upon a time there lived a prince named Orion
who had seven daughters." Both in the sense of being synonymous-
for-me and in the sense of being synonymous as generally used,
the singular affirmative proposition: "Si is P" is, we hold, synony-
mous with some singular existential proposition: "SiP exists."
This large house is, we assume, a real entity, the man Socrates a
real entity. Our proposition: "This large house exists" is, then,
true and: "The man Socrates exists" true. And so our proposi-
tion: "This house is large" is true and our proposition: "Socrates
is a man" true. On the other hand, Socrates, the author of the
Critique of Pure Reason, is, we assume, an unreal entity and
Prince Orion with seven daughters an unreal entity. And so it
follows that, as we explain "falsity," "Socrates, the author of
the Critique of Pure Reason, exists" is false and "A prince named
Orion who had seven daughters and lived at some past date
exists" false; hence "Socrates is the author of the Critique of Pure
Reason" false and "Once upon a time there lived a prince named
Orion who had seven daughters" false.

"This house is large" is, we assume, a true proposition. We
assume, that is to say, that this large house is a real entity, that
this house, considered as a unit enduring from its construction
to its demolition, has the quality of largeness inhering in it. But
what about: "Caesar crossed the Rubicon"? The quality of cross-
ing the Rubicon was not a quality of that phase of Caesar's life
in which he was combatting Vercingetorix or of that phase of
his life in which he was consorting with Catiline. Strictly speak-
ing, the quality of crossing the Rubicon does not inhere in Caesar
taken as a substance enduring from birth to death; rather, it may
be held to inhere in a brief phase of Caesar's life, ia the transitory

117

substance which is Caesar at a momentous instant in his career. 25
If: ' 'Caesar crossed the Rubicon" is synonymous with: "A Caesar
crossing the Rubicon throughout his career exists/' then our:
"Caesar crossed the Rubicon" is false. It is our existential propo-
sition: "Caesar-at-moment-M, having the quality of crossing the
Rubicon, exists" that is, we may say, true. And it is only if it is
synonymous with this latter proposition that: "Caesar crossed the
Rubicon" is true. Generalizing from the example: "This house
is large," it may seem that any proposition: "Si is P" is to be
reduced to a corresponding existential proposition of the form:
"SiP exists." But, both generally and perhaps in this treatise too,
we on occasion refer to some part or related substance by using
words which, if used out of context, would refer to the whole. We
may use the term "France" to refer to the government of France,
may say: "Virgil is difficult to translate" in place of: "The poems
of Virgil are difficult to translate." "Si is P" is, we hold, both
generally and in this treatise, synonymous with some existential
proposition of the form "SiP exists." But the Si occurring in
"SiP exists" may not refer exactly to the entity which our original
subject-term, taken out of context, would normally represent.
"Caesar crossed the Rubicon" is, it would seem, synonymous with
the existential proposition: "Caesar-at-moment-M, having the
quality of crossing the Rubicon, exists," not with the existential
proposition: "A Caesar crossing the Rubicon throughout his
career exists." And "Washington crossed the Hellespont" is, it
would seem, synonymous with the existential proposition: "Wash-
ington-at-some-moment-M, having the quality of crossing the
Hellespont, exists," not with the existential proposition: "A
Washington crossing the Hellespont throughout his career exists."
It is because the former proposition of ours is false, not because
the latter is false, that our: "Washington crossed the Hellespont"
is false.

The question has been asked how the entity represented by the
subject-term of a proposition can be the entity represented by
the predicate-term. 26 It may seem that for S to be P, for "S is P"
to be true, "S" and "P" must refer to the same entity and express
identical mental attitudes. For S, it may be held, can only be S;
it cannot be P in addition. And P, it may be held, can only be
P. As we explain "truth," however, it is not necessary, in order

118

for: "This house is large" to be true, that this house be identical
with largeness. "This house" may express one mental attitude,
"large" another. "This house" may represent a substance and
"large" a quality of that substance. What is necessary, in order
that: "This house is large" may be true as we explain "truth,"
is that this large house exist in our sense of "existence." And this
large house can exist only if there are instances of a quality in-
hering in a substance. Similarly, since: "Socrates is a man" is
synonymous with: "The man Socrates exists," "Socrates is a man"
can be true only if there are individuals that are instances of uni-
versals. Problems concerning substance and quality and prob-
lems concerning the universal and its individual instances will,
however, engage our attention further on in this treatise. 27 It is
in later sections of this treatise that we shall arrive at conclusions
from which it will follow that the S P that is an alleged substance
with its quality, or an alleged universal instanced in an individual,
may be real. And it is our explanation of "truth" in the present
chapter that determines that, when SiP is real in our sense of
"reality," our "Si is P" is true; and that, when SiP is unreal in
our sense of "unreality," our "Si is P" is false.

The singular affirmative proposition: "Socrates is mortal" re-
duces, we hold, to the existential proposition: "Mortal Socrates
exists"; the singular affirmative proposition: "Socrates is im-
mortal" to the existential proposition: "Immortal Socrates exists."
But what shall we say with respect to the singular negative propo-
sition: "Socrates is not mortal"? Let us assume that a mortal
Socrates exists and that a non-mortal or immortal Socrates does
not exist. Whether, then, "Socrates is not mortal" reduces to the
existential proposition: "Mortal Socrates does not exist" or to
the existential proposition: "non-mortal Socrates exists," it is,
when our sense of reality is involved, a proposition which, as we
explain "falsity," is false. But if: "The present King of France
is not bald" is synonymous with our: "The not-bald present King
of France exists/' it is a proposition which is false; whereas if it
is synonymous with our: "The bald present King of France does
not exist," it is a proposition which is true. In either case, "The
present King of France is not bald" seems to be synonymous with
an existential proposition, seems to be true according as some
alleged entity is real or unreal, false according as some alleged

119

entity is unreal or real. In the field in which we can speak with
certainty, in the field of what is synonymous-for-me, "The present
King of France is not bald/' let us say, reduces to: "The present
non-bald King of France exists"; and "Si is not P" reduces to:
"Si: not-P exists/' Whether the: "Si is not P" of some other
writer reduces to one existential proposition of ours or another
or to no existential proposition of ours at all, the: "Si is not P"
that occurs in this treatise is true or false, as we explain "truth"
and "falsity," according as S a : not-P is real or unreal.

As we explain "truth" and "falsity," "Socrates is not mortal," as
it occurs in this treatise, is true or false according as a not-mortal
Socrates is real or unreal, "Socrates is not a man" true or false
according as a Socrates who is not a man is real or unreal. But "A
Socrates who is not a man" is not synonymous with "A man who
is not Socrates." As we explain "truth" and "falsity," "Si is not
P," as it occurs in this treatise, is true or false, that is to say, not
as PI: not-S is real or unreal, but as Si: not-P is real or unreal.

Both in the sense of being synonymous-for-me and in the sense
of being synonymous as generally used, the singular affirmative
proposition: "Si is P" is, we hold, synonymous with some singular
existential proposition: "SiP exists." 28 And both in the sense of
being synonymous-for-me and in the sense of being synonymous
as generally used, the particular affirmative proposition: "Some
S is P" is, it would seem, synonymous with some particular exis-
tential proposition: "Some SP's exist." Our: "Some men are mor-
tal" reduces to: "Some mortal men exist" and is, let us say, true or
false according as 'mortal man' is real or unreal. Our: "Some
men are black" reduces to: "Some black men exist" and is, let
us say, true or false according as 'black man* is real or unreal.
But just as some instances of: "The present King of France is not
bald" express the assertion that there is no bald present King
of France rather than the assertion that a not-bald present King
of France exists, so some instance of: "Some centaurs are not in-
telligent" may express the assertion that some alleged intelligent
centaurs do not exist rather than the assertion that unintelligent
centaurs exist. However uncertain the existential import of some
instance of: "Some centaurs are not intelligent," the instance that
is an expression of ours reduces to: "Some unintelligent centaurs
exist" and is, let us say, true or false according as 'unintelligent

120

centaur' is real or unreal. Just as our "Si is not P" reduces to:
"Si: not-P exists" and is true or false according as the alleged in-
dividual Si: not P is real or unreal in our sense of "reality," so
our: "Some S is not P," let us say, reduces to: "Some S: not-P's
exist" and is true or false according as the alleged universal S:
not-P is real or unreal.

"This house is large" reduces to, and is synonymous with, the
existential proposition: "This large house exists." "Some men
are mortal" reduces to, and is synonymous with: "Some mortal
men exist." And at least in a sense of synonymity lacking univer-
sality in its application, "This house is not large" reduces to:
"This not-large house exists" and: "Some men are not mortal"
to: "Some immortal men exist." It may be one or another existen-
tial proposition with which some singular negative proposition is
synonymous. It may be one or another existential proposition
with which some particular negative proposition is synonymous.
But there appears to be no similar ambiguity with respect to the
universal negative proposition. The universal negative proposi-
tion: "No men are immortal" seems to reduce to the existential
proposition: "No immortal men exist," the universal negative
proposition: "No stone is alive" to the existential proposition:
"Living stones do not exist." 29 There may, to be sure, be in-
stances of "No stone is alive" in which more is asserted than the
non-existence of 'living stone.' Some one may use the sentence:
"No stone is alive" to assert in addition the existence of 'stone/
the existence of lifeless stones. In the uncertain field of general
usage, "No S is P" may be synonymous with the single existential
proposition: "No SP exists" or with the two existential proposi-
tions: "No SP exists" and "S exists," asserted jointly. In the more
limited but more certain field where we explain our terms "truth"
and "falsity" with respect to propositions as they would be used
by me, "No stone is alive" is, let us say, synonymous with the
single existential proposition: "Living stones do not exist" and
is true or false according as the alleged universal living stone'
is unreal or real in our sense of "reality." There are propositions
singular, particular and universal in which the predicate-term
is not "mortal" or "immortal" or "alive," but, rather, "real" or
"unreal" or "true" or "false." Propositions with predicate-terms
of the latter group require special consideration. But with these

121

terms excepted, using "P" here and indeed throughout this
chapter to stand for any predicate-term other than "real," "un-
real," "true" or "false," the universal negative proposition: "No
S is P," as it occurs in this treatise, reduces to "No SP exists" and
is true or false according as the alleged universal SP is unreal or
real in our sense of "reality."

There is the universal negative proposition: "No stone is alive."
And there is the universal affirmative proposition: "All men are
mortal." Some instances of "No stone is alive" are synonymous
with instances of the existential proposition: "Living stones do
not exist," some instances of "All men are mortal" synonymous
with instances of the existential proposition: "Immortal men do
not exist." 80 There may, we have seen, 31 be instances of "No
stone is alive" in which more is asserted than the non-existence
of 'living stone/ And there may be instances of "All men are
mortal" in which more is asserted than the non-existence of 'im-
mortal man/ It is probable that the land-owner whose sign
reads: "All trespassers will be punished" is merely asserting the
non-existence of unpunished trespassers. 32 It is not probable that
he is asserting in addition that there will be trespassers. But just
as "No stone is alive" may be synonymous, not merely with "Liv-
ing stones do not exist," but may in addition express a belief in
the existence of 'stone/ of lifeless stones; so "All men are mortal"
may be synonymous, not merely with "Immortal men do not
exist," but may in addition express a belief in the existence of
'man/ in the existence of men who are mortal. "In the uncertain
field of general usage, "No S is P" may be synonymous with the
single existential proposition: "No SP exists" or with the two
existential propositions: "No SP exists" and "S exists," asserted
jointly." ss And in the uncertain field of general usage, "All S
is P" may be synonymous with the single existential proposition:
"No S: not-P exists" or with the two existential propositions: "No
S: not-P exists" and "S exists," asserted jointly. We have partially
explained our terms "truth" and "falsity" in their application
to propositions occurring in this treatise by reducing "No S is
P" as it occurs in this treatise to: "No SP exists" and by calling it
"true" or "false" according as SP is unreal or real in our sense of
"reality." Let us further explain our terms "truth" and "falsity"
in their application to propositions occurring in this treatise by

122

holding "All men are mortar* synonymous-for-me with "Immortal
men do not exist" and with "Some mortal men exist," asserted
jointly. The proposition: "All S is P" that occurs in this treatise
is true, that is to say, if SP is real and S: not-P unreal; the propo-
sition: "All S is P" that occurs in this treatise is false if SP is
unreal or if S: not-P is real.

"All men are mortal," as it occurs in this treatise, is synonymous
with "Immortal men do not exist" and "Some mortal men exist/'
asserted jointly. Our universal affirmative proposition: "All S is
P," that is to say, reduces to a universal negative existential propo-
sition plus a particular affirmative existential proposition. But
our: "All existing men exist" has not been described as synony-
mous with a corresponding pair of existential propositions. Our
"All existing men exist" has been described as synonymous with
"Some men exist," 84 not with: "Non-existing existing men do not
exist" plus "Some existing men exist." It reduces, that is to say,
to a particular affirmative existential proposition and thus is no
instance of our universal affirmative proposition: "All S is P."

Just as in our "All men exist" the word "all" is not the mark
of what we call a universal affirmative proposition, so in the un-
certain field of general usage the word "all" may occur in proposi-
tions which are not universal propositions. "All the books in the
British Museum would fit into Westminster Abbey" is, it would
seem, a singular proposition. 85 And so is the nursery rhyme: "All
the King's horses and all the King's men could not put Humpty
Dumpty together again." For the latter proposition, despite its
use of "all," appears synonymous with a singular existential prop-
osition of the form: "The individual army exists which is made
up of such and such members and which has the quality: inability
to perform such and such a feat."

Just as the word "all" may occur in a proposition which is
singular rather than universal, so the word "all" may occur in
what seems to be an enumerative proposition rather than a uni-
versal proposition. Unlike the instance of the universal propo-
sition: "All men are mortal" which, as it occurs in this treatise,
expresses a belief in the existence of the universal 'mortal man/
there are propositions of the form: "All S is P" which seem to
express a belief in the existence of various individual SP's. Vari-
ous instances of "All of the pieces of furniture in this room are

123

old" for example, seem not so much to express belief in the
reality of the universal: 'piece of furniture in this room that is
old' and belief in the unreality of the universal: 'piece of furniture
in this room that is not old'; they seem rather to express belief
in the existence of various individuals each of which is presented
as an old piece of furniture in this room. There is, in short, the
instance of "All S is P" which is an enumerative proposition and
which may be read: "Each S is P." And whereas the universal
affirmative proposition as it occurs in this treatise is true if the
universal S: not P is unreal, and the universal SP real, the enu-
merative proposition: "Each S is P" is, let us say, synonymous
with a group of singular propositions, being true if each of them
is true, false if one of them is false. "Each S is P" is true, that is
to say, only if the individuals SiP, S 2 P, S 3 P, . . . exist; "Each S is
not P" true only if the individuals Si: not-P, S 2 : not-P, S 3 : not-P,
. . . exist.

Therels then the universal proposition: "All S is P" and the
enumerative proposition which, whereas it on occasion may also
have the form: "All S is P," is less ambiguous in the form: "Each
S is P." The distinction between them, it is often held, is based,
not so much upon the use of the word "each" in the one case and
the use of the word "all" in the other, as upon the fact that in the
one case each S could be enumerated by the author of the propo-
sition, in the other case not. As it occurs in this treatise, "All men
are mortal" is a universal proposition, not merely because it
makes use of the word "all," but because it expresses an assertion
with respect to the universal 'mortal man' rather than an asser-
tion with respect to individual men. Can it not be, however, that
an author who writes: "All men are mortal" is making an asser-
tion with respect to each individual man? Admittedly, he is not
definitely aware of each individual man. But may he not pri-
marily be holding, not that there are some mortal men, not that
'mortal man' exists, but rather that each individual man is
mortal? "A true proposition," says Hobbes, 86 "is that whose
predicate contains or comprehends its subject or whose predicate
is the name of every thing of which the subject is the name. As,
man is a living creature is therefore a true proposition because
whatever is called man, the same is also called "living creature."
To think of the subject as being included in the predicate is,

124

however, to think of one group of entities as being included
within another group of entities. It is to think of groups, of
classes; in short, it involves taking what is represented by the
subject-term distributively. The fact that the truth of "All S
is P" is held by some writers to be a matter of inclusion, of classes
within classes, evidences the fact that "All S is P" is sometimes
taken distributively, that "All S is P" is sometimes the expression
of an assertion that might have been expressed as "Each S is P."

There are, to be sure, propositions occurring in this treatise
which conform with no one of the categorical forms thus far dis-
cussed. We have still to point out the entities upon whose reality
or unreality the truth or falsity of hypothetical and disjunctive
propositions occurring in this treatise depends. And since "P" as
it occurs in this chapter does not cover the predicate-terms "real,"
"unreal," "true" and "false," 37 we have not yet discussed the
truth or falsity of such propositions as: "This proposition is false"
and "Each of the propositions in this book is true." With these
exceptions, however, our terms "truth" and "falsity," in their
application to propositions occurring in this treatise or as they
might be used by me, have at this point, we hold, been explained.
Propositions are true or false, as we explain our terms "truth"
and "falsity," according as some entity or entities are real or un-
real in our sense of "reality." And each proposition as it occurs in
this treatise or as it might be used by me expresses an assertion of
the reality or unrealityin our sense of "reality" of some entity
or entities.

We have moreover explained our terms "truth" and "falsity"
in their application to various propositions of others who use the
term "existence" as we do. To be sure, the conditions under
which such a writer's: "All S is P" is true may not be the condi-
tions under which our: "All S is P" is true. His "All S is P" may be
true whenever S:not P is unreal, whereas our "All S is P" is true
only if S:not-P is unreal and S P real. 88 And his "Si is not P" may be
true when Si P is unreal, whereas our "Si is not P" is true when
Si: not-P is real. 39 "Even though it should be existence in our sense
of "existence" that is asserted in categorical existential propositions
of others, one writer's proposition that is not a categorical existen-
tial proposition may be synonymous with a certain categorical exist-
ential proposition of ours, another writer's synonymous with an-

125

other of our categorical existential propositions, a third writer's
synonymous with none of our categorical existential propositions
at all/' 40 Our "Ivanhoe is Ivanhoe" is a proposition of the form
"Si is P" and, as "truth" and "falsity" have been explained in
their application to propositions of ours, is true only if Si P is
real, only if an Ivanhoe who is an Ivanhoe is real. But the "Ivan-
hoe is Ivanhoe" of some other writer, even if he uses "existence" as
we do, may be synonymous with no existential proposition at all,
may express no assertion of existence in our sense of "existence"
and may consequently, as we explain "truth" and "falsity," be
neither true nor false.

Since Ivanhoe does not exist, our proposition: "Ivanhoe is Ivan-
hoe" is false. And since the alleged universal 'centaur* does not
exist, our proposition: "All centaurs are centaurs" is false. Our
proposition :"A is A" is true if A exists, false if A does not exist.
As we explain "truth" and "falsity" and as we reduce propo-
sitions to categorical existential propositions, it follows that
"A is A" is not always true. " 'A is A' is always true" is, it may
be held, a formulation of the law of identity. But "A is A' is al-
ways true" has as its predicate-term the word "true." And whether
or not "A is A" is true depends upon the meaning of "truth." The
word "truth" may be assigned a meaning such that it will follow
that "A is A" is always true. Or, as in this chapter, the word
"truth" may be assigned a meaning such that it will not follow
that "A is A" is always true. The law of identity, in short, at least,
the law of identity that may be formulated as: " 'A is A* is always
true" is thus dependent upon, and not independent of, the mean-
ing of "truth." Apart from whatever meaning may be assigned
"truth," it is neither a law of thought nor a law of things. It is
within the framework of our explanation of "reality" that there
is the law of things: A real A is real. And it is within the frame-
work of our explanation of "truth" that, when A is real, our prop-
osition "A is A" is true.

Of any pair of propositions: "Si is P" and Si is not P," at
least one is false. Of any pair of propositions: "All S is P" and
"Some S is not P," at least one is false. Of any pair of propositions:
"No S is P" and "Some S is P," at least one is false. These three
sentences taken together may be said to constitute the law of con-
tradiction. But since the word "false" is the predicate-term in:

126

"One of a given pair of propositions is false/' whether or not one
of a given pair of propositions is false will depend upon the
meaning of "falsity." What, then, is the situation with respect to
our propositions, let us ask, when "falsity" has the meaning
assigned it in this chapter? Can we say that, in our sense of
"falsity," of any pair of our propositions: "Si is P" and "Si is not
P," at least one is false; that of any pair of our propositions: "All
S is P" and "Some S is not P," at least one is false; that of any
pair of our propositions: "No S is P" and "Some S is P," at least
one is false?

In order that our "Si is P" may be true in our sense of "truth,"
Si P must be a real entity. And in order that our "Si is not P" may
be true in our sense of "truth," S x : not-P must be a real entity. Si
P and Si: not-P can not both, however, be real entities. For Si P
and Si: not-P could both be real only if the self-contradictory en-
tity Si: P-and-not-P were real, only if an entity were real that, in
the course of our explanation of "reality," was marked out as
unreal. 41 Again, in order that our "All S is P" may be true in our
sense of "truth," S: not-P must be unreal. And in order that our
"Some S is not P" may be true in our sense of "truth," S: not P
must be real. But S: not-P can not be real when it appears un-
real. For as we have explained "reality," the entity that appears
both real and unreal has been marked out as unreal. Similarly
with our: "No S is P" and our: "Some S is P." Our "No S is P"
is true in our sense of "truth" only if SP is unreal, our "Some S is
P" only if SP is real. It follows then that as we explain "truth" and
"falsity" at least one of our corresponding propositions: "Si is P"
and "Si is not P" must be false, at least one of our corresponding
propositions: "All S is P" and "Some S is not P" false, at least
one of our corresponding propositions: "No S is P" and "Some S
is P" false. For each new meaning of the term "falsity," a new
validation of the law of contradiction is, it would appear, re-
quired. What has just been shown is that, in our sense of "falsity"
and with respect to propositions of ours, at least one of each pair
of what are commonly called contradictory propositions is false.

There are the contradictory propositions: "All S is P" and
"Some S is not P," the contradictory propositions: "No S is P" and
"Some S is P." And there subsists the self-contradictory entity A:
not-A and the self-contradictory entity S: P-and-not-P. As words

127

are commonly used, "contradictory propositions" is no doubt a more
familiar and a less awkward expression than "self-contradictory en-
tities." An object that appears round and not round is unreal, it
may appear, because "This object is round" and "This object is not
"round" are contradictory; not "This object is round" and "This
object is not round" contradictory because a round, not-round ob-
ject is a self-contradictory entity. There are those, we have seen,
who regard truth as prior to reality. 42 And a discussion of truth and
reality that permits "reality" to be explained by a reference back to
'truth' has the advantage of permitting the more familiar expres-
sion: "contradictory propositions" to be introduced before 'the
more awkward expression: "self-contradictory entities." It has
been our choice, however, to discuss reality before discussing
truth, hence to introduce the expression: "self-contradictory en-
tity" before introducing the expression: "contradictory propo-
sitions." 4S But the introduction of our term: "self-contradictory
entity" prior to a discussion of contradictory propositions does
not, I hope, detract from the understanding of our expres-
sion: "self-contradictory entity." There subsists the alleged entity
which appears both straight and not-straight, the alleged entity
which appears both round and not-round. And we do, I hope,
succeed in partially explaining our terms "reality" and "un-
reality" when, even prior to a discussion of contradictory propo-
sitions, we mark out such self-contradictory entities as unreal.

Of our propositions "Si is P" and "Si is not P," at least one, we
have seen, 44 is false in our sense of "falsity." We can not conclude,
however, that, of our propositions "Si is P" and Si is not P," at least
one is true in our sense of "truth." Our proposition: "The pres-
ent King of France is bald" is true only if a bald present King of
France exists; our proposition: "The present King of France
is not bald" only if a not-bald present King of France exists. If,
however, there is no present King of France, a bald present King
of France is unreal and a not-bald present King of France unreal,
hence our proposition: "The present King of France is bald"
false and our proposition: "The present King of France is not
bald" false. When the alleged entity Si is unreal, both our propo-
sition: "Si is P" and our proposition: "Si is not P" are false. In-
deed even when Si is real, "Si is P" and "Si is not P" may both be

128

false. This good deed alleged to be yellow may be unreal and this
good deed alleged to be not-yellow may be unreal. Si, in short, may
be real and yet, having regard to the deductions our explanation
of "reality" permits us to make, Si P may be unreal and S^ not-P
unreal.

Similarly with our contradictory propositions: "All S is P" and
"Some S is not P." Our proposition: "All centaurs are intelligent"
is false in that the alleged universal 'intelligent centaur' is un-
real; our proposition "Some centaurs are not intelligent" false in
that the alleged universal 'unintelligent centaur* is unreal. And
just as the alleged individual: 'this good deed* may be unreal both
when presented as yellow and when presented as not-yellow, so
the alleged universal 'good deed' may be unreal both when pre-
sented as yellow and when presented as not-yellow.

Since the word "false" is the predicate-term in "One of a given
pair of propositions is false," it follows that whether or not one
of a given pair of propositions is false will depend upon the
meaning of "falsity." 45 And since the word "true" is the predi-
cate-term in "One of a given pair of propositions is true," whether
or not one of a given pair of propositions is true will depend upon
the meaning of "truth." "In our sense of 'falsity' and with respect
to propositions of ours, at least one of each pair of what are com-
monly called contradictory propositions is false." 46 But, except
for propositions of the forms "No S is P" and "Some S is P," it
does not follow from our explanation of "truth" that, with re-
spect to propositions of ours, at least one of each pair of what are
commonly called contradictory propositions is true in our sense
of "truth." Within the framework of our explanations of "reality"
and "truth," the law of contradiction in at least one formulation of
it has been deduced as valid with respect to propositions of ours.
Within the framework of our explanations of "reality" and
"truth," the law of identity in at least one formulation of it has
been deduced as valid with respect to propositions of oursbut
only provided the subject-term represents an existent entity. But
it is only within much narrower limits that the law of excluded
middle in at least one formulation of it can be found to be valid
within the framework of our explanations of "reality" and
"truth/'

129

Summary

1 X exists in our sense of "existence," then our proposition
"X exists" is true in our sense of "truth" and our proposition
"X does not exist" false in our sense of "falsity." Thus the ex-
planation of our terms "truth" and "falsity" utilizes and refefs
back to the explanation of our term "existence." Various types
of categorical propositions are considered and the entities pointed
out whose existence or non-existence in our sense of "existence"
determine these propositions to be true or false as we explain our
terms "truth" and "falsity."

The so-called laws of thought are statements about what must
be true or must be false. But we can not say what must be true
or must be false until we know what "truth" and "falsity" mean.
"Truth" and "falsity," like "existence" and "non-existence," are
capable of various meanings. It is only after the meanings of
"truth" and "falsity" have been determined that we are in a posi-
tion to consider the validity of the so-called laws of thought.
When "truth" and "falsity" have the meanings we assign those
terms, the law of contradiction is true, the other so-called laws
of thought only qualifiedly true.

MORE ABOUT TRUE AND FALSE PROPOSITIONS

We have at this point agreed that various sentences are real in
our sense of "reality/* some of them being sentences of ours,
some of them sentences of others. Those propositions of others
which express no assertion of existence or of non-existence in our
sense of "existence" are, to be sure, real. But, as we explain our
terms "truth" and "falsity," they are neither true nor false. 1 Ex-
cept for judgments or facts that may be called "true," it is to sen-
tences expressing assertions of existence or of non-existence in
our sense of "existence" that we are limiting the application of
our terms "truth" and "falsity," to sentences expressing asser-
tions of existence or of non-existence in our sense of "existence,"
whether these sentences be propositions of ours or propositions of
others. It is however only in their application to some of these
sentences that we have thus far explained our terms "truth" and
"falsity." Categorical propositions occurring in this treatise,
whether singular, particular or universal, whether affirmative or
negative, are, provided the predicate-term is not "true," "false,"
"real" or "unreal," true or false according as some entity or en-
tities are real or unreal. Categorical existential propositions oc-
curring in this treatise are likewise true or false according as some
entity is real or unreal. And those propositions of others which are
synonymous with one or more categorical existential proposi-
tions as they might be used by me are true or false according as
all of the categorical existential propositions of ours to which they
may be reduced are true or one of them false.

In explaining our terms "truth" and "falsity" as applied to
propositions of others, little more need be said. It remains, how-

131

ever for us to determine the categorical existential propositions
of ours, if any, to which our non-categorical propositions may be
reduced. And it remains for us to determine whether any of
our propositions can not be reduced to categorical existential
propositions, whether any of our propositions express no assertion
of existence or of non-existence in our sense of "existence,"
whether, consequently, any of our propositions, in accordance
with our explanations of "truth" and "falsity," are neither true
nor false.

Within the framework of our explanations of "truth" and
"falsity" as thus far stated, it may seem that our sentence: "This
proposition is true" may be true, false, or neither true nor false.
For "This proposition is true" is not an explicitly existential
proposition like "Socrates exists" nor, since "P" has been said not
to cover the predicate-terms "true" and "false," 2 an instance of "Si
is P." The only sentences which are neither true nor false, how-
ever, are those which are not propositions and those which express
no assertions of existence or of non-existence in our sense of
"existence." Our sentence: "This proposition is true" is what we
call a "proposition"; 3 it does not express an assertion of mere sub-
sistence; it is not synonymous with: "This true proposition sub-
sists." Rather it expresses an assertion that 'this true proposition*
exists; it is, as we explain "truth," true or false according as 'this
true proposition' is real or unreal.

Let us take as our alleged object: 'the sentence "This proposi-
tion is true," apparently presented as false/ What we seem to have
before us is then a 'this proposition' with the characteristic of be-
ing true and with the characteristic of being false. What we seem
to have before us is a subsisting 'this proposition' which appears
self-contradictory, a subsisting 'this proposition' which conse-
quently is unreal. 'This proposition' is real when presented as
true but not false, unreal when presented as both true and false.
And since "This proposition is true," when not presented as false,
exists, "This proposition is true" is itself, let us say, a proposition
which is true.

Just as our sentence: "This proposition is true" exists when
presented as true and not false, so our sentence: "This proposi-
tion is false" exists when presented as false and not true. Since
'this true-false proposition' does not exist, "This proposition is

132

false" is not true. And yet, -since 'this false proposition' exists, a
sentence which expresses an assertion of the existence of 'this false
proposition* does not express an assertion of mere subsistence and
is consequently either true or false. 'This proposition is false'* is,
it follows, false. To be sure, when the predicate-term is not "real,"
"unreal," "true" or "false," then, when Si P exists, our proposi-
tion: "Si is P" is true. 4 And if "This proposition is false" were to
be treated as an instance of "Si is P," then, since 'this false propo-
sition* exists, "This proposition is false" would be true. But "This
proposition is false" is, as we have just seen, not true. And so it
follows that the conditions determining the truth or falsity of a
proposition of ours whose predicate-term is the word "false" are
not always the conditions determining the truth or falsity of a
proposition of ours whose predicate-term is neither "true" nor
"false" nor "real" nor "unreal/* It is some entity's existence or
non-existence which determines a proposition to be true or false,
in our sense of these terms, rather than neither true nor false. But
in one case where an entity exists, a proposition asserting the
existence of that entity is true; in another case where an entity
exists, a proposition asserting the existence of that entity is false.

Our sentence: "This proposition is true'* is a proposition which
is true, our sentence: "This proposition is false** is a proposition
which is false. But whereas our proposition: "This proposition is
true*' is in all instances true, our proposition: "Proposition A,
which is not this proposition, is true*' is, it would seem, true or
false according as proposition A is true or not. And whereas our
proposition: "This proposition is false*' is in all instances false,
our proposition: "Proposition B, which is not this proposition, is
false" is, it would seem, true or false according as proposition B is
false or not.

There exist, let us agree, propositions whose subject terms are
propositions; there exist propositions whose subject-terms are
propositions which in turn have propositions as their subject-
terms. There exist, that is to say, what we may call the first-order
proposition A, what we may call the second order proposition:
"Proposition A is false," what we may call the third-order propo-
sition: "It is true that proposition A is false.** Generalizing, we
may say that a proposition of the (n+l)th order in which the
predicate-term is the word "true" is true or false according as the

133

proposition of the n th order, which is its subject-term, and not
identical with it, is true or not. And we may say that a proposition
of the (n-fl)th order in which the predicate-term is the word
"false" is true or false according as the proposition of the n th
order which is its subject-term, and not identical with it, is false
or not. We thus elaborate the explanation of our terms "truth"
and "falsity" with respect to propositions of ours of higher and
higher order. But no questions concerning propositions of an
allegedly infinite order are involved. For the most complex propo-
sition whose truth or falsity is to be determined will, however
complex, be a proposition definitely presented to us, a proposi-
tion which is real and of a finite order.

There is our singular proposition: "The proposition 'All men
are mortar occurring on this page is a true proposition." And
there is our enumerative proposition: "Each proposition occurring
on this page is true." There is our singular proposition: "The
proposition 'All centaurs are animals' occurring on this page is a
false proposition." And there is our enumerative proposition:
"Each proposition occurring on this page is false." But, as we use
it, "Each proposition occurring on this page is true" is, let us say,
synonymous with "This proposition is true" and "Each remain-
ing proposition on this page is true." And, as we use it, "Each
proposition occurring on this page is false" is, let us say, synony-
mous with "This proposition is false" and "Each remaining propo-
sition on this page is false." Since our proposition: "This proposi-
tion is true" is always true, our proposition "Each proposition
occurring on this page is true" is true if each remaining proposi-
sition on this page is true. And since "This proposition is false" is
always false, our proposition "Each proposition occurring on this
page is false" is never true. If Lucian had been using "existence"
in our sense of "existence" and if he had ended his "True History"
with the statement: "Each of the propositions in this book is
false," his final proposition would have been false in the sense in
which we are using the terms "falsity" and "truth."

"Given any set of objects such that, if we suppose the set to have
a total, it will contain members which presuppose this total, then,"
say Whitehead and Russell, "such a set can not have a total" and
"no significant statement can be made about all its members." 5
But our statement: "Each proposition occurring on this page is

134

false*' is not without meaning in the sense in which the statement:
"Eeny meeny miny mo" is without meaning. Indeed it expresses an
assertion of the existence of various false propositions and, since it
itself exists as a false proposition, it is a proposition which is false.
It is itself, we hold, a proposition; and hence adds to the number
of propositions on this page. And so if it is the last proposition on
a page containing twenty others, that page, it would seem, con-
tains twenty-one propositions and not propositions having no
total at all.

If Si P exists, our proposition "Si is P" is true; whereas if 'this
false proposition' exists, our proposition: "This proposition is
false" is false. To this extent there is a difference between "truth"
as we explain it in its application to certain propositions of the
first order and "truth" as we explain it in its application to cer-
tain propositions of a higher order. It may seem to be a matter
merely of the choice of words whether, as with Whitehead and
Russell, the distinction is said to be between "truth of the first
order" and "truth of a higher order" or whether, as with us, the
distinction is said to be between the conditions under which cer-
tain propositions of the first order are true or are false, and the
conditions under which certain propositions of a higher order
are true or are false. No doubt, some theory of types, though not
Whitehead and Russell's, might distinguish as we do the condi-
tions under which certain propositions of the first order are true
or are false from the conditions under which certain propositions
of a higher order are true or are false. It is to be pointed out, how-
ever, that it is not the order of a proposition alone that determines
the conditions under which a proposition is true or false as we ex-
plain "truth" and "falsity." The conditions determining the truth
or falsity of our second-order proposition: "This proposition is
false" are, to be sure, not the conditions determining the truth or
falsity of a first-order proposition of the form: Si is P. But the con-
ditions determining the truth or falsity of our first-order propo-
sition "Si is unreal" are likewise not the conditions determining
the truth or falsity of our "Si is P." For our "Si is unreal" is true,
not if an unreal Si exists, but if Si is unreal. And our "Si is unreal"
is false, not if an unreal Si does not exist, but if Si is *;eal. The con-
ditions under which propositions of ours are true or false vary with
the f

135

existence are expressed. But it is always the existence or non-exist-
ence of some entity or entities in our sense of "existence" that
determines a proposition's truth or falsity. It is not existence in
one sense that characterizes entities whose existence is asserted in
first-order propositions, existence in another sense that character-
izes entities whose existence is asserted in second-order proposi-
tions. And it is to this extent not truth in one sense that character-
izes first-order propositions, truth in another sense that character-
izes second-order propositions.

There are second-order propositions whose subject-terms are
first order propositions; there are propositions, that is to say, which
are about propositions. And as there are propositions about prop-
ositions, so there subsist relations between relations, qualities of
qualities, classes whose members are classes. Alleged situations of
these various types may present us with difficulties, with apparent
contradictions. Where such contradictions appear, "the appear-
ance of contradiction/' it has been held, 6 "is produced by the
presence of some word which has systematic ambiguity of type,
such as truth, falsehood, function, property, class, relation, cardi-
nal, ordinal, name, definition." Indeed, as these apparent contra-
dictions may elicit similar diagnoses, so, it may be held, they call
for similar solutions. And so what is said about propositions about
propositions, it may be held, indicates what is to be said about
alleged relations between relations, about alleged qualities of
qualities, about alleged classes whose members are classes. No
doubt apparent contradictions apparently presented to us in con-
nection with alleged qualities of qualities or in connection with
alleged classes whose members are classes require our attention at
some point. It is however in connection with our discussion of
qualities and relations that we shall consider the alleged quality
of being a quality. 7 It is in connection with our discussion of uni-
versals that we shall consider the alleged universal whose instances
are univeisals. 8 And it is in connection with our discussion of
meanings that we shall consider an ambiguity in the term "name."
Let us at this point limit our attention to propositions varying
in form and to the conditions under which propositions varying
in form are true or false as we explain our terms "truth" and
"falsity."

Ever since Aristotle, many logicians hold, propositions of the

136

subject-predicate type have occupied our attention too exclusively.
It is felt that many of the sentences in which we normally express
ourselves fall into the subject-predicate form only by an artificial
and unnatural treatment. "King James was King Charles's son,"
for example, is to be symbolized, it is felt, by "A r B" rather than
by "Si is P." Moreover, it has been pointed out, our neglect of
"A r B" has led us to neglect various valid implications, as, for
example, the implications which are valid when "r" is a transitive
relation. We need, however, merely note these criticisms and pass
on. For our task is not to catalog and discuss the implications
that are valid with respect to propositions of various forms. Nor
is our task to catalog the forms in which we normally express
ourselves. No doubt through the existential proposition: "Anne
exists with the quality of having Ruth as her sister and with the
quality of having Mary as her sister," attention is directed to Anne
as it is not directed to her through the relational proposition:
"Ruth, Mary and Anne are sisters." But the various existential
propositions of ours which are synonymous with our: "Ruth,
Mary and Anne are sisters" need no pointing out. Our task at this
point is to explain our terms "truth" and "falsity" in their applica-
tion to propositions of ours, however these propositions may
vary in form. But the conditions under which our relational prop-
ositions are true or false are, it would seem, clear. For with what-
ever shift in emphasis the reduction of them to existential propo-
sitions may be carried out, however inelegantly the existential
propositions to which they are reduced may have to be expressed,
the existential propositions with which they are synonymous are,
it would seem, clear; hence the conditions under which they are
true or false are clear.

Sentences of others which express no assertions of existence or
of non-existence in our sense of "existence" are, we have said, 10
neither true nor false. There are writers whose term "existence"
has a meaning different from that which our term "existence" has.
And there are perhaps sentencesand certainly clauses which
express no assertion of existence or of non-existence in any sense
of "existence." In the hypothetical sentence: "If A is B, C is D,"
the clause: "If A is B" expresses no assertion that there exists, in
any sense of "existence," an A that is B. There is, for example,
the hypothetical sentence: "If it rains tomorrow, the ground will

137

be wet/' And yet, as this sentence is commonly used, whatever
meaning "existence" has for its author, this sentence's initial
clause expresses doubt, rather that belief, in the occurrence of
rain tomorrow. The statement however is: "If it rains tomor-
row, the ground will be wet," not "If it rains tomorrow, the
ground will be dry." If there is any sense of "truth," any sense of
"falsity," in which the former proposition is true and the latter
false, there would seem to be a corresponding sense of "existence"
in which rain is wet and not dry, a corresponding sense of "exist-
ence" in which rain exists with the quality of causing the ground
to be wet, not with the quality of causing the ground to be
dry. Some other author's: "If A is B, C is D" may express no as-
sertion of existence or of non-existence in our sense of "existence";
his sentence may be neither true nor false in our sense of "truth"
and in our sense of "falsity." Nevertheless there would appear to
be some entity whose existence in his sense of "existence" he is
asserting, some entity which from his point of view is an existent
and supports the statement: "If A is B, C is D" rather than the
statement: "If A is B, C is not D." 1X

There is likewise some entity whose existence or non-existence
in our sense of "existence" determines the truth or falsity in our
sense of "truth" and "falsity" of a hypothetical proposition of
ours. And it is by pointing out the entities whose existence or
non-existence determines the truth or falsity of a hypothetical
proposition of ours that we explain our terms "truth" and "fal-
sity" in their application to that proposition of ours.

A hypothetical proposition of ours is, generally speaking, a
proposition having the form: "If A is B, C is D." But "If it rains
tomorrow" is synonymous with: "If rain tomorrow should exist"; 12
"If some men have six legs" synonymous with: "If the universal
'six-legged man' should exist." 1S And so with our "C is D." There
is the hypothetical proposition: "If rain tomorrow should exist,
then wet grounds tomorrow would exist" and the hypothetical
proposition: "If 'six-legged man' should exist, then 'six-legged
animal' would exist." Many of our hypothetical propositions, that
is to say, may be reduced to instances of: "If entity E should exist,
then entity F would exist," may be said to be true or false accord-
ing as the corresponding instance of: "If E should exist, then F
would exist" is true or false.

138

Our proposition: "If rain tomorrow should exist, then wet
grounds tomorrow would exist" does not express an assertion that
rain tomorrow will exist nor an assertion that there will be wet
grounds tomorrow. Our proposition: "If 'six-legged man' should
exist, then 'six-legged animal' would exist" does not express an
assertion that 'six-legged man* exists nor an assertion that 'six-
legged animal' exists. There are however two-legged men; and
'two-legged man' implies 'two-legged animal/ And there are six-
legged insects; and 'six-legged insect' implies 'six-legged animal.'
Likewise there was rain yesterday which caused wet grounds and
rain a month ago which caused wet grounds. If what may be said
to be analogous to rain tomorrow does not cause what is corre-
spondingly analogous to wet grounds tomorrow, then our propo-
sition: "If rain tomorrow should exist, then wet grounds tomorrow
would exist" is false. And if what may be said to be analogous to
'six-legged man* does not imply what is correspondingly analogous
to 'six-legged animal,' then our proposition: "If 'six-legged man'
should exist, then 'six-legged animal' would exist" is false. E may
not exist and F may not exist. But in order for our proposition:
"If E should exist, F would exist" to be true in our sense of
"truth," some entity in some sense analogous to E must exist in
our sense of "existence"; and some entity correspondingly analo-
gous to F must exist. Indeed, the entity or entities that may be
said to resemble E must really cause the entity or entities that
seem correspondingly to resemble F, must really imply the entity
or entities that seem correspondingly to resemble F, or must really
be synchronous and concomitant with the entity or entities that
seem correspondingly to resemble F. Our proposition: "If E should
exist, F would exist," that is to say, expresses an assertion that
entities in some sense resembling E exist; indeed, that they exist
when presented as entering into certain relational situations with
entities seeming to resemble F. Unless these entities thus pre-
sented are real, our hypothetical proposition, let us say, is false.
Provided these entities thus presented are real, our hypothetical
proposition, let us say, may be true.

There is, moreover, not only an assertion of existence expressed
in our proposition: "If E should exist, F would exist"; there is
also an assertion of non-existence. There may or may not be rain
tomorrow, But an alleged rain tomorrow presented as not caus-

139

ing, or not being concomitant with, wet grounds tomorrow is as-
serted not to exist. 'Six-legged man' may or may not be real. But
'six-legged man/ presented as not implying 'six-legged animal/
is asserted not to be real. Only if E presented as not causing, not
implying and not being concomitant with F is unreal, and only
if entities in some sense resembling E presented as entering into
certain relational situations with entities in some sense resembling
F are real, then and only then is our proposition: "If E should
exist, F would exist" true.

"If it rains tomorrow/' we have said, 1 * is synonymous with "If
rain tomorrow should exist"; "If some men have six legs" synony-
mous with "If 'six-legged man* should exist." Since, however, our
proposition: "No men are immortal" has been reduced to: "Im-
mortal men do not exist," 15 it follows that "If no men are im-
mortal" is synonymous with: "If immortal man should not exist."
There is thus not only our hypothetical proposition: "If E should
exist, then F would exist"; there is our hypothetical proposition:
"If E should not exist, then F would exist," our hypothetical
proposition: "If E should exist and E' not exist, then F would
exist"; our hypothetical proposition: "If E should not exist, then
F would not exist." There is, for example, not only our propo-
sition: "If six-legged man should exist, then six-legged animal
would exist," but also our proposition: "If 'animal' should not
exist, then 'man' would not exist." And there is not only our
proposition: "If rain tomorrow should exist, then wet grounds
tomorrow would exist," but also our proposition: "If there should
be no fire, there would be no smoke."

Our proposition: "If it should rain tomorrow, the ground would
be wet" expresses an assertion that rain tomorrow not concomit-
ant with wet grounds will not exist. And our proposition: "If
there should be no fire, there would be no smoke," we may ten-
tatively say, expresses an assertion that the absence of fire con-
comitant with smoke does not exist. But what is this absence of
fire that is asserted not to exist when presented as concomitant
with smoke? Where there is no fire, there is, let us assume, matter
at a temperature below the point of combustion. It is non-com-
busting matter presented as concomitant with smoke that, it
would appear, we are asserting to be unreal. And it is what might
be alleged to exist on a planet where there are no animals that,

140

it would appear, we are asserting to be unreal when presented
as concomitant with man. In order that our proposition: "If E
should not exist, F would not exist" may be true, what may be
alleged to exist in the absence of E must be unreal when pre-
sented as concomitant with F, must be unreal when presented as
not concomitant with what is alleged to exist in the absence of F.

But when we say: "If E should not exist, F would not exist," is
there anything that we are asserting does exist? Are we asserting
that something does exist in the absence of E and is concomitant
with what exists in the absence of F? Are we asserting at least
that something exists which seems to resemble what might exist in
the absence of E and that this entity is concomitant with an entity
that seems to resemble what might exist in the absence of F? Or
does our: "If E should not exist, F would not exist" merely ex-
press an assertion of non-existence, express no assertion of exist-
ence at all? With respect to that with respect to which we can
speak with certainty, with respect to propositions that are ex-
pressions of mine, let us adopt the last and simplest course. Let us
say that our: "If E should not exist, F would not exist" expresses
no assertion not expressed in: "What is alleged to exist in the
absence of E is unreal when presented as concomitant with F." Let
us consequently say that our proposition: "If there should be no fire,
there would be no smoke" is true or false, in our sense of "truth"
and "falsity," according as non-combusting matter alleged to be
concomitant with smoke is unreal or real. And let us say that our
proposition: "If there should be no animals there would be no
men" is true or false according as there is not, or is, a world
containing men but not animals.

There is our categorical proposition: "All centaurs are animals"
and there is our hypothetical proposition: "If centaurs should
exist, animals would exist." Just as our proposition: "All men
are mortal" is true, as we explain "truth," only if immortal men
do not exist, so "All centaurs are animals" is true only if cen-
taurs who are not animals do not exist. 16 "If centaurs should
exist, animals would exist" is likewise true only if centaurs who
are not animals do not exist. For our proposition: "If E should
exist, F would exist" is true "only if E presented as not causing,
not implying and being concomitant with F is unreal." 1T The two
propositions which we are comparing, one categorical and one

141

hypothetical, both, to be true, require the non-existence of cen-
taurs who are not animals. But they differ in the entities that
must exist, if they are to be true. In order for: "If centaurs should
exist, animals would exist" to be true, there need be no centaurs,
only entities analogous to centaurs whose existence causes or
implies or is concomitant with the existence of animals. But in
order for: "All centaurs are animals" to be true, there must be
some centaur that is an animal. Our categorical proposition: "All
centaurs are animals," it follows, is not synonymous with our
hypothetical proposition: "If centaurs should exist, animals would
exist." For with horses, which may be said to be analogous to
centaurs, being real and being animals, and with centaurs not
being real and not being animals, the hypothetical proposition is
true and the categorical proposition false.

"A hypothetical proposition of ours," we have said, 18 "is, gen-
erally speaking, a proposition having the form: "If A is B, C is
D." But along with the assertions expressed in: "If A is B, C is D,"
there may be the assertions expressed in: "A is not B" as when
we say: "If A were B or had been B, C would be or would have
been D." And along with the assertions expressed in "If A is
B, C is D" and expressed in "A is not B," there may be the as-
sertions expressed in: "C is D." We may be asserting that C is
D but that A is not B; and we may also be asserting that A being
B would cause or imply C being D. We may in short assert that
C is D as though or as if A were B.

In the writings of Vaihinger and others much importance is
attached to fictions. There is the fiction: "All of the sun's mass
is concentrated at the centre." And there is the fictitious or "as
if" proposition: "The earth revolves about the sun in an elliptical
path exactly as if all of the sun's mass were concentrated at the
centre." The fiction itself the proposition, for example: "All of
the sun's mass is concentrated at the centre," may be a proposi-
tion that the physicist finds useful to consider. The mental attitude
which has as its apparent object an alleged sun whose mass is con-
centrated at the centre may lead to other mental attitudes di-
rected upon the behavior of the sun as it actually exists. But when
we assert that C is D as if A were B, we are asserting that A is not
B. We are asserting that A is not B; that C is D; that if A should be
B, G would be D. We are asserting for example: "If the sun's

142

mass should be concentrated at its centre, the earth would revolve
in an elliptical orbit about it." And whereas, in order that this
latter proposition may be true, the sun's mass need not be con-
centrated at the centre, there must be something analogous
to a sun whose mass is concentrated at the centre; and this
analogous entity that is real must really imply, or must really
be concomitant with, an entity analogous to an earth that follows
an elliptical path. It is true, let us suppose, that a laboratory ap-
proximation of a body alleged to have its mass concentrated at
its centre does exist, a body, for example, with a dense core. And
it is true, let us suppose, that the satellite of such a body follows
an elliptical path. If, then, among other assumptions we assume
that the earth's orbit is indeed an ellipse, then the fictitious propo-
sition: "The earth revolves about the earth in an elliptical path
exactly as if all of the sun's mass were concentrated at the centre"
is a proposition which is true; whereas the fiction: "All of the
sun's mass is concentrated at the centre" is a proposition which

is false. tl

Vaihinger distinguishes, however, between what he calls real
fictions" and what he calls "semi-fictions." "Semi-fictions," he
holds," "assume the unreal, real fictions the impossible." But if a
real fiction is to be symbolized by: "The self-contradictory entity
E exists," then the fictitious or "as if proposition that is based
upon it becomes, let us suppose: "F exists as if the self-contra-
dictory entity E existed." "F exists as if the self-contradictory
entity E existed" is, however, true -at least this proposition as it
might be used by me is true,-only if an entity in some sense
analogous to the self-contradictory E is real and only if this
analogous entity really causes, really implies, or is really syn-
chronous and concomitant with, an entity analogous to F. Is
there then, we may ask, a real entity that may be said to be
analogous to the E that is presented as self-contradictory? It
each entity presented as analogous to E appears as self-contradic-
tory as E itself, then no entity analogous to E exists and the
fictitious proposition based upon what Vaihinger calls a real
fiction is false. And if a real entity may be said to approximate and
resemble a self-contradictory one, if a many-sided polygon, for
example may be said to be analogous to a circle bounded by
straight lines, then real fictions and semi-fictions seem to require

143

no separate treatment. For in that case the fictions are equally
false and the fictitious propositions based upon them are equally
likely to be true; in that case, whether our alleged E be self-
contradictory or not, there is a real entity that may be said to be
analogous to it, a real entity whose participation in a particular
relational situation is asserted.

There may be no circle bounded by straight lines. But if there
is a many-sided polygon that may be said to be analogous to such
an alleged circle, then the hypothetical proposition that begins
with the clause: "If a circle were bounded by straight lines" may
be true. There may be no men with six legs. But if 'two-legged
man' may be said to be analogous to such an alleged 'six-legged
man/ then the hypothetical proposition that begins with the
clause: "If some men had six legs" may be true. It may not have
rained last Tuesday. But if there have been other instances of
rain all followed by wet grounds, then the hypothetical propo-
sition: "If it had rained last Tuesday, the ground would then have
been wet" may be true. My alcoholic friend may not be seeing
a snake. But if people have seen snakes and have jumped, my
proposition: "He is jumping as though he were seeing a snake"
may be true.

But if other people have really seen snakes, how can their
experiences which are real be really analogous to an alleged
snake-seeing experience which is unreal? How can yesterday's
rain which was real have the real quality of being analogous to
an alleged but non-existent rain last Tuesday? Real entities, it
would seem, can have only real qualities. Unreal entities, it
would seem, can have only unreal qualities. Last Tuesday's rain
is unreal no matter how it is presented. It is unreal; and its al-
leged quality of being analogous to yesterday's rain is unreal.
And yesterday's rain is real only when presented with qualities
that it really has. The quality of resembling an unreal entity is
unreal. And the yesterday's rain that is presented as resembling
an unreal rain is an unreal subsistent, a subsistent other than the
subsisting yesterday's rain which is real. 20

A real entity, we must agree, can not really resemble an unreal
one. But unreal entities may be presented as apparent objects.
And real entities, which to be sure do not really resemble them,
may subsequently be selected as our objects. There are the real

144

words: "Last Tuesday's rain." And after having these real words
before us, we may subsequently select as our object the real en-
tity: yesterday's rain. There may be no entities really resembling
an unreal E. But our term: "Entities resembling an unreal E" is
real; and this term may suggest other terms which not only are
real but which have real meanings. "If E should exist, F would
exist" is a proposition of ours which is real and which may be
true or may be false. A condition of its truth, we now find, is
not that entities really resembling E enter into relational situa-
tions with entities really resembling F, but rather that the real
entity E 1 , suggested by our real phrase "entities resembling E"
enter into relational situations with the real entity F 1 suggested
by our real phrase: "Entities resembling F."

There is our hypothetical proposition: "If A is B, C is D"; and
there is our alternative proposition: "A is B or G is D." Both in
general usage and as an expression of ours, " A is B or C is D or
E is F" is called true if "A is B" is a true proposition or "C
is D" a true proposition or "E is F" a true proposition. And if
each of the included propositions is false, then the alternative
proposition which includes them is called "false." No matter how
disparate the entities whose existence is asserted or denied in "A
is B" and in "C is D," the alternative proposition: "A is B or C is
D," it would seem, may be true. Thus, since both "Caesar crossed
the Rubicon" and "No centaurs are animals" are true in our sense
of "truth," our alternative proposition: "Caesar crossed the Rubi-
con or no centaurs are animals," let us say, is likewise true in our
sense of "truth." Since our proposition: "All men are mortal" is
true in our sense of "truth," our alternative proposition: "All
men are mortal or Washington crossed the Hellespont" is true
in our sense of "truth." And as we explain our term "falsity"
in its application to alternative propositions of ours, since "The
present King of France is a married man" and "This proposition
is false" are both false, our alternative proposition: "The present
King of France is a married man or this proposition is false" is
false. Our alternative proposition is thus a proposition about
propositions, a proposition that resembles: "At least one of the
propositions on yonder page is true." In its simplest form it is
what we have called a proposition of the second order rather than
a first-order proposition like: "All men are mortal" or like: "If it

145

should rain tomorrow, the ground would be wet/' 21

Let us assume that we have before us a true hypothetical propo-
sition that is, or may be reduced to, an instance of: "If E should
exist, then F would exist." Among the various assertions that this
proposition expresses, there is the assertion that E, presented as
not concomitant with F and presented as neither causing nor im-
plying F, does not exist. 22 Since the hypothetical proposition which
we are considering is assumed to be true, the entity whose non-
existence is asserted in it does not exist. And since an E presented
as not concomitant with F and presented as neither causing nor
implying F does not exist, it follows that an E presented as not
even co-existent with F does not exist. If, that is to say, E does not
exist unless it causes, implies or is concomitant with F, then E
does not exist unless it co-exists with F. Either E does not exist at
all or F is also an existent. Thus at least one of two propositions is
true. Either "E does not exist" is true or "F exists" is true. In
short, if our hypothetical proposition: "If E should exist, then F
would exist" is true, then our alternative proposition: "Either E
does not exist or F exists" is true.

If our hypothetical proposition: "If E should exist, then F
would exist" is true, then our alternative proposition: "Either E
does not exist or F exists" is true. It is not to be concluded how-
ever that if our alternative proposition: "Either E does not exist
or F exists" is true, then our hypothetical proposition: "If E
should exist, then F would exist" is true. Our hypothetical propo-
sition: "If E should exist, then F would exist" expresses an asser-
tion of existence as well as an assertion of non-existence. And even
the assertion of non-existence expressed in it is not the assertion
of the non-existence of an E that is alleged merely not to co-exist
with F. It is an E, alleged not to enter into a particular relational
situation with F, which is asserted to be unreal and which, since
our proposition is assumed to be true, is unreal. E does not
merely not exist without F existing; E does not exist without im-
plying F, or without causing F, or without being synchronous and
concomitant with F. 28 A Caesar who crossed the Rubicon existed;
and rain yesterday existed. The two events co-exist in the sense
that the one is not an existent and the other a non-existent. But
Caesar's crossing the Rubicon did not cause yesterday's rain, did
not imply yesterday's rain, was not synchronous with yesterday's

146

rain. As we explain our terms "truth" and "falsity" in their appli-
cation to propositions of ours, our alternative proposition: "Either
Caesar did not cross the Rubicon or it rained yesterday" is true;
our hypothetical proposition: "If Caesar crossed the Rubicon,
then it rained yesterday" is false.

"It rained yesterday" is a true proposition. There is however
a difference between: "It rained yesterday" being presented as
a true proposition and rain yesterday being presented as an
existent entity, a difference between: "Either it rained yester-
day or Caesar did not cross the Rubicon" being presented as a
true proposition and 'rain yesterday or Caesar not crossing the
Rubicon* being presented as an existent entity. There is no real
entity: 'Rain yesterday or Caesar not crossing the Rubicon.' And
the real entity 'rain yesterday* does not really imply 'rain yester-
day or Caesar not crossing the Rubicon/ The implication in short
is from one true proposition to another, not from the existent re-
ferred to in one proposition to the existent referred to in another.
It is the true proposition P which implies the true proposition: P
or Q; not the entity whose existence is asserted in P which implies
some entity described as "E or F" whose existence might be said to
be asserted in 'P or Q.' There are implications between proposi-
tions, that is to say, which can not be reduced to implications be-
tween the entities that seem to be referred to in these propositions.
There are true hypothetical propositions about propositions,
true hypothetical propositions of the second order, that have no
true hypothetical propositions of the first order corresponding to
them.

It is beyond the scope of this chapter to point out the bearing, if
any, which the remarks of the last few pages have upon proposi-
tions advanced in treatises on symbolic logic. "Existence" may
be assigned various meanings; "truth" may be assigned various
meanings; "implication" may be assigned various meanings. And
the relevance of the distinctions to which we have just alluded will
vary with the meanings selected. Our primary task has been to ex-
plain our terms "truth" and "falsity" in their application to
categorical propositions of ours and to alternative propositions
of ours, to hypothetical propositions of the first order and to hy-
pothetical propositions of the second order. And at this point this
part of our task has, it would seem, been accomplished.

147

Let us however not take leave of the alternative proposition
without some discussion of the dilemma, without some discussion
of the situation in which we are alleged to be confronted by two
equally unsatisfactory alternatives. Consider, for example, the
plight of the ship's barber who has agreed to shave each man on
board ship who does not shave himself and no man on board ship
who does shave himself. 24 The barber is himself a member of the
ship's personnel. If he shaves himself, he is breaking his agree-
ment; since he has agreed to shave no one on board ship who
shaves himself. And if he does not shave himself, he is failing to
shave each man on board ship who does not shave himself. The
barber appearing with the characteristic of shaving all non-shavers
and with the characteristic of shaving only non-shavers, like the
Cretan appearing with the characteristic of making the true asser-
tion that no Cretan ever expresses himself in a true proposition 25
is a subsistent implicitly appearing as self-contradictory, a sub-
sistent that is unreal. Of the two statements the barber may be sup-
posed to have made before entering upon his duties, one is false.
Either "I shall shave each man on board who does not shave him-
self" is false or "I shall shave no man on board who shaves him-
self" is false. The sentence: "Either the proposition 'I shall shave
each man on board who does not shave himself is false or the
proposition 1 shall shave no man on board who shaves himself is
false" is, however, not without meaning in the sense in which the
statement: "Eeeny meeny miny mo" is without meaning. 26 Our
alternative proposition expresses an assertion that, of two alleged
false propositions, one presented as false exists. It expresses an
assertion of existence and, as we explain our term "truth," is
true rather than neither true nor false. 27

There is likewise the dilemma that may be supposed to have
been presented to the court in the hypothetical case of Prota-
goras versus Euathlus. 28 Euathlus is supposed to have agreed to
complete payment for the training he had received only after
winning his first case. When his teacher Protagoras sued him for
the unpaid balance and thus forced upon Euathlus his first and
last case, Euathlus is imagined to have proposed to the court a
dilemma. "Either I shall win this case, in which event the court
will have decided that the balance is voided; or I shall lose this
case, in which event I shall never have won my first case." An

148

alleged correct decision in favor of Euathlus is implicitly pre-
sented with contradictory consequences and is unreal. An alleged
correct decision in favor of Protagoras is implicitly presented with
contradictory consequences and is unreal. The agreement to pay
after the first case no matter what the first case might be, and to
pay only after the first case, like the agreement to shave each
non-shaver and no shavers, turns out to have been an agreement
that cannot be kept. Either our proposition: "Protagoras will re-
ceive payment only after Euathlus wins his first case" is false; or
our proposition: "Euathlus will pay after winning his first case"
is false. Unless the case of Protagoras versus Euathlus was implic-
itly excepted, there was no real agreement at all and judgment
must be rendered on the basis that there was no agreement.

There is our alternative proposition: "A is B or C is D." 29 "A
is B" may be positive or negative, singular, particular or universal.
So with "C is D"; and so with any other propositions that are in-
cluded in our alternative proposition. "A" may moreover be
identical with "C," or "B" may be identical with "D." But how-
ever one alternative proposition of ours may differ from another,
despite the multiplicity of types, nevertheless not every proposi-
tion containing the words "or" or "nor" is an instance of "A is B
or C is D." "All animals are vertebrates or invertebrates," for ex-
ample, is not synonymous with "All animals are vertebrates or all
animals are invertebrates," but, as generally used, seems, among
other assertions, to express the assertion that no animal is both
non-vertebrate and non-invertebrate. And "Neither Taft nor
Wilson is now President" seems, as generally used, to be synony-
mous with: "Taft is not now President; and Wilson is not now
President." It is in short our alternative proposition that we have
been discussing, not every proposition containing the word "or"
or the word "nor."

When we turn to the apodeictic proposition, it is likewise not
each proposition containing the word "necessary" or the word
"must" that concerns us. "S must be P" or "S must exist" may
simply express in more emphatic form what would be expressed
in "S is P" or in "S exists"; "S can not be P" or "S can not exist"
may simply express in more emphatic form what would be ex-
pressed in "S is not P" or in "S does not exist." "S must be P" or
"S can not be P" may simply point to the deep conviction with

149

which "S is P" or "S is not P" is asserted. "I am thoroughly con-
vinced that S is P" is, however, no apodeictic proposition, and "I
am thoroughly convinced that S is not P" no apodeictic proposi-
tion.

Whether they be positive or negative, singular, particular or
universal, existential or not explicitly existential, our categorical
propositions, we have said, express assertions of existence, asser-
tions of non-existence, or assertions of the existence of one entity
and of the non-existence of another. They may each be reduced,
let us say, to an instance of: "F exists," to an instance of: "F does
not exist" or to an instance of: "F exists; and F 1 does not exist."
Likewise each of our apodeictic propositions, let us say, expresses
an assertion that some entity must exist; an assertion that some
entity can not exist; or an assertion that it is necessary that one
entity exist and impossible that another exist. Each apodeictic
proposition of ours, that is to say, may be reduced to an instance
of: "F must exist" or to an instance of: "F can not exist" or to
an instance of: "F must exist; and F 1 can not exist." What, how-
ever, is asserted in our proposition: "F must exist" that is not
asserted in our proposition: "F exists"? And what is asserted in
our proposition: "F can not exist" that is not asserted in our
proposition: "F does not exist"?

As we have explained our terms "reality" and "unreality,"
those subsistents are unreal which appear as self-contradictory,
those subsistents unreal which appear as lacking any date, those
subsistents unreal which appear with various other character-
istics. 30 A distinction suggests itself between those unreal sub-
sistents which explicitly or implicitly appear as self-contradictory
and those unreal subsistents which neither explicitly nor implic-
itly appear as self-contradictory. Perhaps we should call sub-
sistents appearing as self-contradictory "impossible subsistents,"
and should call unreal subsistents not appearing as self-contra-
dictory "unreal subsistents" but not "impossible subsistents." We
might then give "truth" and "falsity" significations from which
it follows that "F can not exist" is to be called "true" if F appears
self-contradictory. The sentences: "F appears self-contradictory"
and "F does not appear self-contradictory," however, merely pre-
sent us with subsistents. They seem to put before us an F appearing
as self-contradictory or an F appearing without the characteristic of

150

being self-contradictory. They express no assertions of existence or
of non-existence, are not what we call propositions and hence, as
we have agreed to use the terms, "truth" and "falsity/' are not true
or false at all. 81 Within the statement: "F appears self-contradictory;
therefore F is unreal," it is not the sentence: "F appears self-contra-
dictory" that is true or false, but only the sentence: "F is unreal."

Moreover the alleged distinction between that which appears
self-contradictory and that which does not even implicitly appear
self-contradictory becomes, with further consideration, less clear-
cut. There is the subsistent which appears with the characteristic
of lacking any date. As we explain our term "reality," this sub-
sistent is unreal. But if, in rejecting this subsistent, it is an alleged
real entity appearing as lacking any date that we are rejecting,
then it is an entity implicitly appearing as self-contradictory that
we are rejecting. For the alleged real entity appearing as lacking
any date implicitly appears as real and as unreal, implicitly appears
with characteristics which seem to contradict one another.

The entity appearing as self-contradictory is unreal; the entity
appearing as lacking any date is unreal; the entity appearing as
generally discredited is unreal. But it is not as mutually exclusive
groups of non-existent entities that we have presented these
subsistents. The entity appearing as lacking any date may appear
as generally discredited; the entity appearing as generally dis-
credited may appear as self-contradictory. It is any entity appear-
ing with any characteristic listed in the closing pages of Chapter
Three that is unreal; and any entity listed among the Y's in the
appendix to that chapter. 82 On the other hand, it is only the entity
not appearing with any of these characteristics that is real, only
the entity not appearing with any of these characteristics that,
explicitly or implicitly, is listed among the X's in that appendix.
Among the entities which are real, however, among the entities
not appearing with certain characteristics and listed among the
X's enumerated in the appendix to chapter three, our proposi-
tions explaining our terms "existence" and "reality" do not per-
mit us to point to some as more real and to others as less real. As
we explain our terms "existence" and "reality" there are no
degrees of reality. There are not some entities which merely
exist and others which have a more exclusive kind of existence
to be called "necessary existence." Thus an alleged distinction

151

between merely existing entities and necessary entities is, one
might say, more repugnant to our explanation of ' 'existence" than
an alleged distinction between merely non-existing entities and
impossible entities.

As we explain our term "existence," there are not existing en-
tities and, among them, entities with a kind of existence called
"necessary existence." And as we express ourselves in the proposi-
tion: "F must exist/' "F must exist" does not express an assertion
that F has a kind of existence not asserted in our proposition: "F
exists." Our proposition: "F must exist," let us say, expresses an
assertion that F exists and is implied by some entity E. Our
proposition: "Some animals must exist" is synonymous with the
proposition: "Some entity exists, as, for example, the universal
'man,' which implies that some animals exist." Our: "F must
exist" expresses what might be expressed in: "Therefore F exists."
For, like "Therefore F exists," it refers back to some entity whose
existence has previously been asserted or whose existence has
implicitly been asserted in the context. Our "F must exist" is true
if F exists and is implied by the E thus referred to. Our "F must
exist" is false if F does not exist or is not implied by this E. And
if we are unable to determine which the alleged entity E is that
is alleged to imply F, then we are unable to understand "F must
exist," unable to determine whether it is true or false.

There is our hypothetical proposition: "If E should exist, then
F would exist"; and there is our apodeictic proposition: "F must
exist." They differ, to be sure, in that in the former the term "E"
occurs within the proposition itself, whereas in the latter it is
neighboring sentences that explicitly or implicitly supply the re-
ference to E. They also differ in that, whereas our apodeictic
proposition asserts the existence of some implication, our hy-
pothetical proposition asserts the existence of some relational
situation which may be one of simultaneity or of cause and effect
rather then one of implication. In spite of the rain yesterday, "The
grounds must have been wet" may not be a true apodeictic propo-
sitkmu For yesterday's rain, it may be said, caused yesterday's wet
grounds, but did not imply them. In view of various instances of
rain followed by wet grounds, including yesterday's sequence, in
view furthermore of the non-existence of rain not followed by wet
grounds, the proposition: "If it should rain tomorrow, the

152

ground would be wet" is true. Nevertheless, unless there is an
implication from rain to wet grounds, the proposition: "The
grounds yesterday had to be wet'* is false. 88

Our hypothetical proposition and our apodeictic proposition
differ, moreover, with respect to the assertion of the existence of
F and with respect to the assertion of the existence of E. "F must
exist" is true only if F exists and only if F is implied by E. And
F is really implied by E only if there is a real E to imply it. If
it is the existence of man that enables us to express ourselves in
the true apodeictic proposition: "There must be some animals/'
'man* must exist, 'animal' must exist, and 'man' must imply
'animal.' But 'man' need not exist in order for: "If there should
be men, then there would be animals" to be true, any more than
'centaur' need exist in order for: "If there should be centaurs,
then there would be animals" to be true. It is the existence of an
entity in some sense analogous to man or in some sense analogous
to centaur that is required if our hypothetical proposition is to
be true. It is an entity in some sense analogous to E that must
exist and that must enter into a certain relational situation with
an entity in some sense analogous to F. 84

The apodeictic proposition that we have thus far discussed is
our apodeictic proposition: "F must exist," "F has to exist," "It is
necessary that F exist." What about our apodeictic proposition:
"F can not exist," our apodeictic proposition: "It is impossible
that F exist"? We may say, to be sure, that it is only when F is un-
real that our proposition "F can not exist" is true. But when F is
unreal, this alleged F, with whatever characteristics it may seem to
be presented to us, is unreal. An unreal F is not really implied by
any entity E. 85 The unreality of F is not really implied by any
entity E. The proposition: "Some entity E implies the unreality
of F" is always false. If, then, in explaining our terms "truth" and
"falsity" in their application to our proposition: "F can not exist,"
we were to say that "F can not exist" is true only when "F does
not exist" is true and only when in addition "Some entity E im-
plies the unreality of F" is true, then it would follow that our
proposition "F can not exist" is never true. If F is unreal, it is
some entity that is alleged to exist in the absence of F that may
be real, some entity alleged to exist in the absence of F that may
really be implied by E. 86 Or it is the proposition: "F is unreal"

153

that exists as a true proposition and it is the true proposition: "F
is unreal" that may really be implied by E. Our proposition: "F
can not exist" is true, let us say, if F is unreal and if some entity E
implies what exists in the absence of F or implies the true propo-
sition: "F does not exist." And our proposition: "F can not exist"
is false, let us say, if F is real or if there is no entity E which either
implies what exists in the absence of F or implies the true propo-
sition: "F does not exist." "Men can not be immortal" is true, for
example, in our sense of "truth," if what exists in the absence of
immortal animals implies what exists in the absence of immortal
men or if our true proposition: "No animals are immortal" im-
plies our true proposition: "No men are immortal."

There are the apodeictic propositions: "F must exist" and "F
can not exist." And there are the problematic propositions: "F
may exist" and "It may be that F does not exist." There are the
apodeictic propositions: "It is necessary that F exist" and "It is im-
possible that F exist"; and there are the problematic propositions:
"It is possible that F exists" and "It is possible that F does not
exist." Just as it is not all propositions containing the word "neces-
sary" or the word "must" that are apodeictic propositions, so it is
not all propositions containing the word "possible" or the word
"may" that are problematic propositions. 87 The: "That may be
John" which is synonymous with: "I rather think but am not sure
that that is John" is not what we shall call a problematic proposi-
tion. And tie: "Oranges may be seedless" which is synonymous
with: "Some oranges are seedless" is no problematic proposition.

When F does not exist, there exists the true proposition: "F
does not exist." And there may, in addition, be some entity which
exists in the absence of F. Our proposition "F may exist" is false,
let us say, if some entity E, referred to in the context in which
"F may exist" occurs, really implies the true proposition: "F does
not exist," or really implies what exists in the absence of F. My
hat being in this room implies the true proposition: "My hat,
presented as being in some other room, does not exist." Within a
context which informs us that my hat is in this room, our proposi-
tion: "My hat may be in some other room" is false. If, on the
other hand, there is no true proposition: "F does not exist," or if,
"F does not exist" being true, there is no entity referred to in the
context that really implies it, then, let us say, our proposition: "F

154

may exist" is true. If my hat is in this room, if, that is to say, there
is no true proposition: "My hat, presented as being in this room,
does not exist," then: "My hat may be in this room" is true. And
"My hat may be in this room" may be true even if my hat is in fact
not in this room, even if: "My hat, presented as being in this room,
does not exist" is true. "My hat may be in this room" is true,
provided there is no entity referred to in the context that implies
the true proposition: "My hat, presented as being in this room,
does not exist." "My hat may be in this room" is true, for ex-
ample, if the context informs me only that my hat is not outside
this house. The "F may exist" that is an expression of ours is, in
short, synonymous with: "Either F exists or, if F does not exist,
the proposition 'F does not exist* is not really implied by E." As
we explain our term "truth" in its application to it, "F may
exist" is true if F is real or if, F being unreal, the proposition "F
is unreal," presented as implied by E, is unreal.

It is in an analogous manner that we explain our terms "truth"
and falsity in their application to our problematic proposition: "It
may be that F does not exist." Assuming that our context tells us
that some men are mortal and assuming that 'mortal man' implies
'mortal animal/ then our problematic proposition: "It may be
that mortal animal does not exist" or: "It is possible that no
animal is mortal" is, let us say, false. Assuming, on the other
hand, that our context tells us merely that some plants are mortal,
and assuming that 'mortal plant* does not imply 'mortal animal/
then, even though 'mortal animal' is real, "It is possible that no
animal is mortal" is, let us say, true. And if 'mortal animal' is
unreal, then, no matter what the context, "It is possible that no
animal is mortal" is likewise true.

"As we explain our term 'existence/ there are not existing en-
tities and, among them, entities with a kind of existence called
'necessary existence.' 88 And as we explain our term "truth," no
sentence is true which merely distinguishes those subsistents which
appear self-contradictory from those subsistents which do not ap-
pear self-contradictory. If our sentence: "Whatever is, is possible"
is to be regarded as a true proposition, this sentence is to be re-
garded as expressing, not the assertion that existent entities, in
addition to being real, have a kind of existence called "possible
existence," but rather the assertion that, if an entity exists, then

155

the proposition that it is possible for it to exist is true. It is in
connection with propositions rather than in connection with en-
tities, intended to be represented by the terms of a proposition,
that the word "possibility" has been considered. And it is in con-
nection with propositions rather than in connection with entities
intended to be represented by the terms of a proposition that the
word "necessity" has been considered. Whatever must be, it may
be said, exists. But what is true is not that entities having a spe-
cial kind of existence also have an existence of a more general
kind. What is true, rather, is that, if the proposition: "S must
exist" is true, then: "S exists" is true.

The world of existent entities has on occasion been described
as something of a hierarchy, with effects pointing up to causes and
with conclusions pointing up to premises until at the apex a First
Cause is reached whose existence is not contingent but necessary.
Contingent existents, on such a view, presuppose other existents;
they presuppose, finally, an entity that presupposes nothing out-
side itself, an entity that has necessary existence. As we use the
term "necessity," however, there is, as has been pointed out, no
kind of existence to be called "necessary existence." And as we
have explained our term "truth" in its application to apodeictic
propositions of ours, the proposition "F must exist" is not true
unless F is implied by some entity E. 89 If the alleged Being pre-
sented to us is a Being which appears with the characteristic of
not being implied by anything referred to in the context, then,
as we have explained our term "falsity," the proposition express-
ing an assertion that this Being must exist is false.

It is often a difficult matter to determine whether an entity is
real or unreal. And it is often a difficult matter to determine
whether a sentence placed before us is true or false or, perhaps,
neither true nor false. Whatever the other difficulties, it is a prime
requisite that we recognize the 'reality' and the 'truth* that are in
question. We have, to be sure, not found it possible to attach to
our terms "reality" and "unreality" a signification which is in
accord with every author's use of these terms. And we have not
found it possible to explain our terms "truth" and "falsity" in
their application to categorical propositions, to hypothetical
propositions, to various other propositions varying in form, in
such a way as to conform with the usage of every logician. But

166

the explanations of our terms "reality" and "truth," now com-
pleted, present a reality and a truth. They place before us the
formal conditions under which an entity is real in one sense of
"reality," the formal conditions under which a proposition is true
in one sense of "truth." In order to determine whether or not con-
sciousness exists, we must understand the term "consciousness"
as well as the term "existence." 40 In order to determine whether
or not the sentence: "Some collections are infinite" is true, we
must understand the term "infinite collection" as well as the term
"truth." With the 'reality' before us that our term "reality" repre-
sents and with the 'truth* before us that our term "truth" repre-
sents, we are, we hold, 41 prepared to turn to what, by contrast,
may be called the less purely formal problems of metaphysics. We
are, we hold, prepared to consider the extent to which entities
discussed by metaphysicians are, in our sense of the word, "real";
and the extent to which propositions which assert the existence or
the non-existence of these entities are, in our sense of the word,
"true."

Summary

Chapter Five continues the explanation of our terms "truth"
and "falsity." It asks: Is the proposition: 'This proposition is
false' true or false in the sense in which we are using the terms
"truth" and "falsity"? And it attempts to point out the entities
whose existence or non-existence determines the truth or falsity,
in our sense of "truth" and "falsity," of various types of proposi-
tions of ours not considered in Chapter Four.

The discussion of "This proposition is false" leads to com-
ments on the theory of types. The discussion of the "as if* propo-
sition has implications for the discussion of the problem of error
in Chapter Eight.

DOES THINKING EXIST?

"I was then in Germany to which country I had been attracted
by the wars which are not yet at an end. And as I was returning
from the coronation of the Emperor to join the army, the setting
in of winter detained me in a quarter where, since I found no
society to divert me, while fortunately I had also no cares or
passions to trouble me, I remained the whole day shut up alone
in a stove-heated room where I had complete leisure to occupy
myself with my own thoughts. One of the first considerations
that occurred to me was . . ."

These opening lines from Part Two of Descartes' "Discourse
on Method" seem to introduce to us a situation in which there
was an instance of thinking. This thinking is alleged to have
occurred in Germany, in a stove-heated room, and in winter;
and presumably it was about man, God and the universe. In this
chapter, however, our primary interest is not in determining the
existence or non-existence of Germany, of winter, or of the stove-
heated room. Nor are we at this point interested in determining
whether or not there exists the 'man,' the God or the universe
about which Descartes may be held to have been thinking. Our
present problem is to determine whether or not thinking exists.
To the extent feasible, let us then at this point disregard prob-
lems concerning the existence of brains which may be held to be
the vehicles of thinking; let us disregard problems concerning
the existence of particular settings in which various instances of
thinking may be held to occur; and let us disregard problems
concerning the existence of objects towards which instances of
thinking may be held to be directed.

158

Let us disregard vehicle, setting and object to the extent to
which we can disregard them. But if we are to concentrate our
discussion upon some specific instance of alleged thinking, as, for
example, that suggested by the lines quoted from Descartes, we
must already have passed over the thinking alleged to be alone
in the world, the thinking that is held to be without vehicle, set-
ting, or object. And if we are to discuss the existence of some
specific instance of thinking in a simple and straight-forward
manner, we must already have acknowledged the existence of
some of the features of the setting in which that instance of think-
ing is alleged to have occurred. Our query must be: Granting
that Descartes had a brain and was in a stove-heated room, was
he thinking? For, with the reality of brain, room and thinking all
in question, we should find ourselves confronted by a host of
questions all clamoring at once for solution and all having to be
answered before the reality of Descartes' thinking could be ac-
knowledged.

To be sure, what we have before us when the meaning of our
term ' 'existence" has been determined is, it may seem, merely an
empty canvas. The method we have agreed to employ, it may be
held, imposes upon us the task of filling in this canvas bit by bit.
In considering whether or not Descartes' thinking belongs on this
as yet empty canvas, our method, it may be said, requires us to
assume the non-existence of everything else. But such candidates
for existence as a thinking alleged to be alone in the world
without vehicle, setting or object are, we find, presented as gen-
erally discredited and are unreal. And such candidates for exist-
ence as the thinking of Descartes' that is presented as having a
vehicle and a setting can be discussed in fewer words and in a less
complicated fashion when, instead of regarding thinking, vehicle
and setting as all mere subsistents, we accept the premise that
vehicle and setting are real. We are the less constrained to regard
vehicle, setting and thinking as all mere subsistents, we are the
less reluctant to make use of the premise that vehicle and setting
are real, in that Descartes' brain and the stove-heated room have
already been listed as existents in the appendix to Chapter Three.
As we begin this chapter, we have thus no empty canvas before
us, but rather a canvas containing all of the entities previously
listed as reaL

159

Yet if this be true, our deduction ends with the appendix to
Chapter Three and all the rest of this treatise is mere commentary.
What then becomes of our decision to discuss particular existen-
tial problems "in the proper order"? What becomes of our deci-
sion to "be on the watch for existential problems so related that
the solution of one may reasonably be expected to aid us in the
solution of the other"? 1 We must, I think, distinguish between
logical objection on the one hand and puzzlement and lack of
concurrence on the other. The reader who has read the appendix
to Chapter Three will agree that the entities there listed as real
are real in the sense in which we have explained our term "real-
ity." But he may feel that some of these entities are listed without
due consideration or that our term "reality" has been assigned
a strange and unacceptable meaning. It is in the effort to dissolve
such objections that an analysis of one entity and a discussion
resulting in the reaffirmation of its reality may aid in the analysis
of some other entity and may be utilized in the discussion result-
ing in the reaffirmation of the latter's reality. Except to the extent
to which the listings in the appendix to Chapter Three are too
enigmatic to be understood and require elaboration, the remain-
der of this treatise is not needed. But it does not follow that the
remaining chapters contain no reasoned arguments and that they
appeal merely for psychological concurrence. A conclusion ar-
rived at from one set of premises may again be arrived at from
similar premises or from other premises. A conclusion arrived
at on a second occasion may be redundant, but it is a logical
conclusion nonetheless. It is then as analysis and argument rather
than as rhetoric that the remainder of this treatise is presented.

Indeed it is only within the framework of some explanation of
the term "truth," only after some such section as is incorporated
in Chapter Four, that there is valid argument that may be recog-
nized as valid, and true conclusions that may be recognized as
true. For just as "A is A" may be true in one sense of "truth" but
not in another, 2 so "A implies B" may be true in one sense of
"truth" but not in another. The existential conclusions to be
arrived at in the remainder of this treatise thus not only describe
the entities whose existence is asserted in greater detail than was
possible at the end of Chapter Three, but, in contrast to the con-
clusions of Chapter Three, they follow as conclusions that can be

160

recognized as validly deduced. It is not then an empty canvas
that confronts us as we begin this chapter but rather a canvas
which, although well-filled, requires minute criticism and re-
affirmation. It is not as the painter putting on the initial daubs
of oil that we approach the canvas; but rather as the painter-
critic who concentrates his attention on minute sections of his
work in turn, at each point regarding the rest of the work as un-
questioned and making such adjustments as the section under
consideration requires.

It is then with the premise that there are such entities as brains
and rooms that we inquire whether thinking exists. Yet our ques-
tion is not whether all subsisting instances of thinking exist. For
just as: "All subsisting men exist" is false as we have explained
''truth" and "falsity," so: "All subsisting instances of thinking
exist" is false. 3 And "some subsisting instances of thinking exist"
is true only if there are some such installs as the thinking that
is alleged to have characterized Ooft&rtOG as he paced up and
down the stove-heated room and pondered, or seemed to ponder,
the existence of man, God and the universe. We choose as our
question then whether the instance of thinking that allegedly
characterized Descartes was real. Granted the existence of Des-
cartes' brain rather than the existence of brains generally, and
granted the existence of Descartes* stove-heated room rather than
the existence of settings of all sorts, our query is: Was Descartes
thinking?

Yet, whereas we have what may be described as an individual
situation as our apparent object, it is not clear at this point what
element in this alleged situation is being called: "Descartes'
thinking." There are various alleged entities that need to be
untangled. There is the alleged public object, such as God him-
self, to which Descartes' thinking may be alleged ultimately to
refer. There is an alleged mental attitude which is not in the first
instance content, but said to be directed upon content. And there
is allegedly private content, such as Descartes' idea of God, which
may be held to refer beyond itself to some such public object as
God himself. But it is not the public object that we choose to
call "Descartes' thinking" and not private content. Tfce thinking
whose existence we are primarily questioning in thio chapter J
mental attitude rather thail priVaW ttHlttHtit. It is some such entity

161

as Bsageptes-*- alleged mental activity rather than any image or
picture or obiect. it is, in a word, thinking rather than whatTs
thought.

At first sight the distinction between what is alleged to be
mental attitude and what is alleged to be content, whether pri-
; vate or public, seems clear. And yet this distinction becomes less
clear-cut when we attempt to introspect and to make mental atti-
tude a part of content. fopr mypart," says Hume, 4 "when I enter
most intimately into whafl call rriyself, I always stumble on some
particular perception or other, of heat or cold, light or shade,
love or hatred, pain or pleasure. I never can catch myself at any
time without a perception, and never can observe anything but
the perception.'/" Or, as Lovejoy 5 puts it, "what I seem to discover
when perception occurs is not a perceiving, but a certain complex
of content which is subject to conscious change/* On the other
hand, there are those who hold that what we call "mental atti-
tude" may be introspe'ctecf. And there are others who hold that
mental attitude is an object to be inferred, an object to be in-
ferred even from the circumstances reported by Lovejoy and
Hume. Let us not at this point exclude from the denotation of
"mental attitude" the alleged mental attitude which is presented
as on occasion being an object. So far as we have yet seen, it may
be that, if what we call "thinking" exists, it can be apprehended
by a second act of thinking. It may be that, with respect to such
a second act of thinking, thinking is revealed as content of one
sort or another. And since thinking may be held to be revealed
in introspection or otherwise given as an object, this thinking
whose existence we are to consider is not to be described as some-
thing that is never content. It is rather to be described as some-
thing that, if given as content, is given as attitude, attitude that
is perhaps directed towards other content.

Indeed the possibility of thinking, if it exists, becoming content
is not the only consideration that blurs our initial distinction
between thinking on the one hand and the object of thought on
the other. When we distinguish between thinking or mental
attitude or mental activity on the one hand and object of thought
or private jgtfitent or what is pre^nted as a datum on the other,
we makaaise^f .such terms as "activity" and "passivity," terms
which, it would seem, apply to things which move or are moved,

162

things which attack or are attacked, rather than to such alleged
entities as Descartes' thinking on the one hand and Descartes'
idea of God on the other. We do very little to clarify the distinc-
tion between the mental attitude whose existence we are to con-
sider and the idea of God whose existence we are not at this
point to consider by calling the former "active" and the latter
"passive/' To be sure, Descartes* alleged thinking is not presented
as a mental picture or image, not presented as passive in the way
in which a picture or image is usually passive. But it need j^ot be
presented as manipulating its content in the way in which an
active organism may be held to bring about changes in entities
in its environment. The distinction between what we call "think-
ing" and wihat we call "private content" must at this point remain
a bit blurred. What we call "mental attitude" may by some be
included in what they would call "private content/ 1 Yet what we
call "mental attitude" is presented as not a mental picture or
image and it is presented as not being content except in so far
as it is the object upon which it or some further mental attitude
is directed.

Our problem is whether or not thinking exists. More specifi-
cally, our problem is whether or not, as Descartes paced up and
down his stove-heated room, there existed a mental attitude
apparently directed upon man, God and the universe. We have
at this stage made it clear that we are not in this chapter con-
cerned with the existence of the stove which is in Descartes'
environment, with the existence of God, upon whom Descartes'
thinking is alleged to be directed, or with the existence of a pic-
ture or description of God which may be alleged to be part of
the private content of Descartes* mind. It remains for us to dis-
tinguish what we call Descartes' thinking from certain physical
activities in which Descartes was engaged. Descartes, let us say, was
pacing up and down the room, knitting his brows, staring past
the furniture that was around him. These, to be sure, were phy-
sical activities, whereas his alleged thinking may be said to be a
mental activity. But the thinking whose existence we are ques-
tioning is not at this point being presented as non-physical. Our
query is as to the existence of Descartes' mental attitude, whether
it be non-physical or an aspect of his total bodily reactions. The
mere words "mental" and "physical" do not at this point point

163

to mutually exclusive entities, do not at this point mark off Des-
cartes' alleged thinking from what is roughly called his behavior.

It may be held, to be sure, that what we call Descartes' thinking
is presented as subject to observation by none but Descartes him-
self. Whereas Descartes' behavior may be an object for others,
his thinking, it may be said, is, if it exists, an object for him alone.
We might well make use of this difference, it may be held, to
distinguish the mental attitude whose existence we are to con-
sider from the behavior whose existence we in this chapter as-
sume.jWe may, however, say at once that Descartes' thinking, if
it exists, is not an object for Descartes alone. It is Descartes'
alleged thinking that you and I are now considering, an instance
of thinking, consequently, that, at least implicitly, is presented
as apparently an object for you and for me. 6 Indeed it is only
the thinking, not presented as an object for Descartes alone, that
is presented as free from self-contradiction; only the thinking,
not presented as an object for Descartes alone, that may be real.
Hence it is not in being an object for Descartes alone that Des-
cartes' thinking, if it is real, differs from Descartes' behavior.

But being an object, it may be said, is one characteristic; being
an object which is sensed another. And whereas Descartes' think-
ing and Descartes' behavior are both presented as objects for you
as well as for Descartes, Descartes' thinking, it may be said, is
presented as not only an object for Descartes, but as sensed by
Descartes. However, we do not care to restrict our attention to
an alleged thinking that is presented as having been sensed by
Descartes; or to an alleged thinking that is presented as an entity
that Descartes might have sensed. We do not care to exclude from
our consideration the alleged instance of thinking that may be
alleged not to have been sensed by Descartes. What we are to
consider is a mental attitude of Descartes' that he may or may
not have sensed, a mental attitude that he may or may not have
been able to sense. And with this latitude in the entity which we
are to consider, we can not distinguish Descartes' alleged thinking
from his behavior by a reference to the manner in which that
thinking was apprehended by Descartes.

Is there not a difference, however, between the manner in
which Descartes' contemporaries apprehended his behavior and
the manner in which, if they apprehended it at all, they appre-

164

hended his thinking? His behavior, it may be held, is something
which they saw, his thinking something which they inferred from
what they saw. We have agreed not to limit the entity tinder
consideration to the mental attitude alleged to have been sensed
by Descartes. But shall we not at least describe the entity under
consideration as a mental attitude that is not sensed by others?
Here however we run into the difficulty of distinguishing what is
a sense-datum from what is inferred. "When looking from a win-
dow and saying I see men who pass in the street, I really do not
see them, but infer that what I see are men. . . . What do I see
from the window," asks Descartes, 7 "but hats and coats which may
cover automatic machines? Yet I judge these to be men/' Our
inference however, if it be called inference, is so inseparable from
our apprehension of what is sensed, that we are at once aware of
men. We see two converging tracks with our experienced eyes
and we see the distance. We look at a picture of a landscape and
we see, not a two-dimensional manifold, but a scene which goes
back from foreground to horizon. As Bode says, 8 "we do not first
observe and then supply a context, but we observe by seeing
things as existing in a context/' So, if Descartes' thinking exists,
the contemporary observer may be held to have seen not only Des-
cartes' knitted brow and distant stare, but also the thinking im-
plicit in his total behavior. When we look at Rodin's "Thinker,"
we seem to be aware at once of the alleged thinking; just as we
seem to be aware of depth as soon as we look at a landscape paint-
ing. In both cases it is, one might say, when we attend to the
artist's technique that we distinguish the sense-datum from what
then appears to us to have been inferred. The thinking of Des-
cartes' that we are to consider is presented as likely to be given
to an outside observer as soon as is Descartes' knitted brow or
distant stare. Whether it be physical or non-physical, Descartes'
thinking, if it exists, is as an object so commingled with his other
behavior that any study of his total behavior must include a study
of what we call his thinking.

The distinction between total behavior and thinking is, as we
choose to describe it, not so much the distinction between the
immediately given and the subsequently inferred, as it is the dis-
tinction between the unanalyzed whole and an alleged selection
from this whole. Given the pacing, the staring and the alleged

165

thinking which characterize Descartes, we can say that the pacing
is not the entity whose existence we are to examine; and that the
staring is not this entity either. We may pass from a consideration
of Descartes' total behavior to a consideration of his knitted brow
or distant stare. Or we may pass to a consideration of his alleged
thinking. Indeed, if we accept a suggestion of Alexander's, 9 we
will agree that thinking is normally presented to us as an object
before the knitted brow and the distant stare. It is by separating
out of Descartes' total behavior his alleged interest in man, God
and the universe, it is by concentrating our attention upon one
alleged element in his total behavior, that we come to have as
our apparent object the alleged entity that we call Descartes'
thinking. For whether Descartes' thinking is in his body or merely
associated with his body, it is, if it exists, so intimately associated
with his body that, in having Descartes before us as an unanalyzed
whole, his alleged thinking is within, rather than outside, the
entity before us.

The thinking of Descartes' that may be real may be presented
as a characteristic of Descartes' body like his knitted brow or
distant stare. Or the thinking of Descartes' that may be real, where-
as alleged to be an element abstracted as an object from his total
behavior, may be presented as an entity that is merely associated
with his body, may be presented as an entity that in itself lacks
position and extension. Whereas we may be led to consider Des-
cartes' alleged thinking through having Descartes' total behavior,
Descartes as an unanalyzed whole, as our apparent object, the
alleged thinking that we come finally to consider is, it may be
held, an entity that has no position within Descartes' body and
no position anywhere else, but is rather an entity that is non-
spatial and merely associated with Descartes' body.

We find no clearer exposition of the view that thinking is
immaterial and non-spatial, and merely associated with the body,
than in the writings of Descartes himself. Thinking is for him
the sole attribute of a thinking substance. And this substance
whose sole attribute is thinking and with it the thinking that
is presented to him as inhering in a substance which has no
position and no extension is real, he holds, 10 "because, on the
one side, I have a clear and distinct idea of myself inasmuch as I
am only a thinking and unextended thing, and as, on the other,

166

I possess a distinct idea of body inasmuch as it is only an extended
and unthinking thing/*

Now we shall not deny that an instance of thinking with no
position and no extension is an apparent object. For it is such
an apparent object, such a subsistent, whose claim to reality we
are here attempting to evaluate. Something may, to be sure, be
said with respect to its clarity and distinctness. As Arnauld pointed
out, 11 we appear to apprehend a right triangle clearly and dis-
tinctly even when we do not apprehend the fact that the square
on its hypotenuse equals the sum of the squares on its other
sides. Nevertheless we do not conclude from this that the right
triangle exists without the square on its hypotenuse being equal
to the sum of the squares on its sides; and, he holds, we should
not conclude that thinking is unextended, merely because we
seem clearly and distinctly to apprehend it without extension.
It appears however to be Descartes' more matured opinion that it
is only when two substances art clearly and distinctly appre-
hended without either of them being presented with the essential
qualities of the other, it is only then that we can conclude that
these entities exist as they appear to us. If we could apprehend
the substance 'right triangle* clearly and distinctly without appre-
hending the substance 'triangle the square on whose hypotenuse
is equal to the sum of the squares on its sides' and if we could
likewise apprehend clearly and distinctly the substance 'triangle
the square on whose hypotenuse is equal to the sum of the squares
on its sides' without apprehending the substance 'right triangle/
only then, Descartes would seem to hold, could we conclude that
right triangles exist without this ratio obtaining between their
sides and hypotenuse.

But even with this emendation, even if we limit ourselves to the
cases in which two entities are presented as substances and each
appears without the attributes of the other, how can we "conclude
that the substances are really distinct one from the other from the
sole fact that we can conceive the one clearly and distinctly with-
out the other?" 12 To arrive at a valid conclusion which expresses
an assertion of existence, there must be a reference to existence in
our premises. The validity of Descartes* conclusion depends upon
the validity of some implicit premise which ties together exist-
ence and alleged entities that are clearly and distinctly appre-

167

hended. It is an essential pan of his argument that "God can
carry into effect" (i.e., into existence) "all that of which we have a
distinct idea." And so we find Descartes' argument for the im-
materiality of the soul, his argument for the inextendedness of
thinking, resting upon what in a previous chapter we decided was
an implicit determination of the meaning of the term "exist-
ence." 18

It would ill become us to speak slightingly of an argument be-
cause it makes use of a proposition determining the meaning of
the term "existence." For it has been our thesis that any valid ex-
istential proposition must point back to some proposition in which
the term "existence" is explained. We make no reference to the
'clear and distinct" in our own explanation of "existence." And so
we find Descartes' argument, culminating in the conclusion that
thinking is concomitant with no extension, without relevance to
our own problem. What this suggests, however, is that we turn to
our own explanation of "existence" to determine the existence or
non-existence, in our sense of "existence," of a thinking that is
presented as non-spatial. And when we recall that, for an entity
to be real in our sense of "reality," it may not be presented as
lacking all position, we realize that the alleged instance of think-
ing which is presented as non-spatial is, in our sense of the term,
"unreal." Descartes' alleged thinking as he paced up and down
the stove-heated room may or may not exist. But if it exists, it is
not an entity that is utterly non-spatial.

In rejecting the mental attitude which is presented as non-
spatial however, perhaps we eliminate the possibility of Descartes'
alleged mental attitude being real, however presented. Perhaps the
alternative an alleged mental attitude presented as being spatial,
as having position is so absurd that the unreality of non-spatial
thinking involves the unreality of thinking of any sort. What is
presented as mental, it may be said, is presented as quite different
from what is presented as spatial. It is presented as so different,
it may be said, that any instance of thinking presented as having
position is implicitly presented as generally discredited. Such a
spatial thinking, it may consequently be held, is just as unreal in
our sense of "reality" as the non-spatial thinking which we have
already eliminated.

For one thing, it may be said, there are inorganic phenomena

168

to which scientific formulae apply; and there are organic phenom-
ena to which these formulae do not apply with equal force. There
are organic phenomena, it may be said, which are wayward and
unpredictable and which point to the existence of some entity
whose activities do not fall within the scope of scientific formulae.
It is the waywardness and unpredictability of organic phenomena
which point back, it may be held, to a mental attitude that is non-
spatial and which make incredible a mental attitude that is alleged
to have position with respect to the spatial entities that are its
contemporaries.

So long as we focus our attention upon some alleged mental
attitude presented to us and disregard the organic and inorganic
phenomena that are alleged to be its contemporaries, we can not
come to grips with such a doctrine. Let us then in this chapter
agree that there exist organic phenomena and inorganic phenom-
ena, and that the scientific formulae that have come to be accepted
can on the whole be applied more readily and more satisfactorily
to the latter than to the former. Each organism, let us agree, seems
to have a structure and to develop along the lines that its nature
determines for it. It seems to maintain its own course of develop-
ment with a persistency which is not altogether at the mercy of the
environment. The motions of inorganic bodies, on the other hand,
let us agree, seem to be completely dependent upon the forces
which act upon them. They seem to be such that similar actions
call forth similar reactions; whereas, in the case of organisms,
'learning' takes place and the reaction to a stimulus applied a
second time may not be identical with the first reaction.

With this much common ground established, let us consider
the status of the scientific formulae which, we have agreed, can on
the whole be applied to inorganic phenomena more readily and
more satisfactorily than to organic phenomena. In so far as a
formula is valid, it is, it would seem, both a generalization and a
tool enabling prediction. But, both as a generalization and as a
tool for prediction, it applies, it would seem, not so much to be-
havior as a whole as to qualities which are numbered qualities.
It is not the concrete behavior of an entity that some scientific
formula enables us to predict, but rather, it would seem, some
particular measurable characteristic, as, for example, the number
that is to characterize that entity's speed or the number of de-

169

grees that is to characterize its heat. The admission then that
scientific formulae can be applied to inorganic phenomena more
readily and more satifactorily than to organic phenomena turns
out to be the admission that numbered qualities to which scien-
tific formulae apply are to be found among inorganic phenomena
to a greater extent than among organic phenomena. The mental
attitude which we are to consider is presented as manifested
in organic phenomena which are poor in numbered character-
istics to which scientific formulae apply. The question is whether
a mental attitude so presented may be presented as spatial with-
out being presented as generally discredited.

Organic phenomena are to be called "wayward," it would seem,
if they have characteristics which are numbered and to which
scientific formulae are ready to be applied, and if, nevertheless,
they fail to conform to these formulae. There seems, however, to
be no specific scientific law ready to be applied to organic behavior
that is in fact violated by the apparently teleological behavior of
organisms, no specific scientific law ready to be applied to organic
behavior that 'learning' violates. The disorder that is implicit in
waywardness, as we have described waywardness, seems not to be a
fact. And so, with waywardness described as we have described it,
the mental attitude which is alleged to give rise to wayward
organic phenomena comes to be presented as discredited and un-
real. Just as Descartes' alleged mental attitude does not exist
when presented as non-spatial, so Descartes' alleged mental atti-
tude does not exist when presented as giving rise to organic
phenomena which in our sense are wayward. The mental attitude
which we are to consider is the mental attitude alleged to be mani-
fested in organic phenomena which are poor in numbered quali-
ties. Or it is the mental attitude alleged to be spatial and to be
manifested in organic phenomena which have numbered qualities,
qualities, however, to which, in large part, scientific formula do
not apply or are not ready to be applied. But the mental attitude,
alleged to contravene specific scientific formulae applicable to it,
we at this point reject as discredited and unreal.

Neither the absence of numbered qualities nor the absence of
scientific formulae applying to what numbered qualities there
are, seems to point to an entity that is non-spatial. Before there
were thermometers to measure heat, when heat was presented as

170

an unnumbered quality, heat was not generally presented as non-
spatial or as the manifestation of something non-spatial. And when
heat is presented as a quality that can be measured and assigned a
number, and yet no scientific formula presented which is applic-
able to the relation between the heat of one day and the heat of
another, we still do not think of heat as the manifestation of
something non-spatial. For, the quality that is not numbered,
we seem generally to hold, may perhaps be numbered. And the
qualities which are presented without some scientific formula
which applies to them need not be presented as incapable of
having such a formula apply to them. The mental attitude which
is presented as manifested in organic phenomena which are poor
in numbered qualities is not presented as generally discredited,
we find, when presented as spatial. And neither is the mental
attitude which is presented as manifested in organic phenomena
having numbered qualities to which, in large part, scientific for-
mulae are not ready to be applied.

There is a distinction to be made, however, between the entity
alleged to be poor in numbered qualities and the entity which, it
is alleged, cannot have numbered qualities; a distinction to be
made between the entity presented as having numbered qualities
to which scientific formulae are not ready to be applied and the
entity presented as having numbered qualities to which scien-
tific formulae cannot be applied. We read in McDougalPs "Body
and Mind" that "the soul has not the essential attributes of
matter, namely, extension (or the attribute of occupying space)
and ponderability or mass"; for, says he, 14 "if it had these attri-
butes it would be subject to the laws of mechanism, and it is just
because we have found that mental and vital processes can not be
completely described and explained in terms of mechanism that
we are compelled to believe in the cooperation of some non-
mechanical teleological factor." But when, as with McDougall,
the phrase is "can not be described" rather than "is not de-
scribed," the inference would seem to be from what is non-spatial
to what is not subject-matter for scientific formulae rather than
vice versa. The mental attitude which is alleged to have mani-
festations to which no scientific formula could ever be applied is,
we find, already presented as non-spatial. It is in view of the non-
spatiality with which it is implicitly presented that the mental

171

attitude, alleged to have manifestations to which no scientific
formulae could be applied, is presented as discredited when also
presented as spatial.

The mental attitude is unreal which is alleged to be spatial
and also alleged to be non-spatial. So is the mental attitude which
is alleged to be spatial and implicitly alleged to be non-spatial,
the mental attitude, for example, which is alleged to be spatial and
also alleged to be manifested in phenomena to which scientific
formulae cannot be applied. Not only however is the mental atti-
tude unreal which is alleged to be both spatial and non-spatial;
"the alleged instance of thinking which is presented as non-
spatial" is, we have found, likewise unreal. 15 There remains as
an entity that, so far as we have yet seen, may be real the mental
attitude which is alleged to be spatial and not alleged to be non-
spatial. There remains the mental attitude which is alleged to
be spatial and not alleged to be manifested in phenomena to
which scientific formulae cannot be applied. And this mental
attitude may be real whether it be presented as having or lacking
manifestations which are in fact numbered, whether it be pre-
sented as having or lacking manifestations to which scientific
formulae are in fact or will in fact be applied.

The mental attitude which may be real is the mental attitude
which explicitly or implicitly is alleged to have position with re-
spect to the spatial entities that are its contemporaries. The men-
tal attitude of Descartes' which may be real is the mental attitude
of his which is not merely associated with his body but is alleged
to have position with respect to the phase of his brain and the
phase of the stove that are its contemporaries. Position, to be sure,
may be definite position, position of a sort that a point is alleged
to have; or it may be indefinite position, position of a sort that an
extended entity is alleged to have. Let us however dismiss at once
the mental attitude which is alleged to be at a point. Let us mark
out as unreal the mental attitude of Descartes' which is alleged
to have position with respect to brain and stove, but no extension.
We thus find ourselves considering the mental attitude which is
alleged to have not only position with respect to its contempo-
raries but also extension. We thus find ourselves holding that, if
Descartes had any mental attitude at all as he paced up and down

172

the stove-heated room, that mental attitude had, or was concomit-
ant with, an extension.

Is however an -extended mental attitude at all plausible? Is not
a mental attitude or instance of thinking that is alleged to be
extended presented as discredited and unreal? There are, let
us agree, distinguishable mental attitudes which form an inte-
grated whole and which can not be separated one from the other
as my foot can be severed from the rest of my body. Does it how-
ever follow that "there is a great difference between mind and
body, inasmuch as body is by nature always divisible and the
mind is entirely indivisible"? 16 To be sure, the bolt of blue cloth
or the gallon of water that is presented as extended is implicitly
presented as in some sense divisible. And the mental attitude or
complex of mental attitudes that is presented as extended is like-
wise implicitly presented as in some sense divisible. There is
however a sense in which an extended substance is divisible and
the quality of an extended substance likewise divisible. And there
is another sense in which a quality, without regard to the sub-
stance in which it inheres, is, or is not, divisible. The gallon of
water can be divided into four quarts of water, the bolt of blue
cloth into small pieces of blue cloth. The blueness of the bolt of
cloth is divisible in the sense that the bolt of doth in which it
inheres is divisible. And if thinking or mental attitude is a qual-
ity of an extended substance, dunking is divisible in the sense
that the substance in which it inheres is divisible.

It may however well be another sense of divisibility that is
suggested when we say that blue is a primary color, purple not; or
when we say that some complex of mental attitudes is divisible or
indivisible. The assertion that blue is a primary color is generally
the assertion that blue can not be analyzed or reduced to other
colors, not the assertion that bolts of blue cloth can not be sepa-
rated into parts. And the assertion that thinking is indivisible may
well be the assertion that thinking is not to be analyzed rather
than the assertion that thinking does not inhere in an extended
substance. Whether blue be alleged to be capable of analysis or
not, the blueness that is alleged to be the quality of an extended
and divisible substance is, I find, not presented as incredible and
unreal. And whether the mental attitude of Descartes' that we are

173

considering be alleged to be capable of analysis or not, this en-
tity, alleged to be the quality of an extended and divisible sub-
stance, is, I find, likewise not presented as incredible and unreal.

It may, to be sure, be pointed out that whereas the segments
into which a bolt of blue cloth is cut are all blue, the segments
of some extended substance in which thinking is alleged to in-
here are substances in which no mental attitudes inhere at all.
Yet a round plate may be circular, it would seem, without any of
the fragments into which it is broken being circular. And a mole-
cule may have properties which none of its constituent atoms
have. The alleged circularity of a round plate is not presented as
incredible when the segments are alleged not to be circular. And
the thinking that is alleged to be a quality of some extended brain
or nerve-fibre is not presented as incredible when a segment of
that brain or nerve-fibre is alleged not to be thinking, alleged
not to have a mental attitude inhering in it as a quality.

But if thinking is extended, or the quality of an extended sub-
stance, then the extension of the substance that thinks about a
gallon of water, it may be said, must be four times the extension
of the substance that thinks about a quart of water. If thinking is
extended at all, it may be held, the extension with which it is
concomitant must be proportionate to the extension of the object
upon which it is directed. Thinking about a house would then
have to have the shape of a house, thinking about the moon the
shape of the moon; and a mental attitude apparently directed
upon an inextended object would have to be inextended, and
thus be both extended and inextended at once. If the thinking
that we are considering, the thinking that is alleged to be ex-
tended, had such implicit characteristics as these, it would, to be
sure, be presented as discredited and unreal. But if my uncle is
twice as big as yours, it does not follow that I am twice as big
as you. And if my uncle is twice as big as yours and yet I not
twice as big as you, it does not follow that neither you nor I are
extended at all. Two entities may both be extended and yet the
ratio between their extensions not be equal to the ratio between
the extensions of the entities to which they are respectively re-
lated. And two thinking substances may both be extended and
yet, assuming that they have objects, the ratio between their
extensions not be equal to the ratio between the extensions of

174

their respective objects. The mental attitude o Descartes' that
we are considering is alleged to be concomitant with an extension;
and it is alleged to be the quality of an extended substance whose
extension does not depend upon the extension of the object, if
any, upon which the mental attitude is apparently directed. The
mental attitude that we are considering is not presented with the
implicit characteristics just considered; it is not, so far as we have
yet seen, presented as discredited and unreal.

In order that we might have a suitably limited framework
within which to consider the reality of Descartes' alleged mental
attitude, we have in this chapter agreed that Descartes' brain is
real and his stove-heated room real. 17 Let us likewise agree that
there is some real entity distant from Descartes' body upon which
his alleged mental attitude is alleged to be directed. And let us
agree that there is some real entity distant from Descartes' body
that is alleged to be causally related to his alleged mental atti-
tude. Whether or not the moon is really the object of a mental
attitude of Descartes,' and whether or not the moon really brings
about the mental attitude that may be alleged to be directed
upon it, the moon, let us in this chapter agree, is real and really
distant from Descartes' body. Yet if the moon is real and distant
from Descartes' body, how can a mental attitude which is con-
comitant with an extension within Descartes' body be either
affected by the moon or aware of it? Given a real moon that is
there and an alleged mental attitude that is alleged to be extended
and here, any alleged relational situation, whether of cause and
effect or of subject and object, is so incomprehensible, it may
be said, that the mental attitude, which is presented as extended
and here, comes to be presented as incredible and unreal.

The statement that the distant moon affects a thinking sub-
stance which is extended and here would seem to be less per-
plexing than the statement that the distant moon affects a think-
ing substance which is here but at a point. And the statement
that the moon affects a thinking substance which is at a point
would seem to be less perplexing than the statement that it
affects a thinking substance which has no position at all. The al-
leged influence of one extended substance upon another extended
substance, distant from it, has the advantage of seeming some-
what analogous to the generally credited influence of the moon

175

Upon the tides, or to the generally credited influence of the sun
upon vegetation on the earth. Similarly, an alleged situation in
which there is a subject-object relation between a thinking sub-
stance which is extended and a substance which, although distant,
is likewise extended has the advantage of seeming somewhat analo-
gous to the generally credited situation in which two extended
substances have the relation of being distant from one another or
in which two extended substances, although distant, are like one
another.

We may, to be sure, wonder how any substance can influence
another, distant from it. We may wonder through what media a
distant entity comes to affect a substance that is characterized by
a mental attitude; and to what extent the mental attitude is due
to the media rather than to the distant entity itself. We may
likewise wonder how the mental attitude can, figuratively speak-
ing, reach to the distant substance and have it as an object. And
we may perhaps conclude that a mental attitude can not have a
distant entity as its object, that it either has no object at all or
only an object that is where it itself is. These however, are ques-
tions for subsequent chapters. In this chapter our question is
whether Descartes was thinking, not whether that thinking had
an object, much less how thinking and the distant entity, al-
leged to be the cause of that thinking, come to be related. What
we are considering is the alleged mental attitude of Descartes'
that seems to be directed upon man, God and the universe; or
that seems to be directed upon the moon. The mental attitude
of Descartes' that we are considering is presented with the char-
acteristic of seeming to be directed upon the moon, whether or
not it is presented in addition as having the real moon as its
object. In a later chapter the mental attitude, seeming to be
directed upon the moon, that is alleged to have the real moon
as its object, may be found to be presented as incredible. Or the
subsistent then found to be presented as incredible may be the
mental attitude, seeming to be directed upon the moon, that is
alleged to be directed only upon private content; or the mental
attitude, seeming to be directed upon the moon, that is alleged
to have no object at all. At this point the mental attitude under
consideration is presented without any claim as to what, if any-
thing, is its real cause or what, if anything, its real object. Our

176

subsistent is the mental attitude which pretends to be directed
upon the moon. And this subsistent presented with no claim as
to its real cause or real object need not, so far as we have yet
seen, be presented as incredible.

Our subsistent has been Descartes' mental attitude as he paced
up and down his stove-heated room and seemed to be thinking
about the moon or as he "seemed to ponder the existence of man,
God and the universe." 1S But it is no longer this alleged entity
presented as non-spatial which has been found to be unreal
that we have to consider; 19 nor is it this alleged entity presented
as having position, but not extension. 20 Our subsistent is Des-
cartes' mental attitude presented, not as itself an extended sub-
stance, but as the quality of an extended substance such as Des-
cartes' body or such as the brain, the cortex or a nerve-fibre within
Descartes' body. On the view which seems to remain before us for
our consideration, Descartes' body, or part of his body, has such
qualities as extension, weight and color, qualities which may be
called "non-mental." And it is to such qualities that our atten-
tion is directed when the substance in which these qualities inhere
is called "Descartes' body" or "Descartes' cortex." But the sub-
stance in which these qualities inhere is also, on the view which
we are examining, the substance in which Descartes' thinking
inheres as a quality. For Descartes' thinking, on this view, "is an
event and not a thing or stuff; and it is an event adjectival to the
brain." 21 In order to think of Descartes' brain or Descartes' body
as a substance in which not only non-mental qualities but also
thinking inheres, we must, says Sellars, "enlarge our conception
of a cerebral state over that which physiology gives." 22 And to
give recognition to the differing types of qualities which, on this
view, this substance has inhering in it, this substance may be
called, not Descartes' brain or Descartes' nerve-fibre, but rather
Descartes' mind-brain or Descartes' mind-nerve-fibre. It is this
mind-brain or mind-nerve-fibre which, on this view, thinks. And
since this mind-brain or mind-nerve-fibre is extended, Descartes'
thinking is concomitant with the quality of extension and may
to this extent be said itself to be extended.

When it is alleged that there is a substance to be called Des-
cartes' brain or Descartes' mind-brain, a substance in which think-
ing and extension inhere as qualities, there are various questions

177

that may be raised with respect to substances in general and with
respect to qualities in general. It may be asked what "substance"
means and what "quality" means. And it may be asked how a
substance can have qualities inhering in it, how a substance, for
example, can be thinking or can be extended. To discuss such
questions at this point would, however, carry us far afield. Des-
cartes' alleged mental attitude "presented as having a vehicle and
a setting can be discussed in fewer words and in a less complicated
fashion when, instead of regarding thinking, vehicle and setting
as all mere subsistents, we accept the premise that vehicle and
setting are real." 23 And Descartes 1 alleged mental attitude, pre-
sented as the quality of an extended substance, can be dis-
cussed in fewer words and in a less complicated fashion when
we assume that there are substances and that an instance of
thinking and an instance of extension can, if real, each be
the quality of a substance. "It is not as the painter putting on
the initial daubs of oil that we approach the canvas, but rather
as the painter-critic who concentrates his attention on minute
sections of his work in turn, at each point regarding the rest of
the work as unquestioned." 24 In our present discussion, let us
then make use of the fact that certain substances are listed as real
in the appendix to Chapter Three; and certain qualities. And let
us reserve for subsequent chapters discussions that deal with
substance in general rather than with Descartes* alleged mind-
brain or mind-nerve-fibre and discussions that deal with quality
in general rather than with Descartes' alleged mental attitude.
Let us in this chapter agree that Descartes' brain is a real sub-
stance and extension a real quality inhering in it. Indeed let us
agree that there are some qualities of the sort that are generally
called "secondary qualities." Let us agree that a certain piece of
metal is a real substance which is really hot and really red. And
let us agree that the electric bulb on the desk before me is really
bright and incandescent. Let us further agree that on some occa-
sion before our piece of metal was placed in a furnace, it was not
yet red; and that, after I have turned the switch, the bulb is no
longer incandescent. It is in some such fashion as this that the
alleged mental attitude of Descartes' that remains for our con-
sideration may be held to qualify the substance in which it in-
heres. Just as redness may be held to be a quality of the metal

178

which, before it was heated, was not red, and just as incandescence
may be held to be a quality of the bulb which, after I turn the
switch, is no longer incandescent, so Descartes' alleged thinking,
on the view that remains for our consideration, is presented as a
quality of an extended mind-brain or mind-nerve-fibre which in
some earlier phase may not have been thinking and in some later
phase may again not be thinking. So far as we have yet seen, the
mental attitude of Descartes' may be unreal that is alleged to be
the quality of an extended substance and alleged to be the quality
of a substance which in other phases is not thinking. But if this
subsistent is unreal, it is not unreal in so far as it is presented as
a quality concomitant with extension or in so far as it is presented
as the quality of one phase of a substance but not of another.

Descartes' alleged, mental attitude is not presented as incredible*
in so far as it is presented as a quality concomitant with extension.
For there are qualities concomitant with extension. But this
alleged mental attitude, it may be said, is a peculiar quality, un-
like redness or incandescence. And, it may be held, whereas in-
stances of redness and of extension inhering in the same substance
are plausible, alleged instances of thinking and of extension in-
hering in the same substance are not. Descartes' brain and its ex-
tension, it may be said, were objects for Descartes' contemporaries
but not for Descartes; whereas his mental attitude was an object
for him alone. With the piece of metal and its extension on the
one hand and that metal's redness on the other, or with the con-
cave side of an arc on the one hand and its convex side on the
other, the situation, it may be said, is different. For, both the
metal and its redness can be objects for the same observer. And
"when two percipients observe different sides of the same thing,
like the hasty knights in the fable, they can," as Ward says, 25
"change places and each connect the two aspects in one experience
of an object."

We have in effect agreed, however, that, if Descartes' alleged
mental attitude is real, it and the mind-brain ^in which it is al-
leged to inhere may be objects for the same observer. It is Des-
cartes' alleged mind-brain that you and I are now considering and
Descartes' alleged mental attitude that you and I are now discuss-
ing. 26 To this extent each of us may be said to connect a substance,
its extension and its alleged thinking in one experience, just as

179

we connect the metal, its extension and its redness in one experi-
ence, and just as each knight connects the two sides of the arc
in one experience. It may still be pointed out, however, that, even
if a substance, its extension and its alleged thinking are apparent
objects for the same observer, the substance and its extension
seem to be apprehended in one way, its alleged thinking in an-
other. By Descartes, it may be said, his thinking is sensed, his
brain and its extension inferred; by others, it may be said, Des-
cartes' brain and its extension may be sensed, but his thinking
must be inferred. But in spite of the fact that we see the metal and
its redness and do not see but feel its heat, we do not^eem to dis-
believe that the metal is both red and hot. And when we do not
feel the heat but only infer it from what we see, we likewise do
not seem to disbelieve that the metal is both red and hot. Qualities
are generally believed, to inhere in the same substance, even when
they are perceived through different senses, and even when one
is sensed and the other inferred. And an instance of thinking and
an instance of extension alleged to inhere in the same substance
need not be presented as incredible when they are presented as
being apprehended in different ways or when they are presented
as one being a sense-datum, the other an object which is inferred.
The mental attitude which is alleged to be a quality inhering
in a mind-brain or mind-nerve-fibre need not be presented as
incredible in that mental attitude and nerve-fibre are presented
as being apprehended in different ways. But such an alleged
mental attitude is presented as incredible, it may be said, in that
mental attitudes and nerve-fibres appear totally incommensurate
with one another. "If we know so little what we mean by a 'nerve-
process' that it may turn out ... to be an emotion or a tooth-ache,"
then, says J. B. Pratt, 27 "we have no business to use the term
'nerve-process' at all/* When, however, Descartes' thinking is pre-
sented as a quality of his nerve-process, it need not be presented as
itself the iferve process. In order for a nerve-process to have the
quality of thinking, the terms "thinking" and "nerve process"
need be no more synonymous than the terms "redness" and "piece
of metal" or the terms "incandescence" and "electric bulb." It
may be held, to be sure, that "thinking nerve-fibre" is a perplex-
ing combination of terms, that "thinking" and "nerve-fibre"
joined together seem to represent no apparent object at all. It is

180

however such an apparent object, such a subsistent, that we have
through several paragraphs been describing. Having eliminated
as unreal Descartes' thinking appearing with the characteristic
of being utterly non-spatial, Descartes' thinking appearing as
having position but not extension, and Descartes' thinking ap-
pearing as itself an extended substance, the subsistent remaining
for our consideration is: Descartes' mental attitude seemingly
directed towards man, God and the universe, a quality of, and
abstractable from, all or part of the breathing, reacting, extended
substance that may be called Descartes' mind-body. Although
perhaps perplexing, this subsistent, so far as we have yet seen, need
not be presented as generally discredited. So far as we have yet
seen, this alleged mental attitude of Descartes' may be real.

We have agreed that qualities exist; and substances in which
they inhere. Let us further agree that, when a quality inheres in
a substance, there is a sort of parallelism between them. When an
electric bulb is destroyed, its incandescence disappears. And when
its incandescence disappears, the electric bulb is different from
what it was before, if only in that it no longer has the quality of
incandescence. The view which we are considering, the view that
thinking is a quality of an extended mind-brain or mind-nerve-
fibre which thinks, has implicit in it the view that, as thinking
changes, there is some change in the extended substance which
thinks. The change in the substance may be merely the change
from a phase which has a given mental attitude inhering in it to
a phase which has no mental attitude or a different mental attitude
inhering in it. Or there may be held to be other qualities of this
substance, non-mental qualities, that change when its thinking
changes. If however a change in the thinking extended substance
parallels a change in its mental attitude, then a change in mental
attitude need be no more dependent on a change in the substance
in which it inheres than a change in that substance need be de-
pendent on a change in its mental attitude.

We may ask how an outside stimulus causes both a change in
thinking and a change in the substance which thinks. And we may
ask whether thinking and certain non-mental qualities inhering
in the same substance change together. But the view which we are
considering involves no epiphenomenalism. On the view which we
are considering, there may be changes in non-mental qualities just

181

prior to a change in mental attitude; so that an examination of
non-mental qualities may enable us to predict mental attitudes.
There may likewise be a change in mental attitude just prior to
changes in certain non-mental qualities; so that an examination of
mental attitudes may enable us to predict non-mental qualities.
But in so far as mental attitude, substance and non-mental quali-
ties change simultaneously, they are, on the view which we are
considering, presented as interdependent. "Take away the neural
process/' says Hodgson, 28 "and there is no sensation. Take away
the sensation it can not be done save by taking away the neural
process. There is therefore," he continues, "dependence of the sen-
sation on the concomitant neural process but not vice-versa." But
if thinking and neural process are concomitant, we do not take
away the neural process and then take away the thinking. If we
take away the neural process, we take away simultaneously the
thinking and whatever non-mental qualities inhere in this sub-
stance. If we take away, not the substance, but those of its non-
mental qualities, if any, that occur only when its thinking occurs,
we take away its thinking. Similarly, however, if we take away its
thinking, we take away those of the substance's non-mental quali-
ties that occur only when thinking occurs; and we change the
substance in which both they and the thinking concomitant with
them formerly inhered.

Even if we disregard those non-mental qualities inhering in the
thinking substance that may be alleged to change before or after
there is a change in mental attitude, even if we restrict our atten-
tion to those non-mental qualities, if any, in which a change is
alleged to occur simultaneously with a change in mental attitude,
we may, to be sure, find it convenient to explore what happens
to non-mental qualities more intensively than we explore what
happens to thinking. Let us assume that, when light disappears
from an electric bulb, the bulb simultaneously ceases to have an
electric current running through it. Let us assume that the qual-
ity of being lighted and the quality of being affected by an
electric current are interdependent; that the occurrence of the
quality of being lighted does not precede, and does not enable us
to predict, a subsequent occurrence of the quality of being affected
by an electric current; and that the occurrence of the quality of
being affected by an electric current does not precede, and does

182

not enable us to predict, a subsequent occurrence of the quality
of being lighted. We may nevertheless, it would seem, find it con-
venient to explore the onset and disappearance of electric cur-
rents more intensively than we explore the onset and disappear-
ance of the quality of being lighted, a quality that is concomitant
with it. Such priority, however, as under these circumstances we
might give to the quality of being affected by an electric current
over the quality of being lighted would be a priority in attention
and would not imply that one quality is temporally prior to the
other or that one quality is real and the other unreal. In a
similar fashion, it would seem, priority in attention may be given
to certain non-mental qualities of a thinking, extended substance
rather than to the mental attitude which is alleged to vary with
them. Descartes' mental attitude seemingly directed upon man,
God and the universe may be presented as the quality of an ex-
tended thinking substance. And other qualities of this substance,
non-mental qualities, may be presented as varying with this
attitude, as being present when it is present, absent when it is
absent. But these alleged other qualities may be presented as be-
ing more promising to investigate without being presented as
being temporally prior to the mental attitude seemingly directed
upon man, God and the universe. And they may be presented as
more promising to investigate without this alleged mental attitude
being presented as unreal.

Certain non-mental qualities, let us agree, offer a more fruit-
ful field for investigation than the mental attitudes which, if they
exist, are concomitant with them. These non-mental qualities
along with others, which all together may be said to constitute an
organism's behavior, have been the subject of much study on the
part of behaviorists. Organisms have been confronted by various
stimuli and the organisms' responses noted. "The desire in all
such work/' says Watson, 29 "is to gain an accurate knowledge of
adjustments and the stimuli calling them forth. The reason for
this is to learn general and particular methods by which behavior
may be controlled. The goal is not the description and explana-
tion of conscious states as such." As a result of such work, how-
ever, one may come to hold that we can disregard mental atti-
tudes even if they exist, that information with respect to be-
havior alone will teach us all that we can know that might en-

183

able us to predict and control what organisms will do. If a men-
tal attitude exists and is concomitant with some non-mental qual-
ity, a given stimulus may be said to bring about both the non-
mental quality and the mental attitude; and a given response may
be said to be due both to the non-mental quality and the mental
attitude. But if a study of the causal relation from stimulus to
mental attitude to response gives us no ability to predict and con-
trol not given us by a study of the causal relation from stimulus
to non-mental quality to response, then the alleged mental atti-
tude, it may be said, is, like LaPlace's God, an unnecessary
hypothesis.

The mental attitude which is alleged to be the quality of an
extended mind-brain or mind-nerve-fibre need not be presented
as incredible, we have seen, when it is presented as less promising
to investigate than the non-mental qualities with which it is
alleged to be concomitant. But is it not presented as incredible
when it is presented as unnecessary for prediction and control?
We may imagine two worlds before us, only one of which is the
world of real entities. In the one, an organism is stimulated; the
stimuli bring about non-mental qualities in the brain; and these
non-mental qualities lead the organism to make characteristic re-
sponses. In this imagined world, however, organisms are like
robots; there are no mental attitudes. In the other, organisms
behave just as they behave in the world just described. But, in-
tervening between stimulus and response there are not only the
brain's non-mental qualities, but the brain's mental attitudes,
its thinking, as well. Entities, according to the dictum attributed
to William of Occam, are not to be multiplied beyond what are
necessary. Admitting, then, that mental attitudes, if they exist,
are not needed to enable us to predict and control what organisms
will do, should we not accept the world with fewer entities and
reject the other? In view of its being presented as not needed, do
we not find the mental attitude alleged to be an additional quality
inhering in an extended brain or extended nerve-fibre presented
as incredible and as generally disbelieved?

Let us first remark that, whereas one writer may hold that en-
tities are not to be multiplied beyond what are necessary, an-
other may hold that all entities are real that can be real, all en-
tities, that is to say, that are not inconsistent with some entity

184

that is real. Both assertions ascribe characteristics to what is taken
to be "reality." Each assertion, we hold, depends for its truth upon
the signification that is assigned the term "real." As we have
explained our term "reality," the world o real entities, so far as
we have yet seen, need be neither a world with the maximum
number of compossible entities nor a world with the minimum
number needed for prediction and control. As an element in the
explanation of our term "reality," we have, however, said that
"whatever explicitly or implicitly appears as generally discredited
is unreal." 30 Consequently, it remains for us to determine whether
an entity may be presented as not needed for prediction and con-
trol; and yet not be presented as generally discredited.

There is a distinction to be made, let us suggest, between the
entity which is proposed in order that we may organize our
knowledge, in order that the facts that we know may be known to
be related, or in order that we may predict and control future
events, and the entity which is not proposed with the purpose of
accomplishing any of these objectives. When facts are puzzling and
hypotheses proposed in order that we may become aware of re-
lations between these facts, it would seem that, on the whole,
we accept the simpler hypothesis, the hypothesis which introduces
and proposes fewer entities; and that we reject the more compli-
cated hypothesis, the hypothesis which introduces and proposes a
greater number of entities. When there are similar objectives
and when alternative hypotheses are proposed, we likewise seem
on the whole to accept the hypothesis which accounts for a large
number of facts and to reject the hypothesis which accounts for a
lesser number of facts. But these observations do not apply to the
entity that is not introduced in order that we may predict and con-
trol, not introduced in order that we may become aware of other
entities as mutually related. An alleged God may be proposed, not
after miracles have been experienced and found puzzling, but as
an entity that is itself experienced. The suggestion that my electric
bulb is bright may be made, not to suggest a cause for the waves
that travel out from the bulb, the electric current running
through the bulb may already have been accepted as such a cause,
but the suggestion may be made on the basis of other evidence,
on the basis of independent belief. Whatever may be the situation
with respect to entities that are proposed in order that we may

185

become aware of other entities as mutually related, entities that
are not proposed and introduced with such a purpose need not, we
hold, be presented as incredible when they are alleged not to be
needed in order that we may become aware of other entities as
mutually related. The alleged brightness of my electric bulb need
not be presented as incredible when it is alleged not to be needed
in order that we may become aware of a cause of the waves travel-
ling out from the bulb. And the mental attitude that is alleged to
be a quality of an extended substance need not be presented as in-
credible when it is alleged not to be needed to enable us to pre-
dict and control the organism's responses.

The subsistent that seems to be before us is Descartes' mental
attitude seemingly directed upon man, God and the universe, a
mental attitude which is alleged to be the quality of an extended
mind-brain or extended mind-nerve-fibre. It is a mental attitude
which, along with such concomitant non-mental qualities as vary
with it, is alleged to be a result of certain stimuli and a cause of
certain responses; a mental attitude, nevertheless, which is alleged
to offer a less promising field for investigation than the non-men-
tal qualities which accompany it; and a mental attitude which is
alleged not to have been proposed in order that we might be able
to predict and control Descartes' responses. Such an alleged men-
tal attitude need not be presented as incredible. And yet, since
there are certain behaviorists who reject it, this alleged mental
attitude is presented as being in some quarters disbelieved.

To what extent, however, do behaviorists disbelieve in the par-
ticular subsistent that we are considering? A behaviorist may assert
that there are no entities which lack position altogether. He may
assert that he disbelieves in an entity which is alleged not to in-
here in any extended substance. He may hold that nothing exists
outside what we have called "total behavior." 81 And he may sum
up his position by stating that thinking is behavior. The state-
ment, however, that thinking is behavior may not be incon-
sistent with the statement that thinking and non-mental qualities
inhere in the same substance, with the statement that thinking and
certain non-mental qualities which are concomitant with that
thinking vary together. And the statement that there is no ob-
servable object outside total behavior may not be inconsistent
with the statement that, "in having Descartes before us as an un-

186

analyzed whole, his alleged thinking is within, rather than out-
side, the entity before us." 32 There are, let us agree, behaviorists
who seem to disbelieve in the subsistent that appears to be before
us. There are behaviorists who find that, when they attempt to
abstract thinking, mental attitude, mental activity, from Des-
cartes' mind-brain or mind-nerve-fibre or total behavior, there is
an irruption of disbelief similar to that which breaks in upon us
when we attempt to abstract from this rectangular desk its al-
leged roundness. To a considerable extent, however, it is some
other subsistent, and not the subsistent which we are considering,
that seems to be the object of their disbelief. The subsistent which
we are considering is presented as seemingly disbelieved by some
behaviorists, but not as generally disbelieved by behaviorists.

Some behaviorists seem to disbelieve in the alleged mental
attitude which we are considering. And some epistemologists who
assert the existence of ideas seem likewise to disbelieve in this
subsistent. Such epistemologists may agree that, in addition to
Descartes' non-mental behavior, there is a real mental entity to be
abstracted or to be inferred from his mind-brain or from his mind-
nerve-fibre or from his total behavior. But they may hold that
whatever mental entity is thus to be really abstracted or inferred
is what they would call "content" or "idea," not what they would
call "mental attitude" or "mental activity" or "thinking." Just,
however, as the behaviorist who asserts that thinking is behavior
may not disbelieve in the alleged mental attitude which we are
considering, so the epistemologist who denies the existence of
mental entities other than what he calls "ideas" may not dis-
believe in the alleged mental attitude which we are considering.
For "what we call 'mental attitude' may by some be included in
what they would call 'private content.'" 33 To be sure, what we
call "mental attitude" has not been presented as private content,
as an object for Descartes alone. 3 * But it has been presented as an
entity that may be held to be sensed by Descartes alone. And
whereas what we call mental attitude has been presented as
think-mg rather than as what is thought, it has been presented as
an entity that may be held to be an object "upon which it or some
further mental attitude is directed." 38 The epistemologist who
holds that there are no mental entities that are not pictures or
images disbelieves, we may say, in what we are calling "mental

187

attitudes." But what is alleged to be an idea and not alleged to be
a picture or image may be what we should call a mental attitude
presented as an object for some further mental attitude.

The alleged mental attitude which we are considering is pre-
sented as seemingly disbelieved by some behaviorists, but not as
generally disbelieved by behaviorists. And it is presented as
seemingly disbelieved by some epistemologists who assert the
existence of ideas, but not as generally disbelieved by epistemolo-
gists who assert the existence of ideas. Indeed when we turn from
the opinions of behaviorists and epistemologists to the opinions of
men generally, we seem to note a general belief that men are
not robots and that their mental life is not made up of pictures
and images. In addition to the -words "idea" and "thought/* there
are in common use the words "thinker" and "thinking"; and the
statement that there are thinkers who think would seem to ex-
press a belief in entities that are not pictures or images but are
rather what we in this chapter have called mental attitudes. In
any case the particular mental attitude which we have been con-
sidering, and which we are now considering, is not presented as
generally discredited. And this alleged mental attitude is listed as
real in the appendix to Chapter Three. As Descartes paced up and
down his stove-heated room, he was, we conclude, thinking. He
had a mental attitude which seemed to be directed upon man, God
and the universe, and which was a quality inhering, along with
extension and other non-mental qualities, in his mind-brain or
mind-cortex or mind-nerve-fibre.

Summary

Positive statements about what exists in our sense of "existence"
should, according to our program, be statements about individual
subsistents carefully identified. In this chapter we select an alleged
instance of thinking what we call a "mental attitude" distin-
guishing it from object, from mental content and from non-
mental behavior.

Such an instance of thinking may be presented as spatial or as
non-spatial. Presented as non-spfettial it is unreal. Presented as

188

spatial it may be real; for the arguments which have been ad-
vanced against spatial and extended thinking are unconvincing.

The entity we present is an instance of thinking that is a quality
of an extended substance. Even though this entity is presented as
something that need not be considered in investigations into be-
havior, it is not presented as generally discredited and is real.

Share this with your friends