Geach, Peter (1954) - Form and Existence

Form and Existence
Author(s): P. T. Geach
Source: Proceedings of the Aristotelian Society, New Series, Vol. 55 (1954 - 1955), pp. 251-272
Published by: Wiley on behalf of The Aristotelian Society
Stable URL: http://www.jstor.org/stable/4544548
Accessed: 06-05-2015 16:36 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
The Aristotelian Society and Wiley are collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the
Aristotelian Society.
http://www.jstor.org
This content downloaded from 128.103.149.52 on Wed, 06 May 2015 16:36:56 UTC
All use subject to JSTOR Terms and Conditions
Meetingof theAristotelianSocietyat 21, BedfordSquare,London,

W.C.I, on May 9th, 1955, at 7.30 p.m.
XI.-FORM
AND EXISTENCE.
By P. T. GEACH.
In this paper I shall discuss what Aquinas meant by
his term esse, or actusessendi,"act of existing ". Another
synonym that he uses-quo aliquid est, " that by which a
thing is (or: exists) "-suggests a convenient division of
the subject: we can first discuss Aquinas's philosophical
use of quo, " that by which ", and then consider which
sense of est, which sort of existential statements, may be
relevant to Aquinas's doctrine of esse. But we shall see that,
having got thus far, we cannot arrive at the meaning of
the whole phrase quoaliquidest, or the reasons for the way
Aquinas uses it, simply by combining our separate considerations about quoand est.
I.
Beginning with Aquinas's use of quo brings a great
immediate advantage. The predicate est," is " or " exists",
is at least a peculiar one, and many people would deny
that it is properly a predicate at all; but Aquinas uses quo
not only with est but also with unexceptionable predicates.
In this use, quofollowed by a noun- ubject and an (ordinary)
predicate is synonymous with the phrase formed by the
abstract noun answering to the predicate followed by the
genitive of the noun that was subject; quoSocratesalbusest
is synonymous with albedoSocratis,and so on. Either kind of
phrase is thus used in order to designate what Aquinas calls
Forms; to understand his use of quowe must exaniine his
notion of forms, which moreover is intimately connected
in other ways with his doctrne of esse.
For Aquinas, the real distinction between a form and
thesself-subsistent individual (suppositum)whose form it is
comes out in the logical distinction between subject and
2E
252
P.
T.
GEACH.
predicate. (Ia q. 13 art. 12; q. 85 art. 5 ad 3 um). I

think this is the way to introduce his notion of form to
modern philosophers. There are, however, strong prejudices against allowing that this logical distinction answers
to any real distinction. One such obstacle is the old twoname or identity theory of predication, which flourished in
the Middle Ages, and still keeps on appearing in new guises:
the theory that a true predication is effected by joining
different names of the same thing or things, the copula
being a sign of this real identity. I shall not waste time on
this logically worthless theory. Anybody who is tempted
by it may try his hand at explaining in terms of it how we
can fit together the three terms " David " " father ", and
" Solomon " (which on this theory are three names) to
form the true predication " David is the father of Solomon ".
The futility of the two-name theory comes out clearly at
the beginning of Lewis Carroll'sGameof Logic. Lewis Carroll
professes to find a difficulty over saying " some pigs are
pink "; as it stands, this suggests an impossible identity
between certain things (pigs) and a certain attribute
(signified by " pink ") ! He seeks to remove this difficulty
by expounding the statement as meaning " some pigs are
pink pigs ", where " are " signifies real identity. But " pink
pigs " means "pigs that are pink ", and there is as much
or as little difficulty about this phrase as about the predication " pigs are pink " at which he stumbles.
What is right in the two-name theory is that it is
possible to state the truth-condition of an affirmative predication as an identity of reference between two names;
" a man runs ", let us say, is true if and only if " man
and " thing that runs " are two names of the same individual. Aquinas uses this way of stating truth-conditions
quite often, and has in consequence been wrongly regarded
as holding the two-name theory. But it is not the name
" thing that runs " that is used in the sentence " a man
runs
',
but the predicate " runs " from which this name is
formed; and " runs " and " thing that runs " are by no
means synonymous; the relation between their modisignificandi in fact raises over again the same problem as the
253
FORM AND EXISTENCE.
relation of subject and predicate, a problem that is thus

merely shifted by expressingthe truth-conditionsof predication in terms of identity of reference.
Modern philosphers have pretty generally abandoned
the two-name theory; at least to the extent of admitting
that a logical subject and a predicate have radically different
modisignificandi. But need we admit also a difference of
type as regards the res significatae? Surely what distinguishes a predicate from a name is just the fact that it
does not nameanything, but is rather true or false of things;
a true predication is one in which the predicate is trueof what
the subject names. Are we not blurring this distinction if
we say that predicates stand for (why not, that they name?)
a type of entity other than that which names stand for ?
Have not philosophers said the queerest things about the
entities that predicates are supposed to stand for ? No
wonder; such paradoxes are bound to arise if you treat
as a name what is not a name; like the parodoxes about
Nobody in Throughthe Looking-glass.
But, whatever difficulties it may involve, I think we
have to allow that logical predicates do stand for something,
as well as being true or false of things. For when a question
howmanyis asked and answered, we can surely ask: To what
is this manyness being ascribed ? And in any concrete
instance we shall find that in asserting manyness we use a
logically predicative word or phrase, and are ascribing
manyness to what this stands for. " How many ducks
are swimming in the Chamberlain Fountain ? " " Three ".
If this answer were true, there would be objects of which

" duck swimming in the Chamberlain Fountain " could
truly be predicated; and my number-statement is about
what this predicate stands for. You cannot say my statement
ascribes a property (threeness) to a certain set of individuals
-the ducks swimming in the Chamberlain Fountain; for
(so far as I know) no ducks are swimming there. And it
makes no difference whether there are any or not; for
the sense of the question how many such ducks there are
cannot depend on what the right answer to the question is;
and the answer " three " cannot be taken as a predication
2E
254
P.
T.
GEACH.
about a set of ducksunless the question answeredis a question

about them; but the question how many such ducks there
are admitsof the answer "0 ", and noughtishness certainly
is not here being taken as a property of any set of ducks.
You may indeed rightly say that my statement is about ducks
(not: the ducks) swimming in the Chamberlain Fountain.
The omission of the definite article is here significant. A
statement that could rightly be called " a statement about
the ducks (etc.) " would have to refer (or at least profess to
refer) to a certain set of ducks; but in speaking of " a
statement about ducks (etc.) " I am not implying that the
statement mentions any individual ducks, but on the contrary that it is about what the predicate " duck swimming
in the Chamberlain Fountain " stands for.
It is only to what such a predicative expression stands
for that we can even falsely ascribemanyness. It is nonsense,
unintelligible, not just false, to ascribe manyness to an
individual; what can be repeated is always and only a
caderepluralitas
common nature. Non enimpotestin intellectum
huiusindividui(Ia. q. 13 art. 9) Because, for Aquinas, forms
are what answer in rebusto logical predicates, it is consistent
for him to say that forms are as such multipliable (Ia. q.
3 art. 2 ad 3 um: q. 7 art. 1).
I cannot help being reminded here of the very similar
language that we find in Frege. Frege, like Aquinas, held
that there was a fundamental distinction in rebusanswering
to the logical distinction between subject and predicatethe distinction between Gegenstand(object) and Begrif
(concept). (In using Begrif as a term for what logical
predicates represent, Frege was not accepting any form of
conceptualism; on the contrary, he explicitly denies that the
Begrif is any creature of the human mind-it is, he says,
" objective ".) And for Frege the Begrif, and it alone,
admits of repetition and manyness; an object cannot be
vor.
repeated-kommt nie wiederholt
Understood in this way, the distinction between individual and form is absolutely sharp and rigid; what can
sensibly be said of one becomes nonsense if we try to say it
of the other. (Aquinas's " subsistent forms " might seem
FORM AND EXISTENCE.
255
to bridge the gulf; but, as we shall presently see, they do

not, nor did Aquinas really think they do.) Just because of
this sharp distinction, we must reject the Platonic doctrine
that what a predicate stands for is some single entity over
On the
against its many instances, E'V Er' 7roXXcov.
contrary: the common nature that the predicate " man "
(say) stands for can be indifferently one or many, and
neither oneness nor manyness is a mark or note of human
nature itself. This point is very clearly made by Aquinas
in De Enteet Essentia. Again we find Frege echoing Aquinas;
Frege counts oneness or manyness (as the case may be)
among the properties (Eigenschaften)of a concept, which
means that it cannot at the same time be one of the marks
or notes (Merkmalen)of that concept. (Frege's choice of
words is here unfortunate; his saying that it is an Eigenschaft
e.g., of human nature to be found in many individuals has
led people to suppose that he regards this as a propriumof
human nature-although so to read him makes complete
nonsense of his distinction. Aquinas's saying that oneness or
manyness is incidental (accidit)to human nature is a much
clearer expression.)
The Platonic mistake about the nature of forms goes
with a liberal use of what we may class together as abstractsingularexpressionslike av7-o something-or-otherin Greek,
or " X-ness " or " the attribute of being X "; these
expressions are not just grammatically but also logically,
argumentatively handled as though they were proper
names. I do not say that such abstract expressionslooking
like proper names should be totally banned; it would make
things very difficult for philosophers. (I myself used
" human nature " in this way in the last paragraph.) But
I do say that anyone who uses them ought to be ready to
replace them on demand by use of the concrete predicates
from which they are derived. (Thus: for " neither oneness
nor manyness is a mark of human nature itself" read
" whether there is one man or many men is irrelevant to
what X must be if X is a man ", or something like that.)
Sometimes this replacement is stylistically better, sometimes
not. But it must be possible; a sentence with an irreducible
256
P. T. GEACH.
abstract " proper name " in it (say: " Redness is an eternal

object ") is nonsense.
All the same, Platonism of this sort is a very great
temptation; and I think it is instructive to watch Frege's
unsuccessfulstruggles against temptation in his paper Ueber
Begriff und Gegenstand(" On Concept and Object ".) I
quote: (op. cit. p. 197): " In logical discussionsone quite
often needs to assert something about a concept, and to
express this in the form usual for such assertions viz. to
make what is asserted of the concept into the content of the
grammatical predicate. Consequently, one would expect
that the reference of the grammatical subject would be the
concept; but the concept as such cannot play this part, in
view of its predicative nature; it must first be converted
into an object, or, speaking more precisely, represented by
an object. We designate this object by prefixing the
words 'the concept': e.g.
'The concept manis not empty'.
Here the first three words are to be regarded as a proper
name ". And later on (p. 198): " In my way of speaking
expressionslike ' the concept F ' designate not concepts but
objects ".
Of course Frege has gone astray here; he does not clear
himself of the charge of having made a concept into an
object just by saying that " the concept man" does not
stand for a concept but for an object that " represents" a
concept; no more than a writer can escape the charge of
vulgarity by a parenthetical " to use a vulgarism ". But
how then are we to get into the subject of predication a
direct reference to what Frege calls a concept and Aquinas
calls a form ? I think the solution, the way to avoid the
mistake of Platonism, is that an abstract noun (or nounphrase) referring to the form can indeed occupy the place
of the subject, but cannot be the whole of the subject; the
form being signified, in rectoas Aquinas would say, by an
abstract noun, we must add a mention in obliquoof the
individual whose form it is; " the wisdom of Socrates"
and " the redness of Socrates's nose " give us designations
FORM AND EXISTENCE.
257
of forms, the spurious proper names "wisdom" and

" redness " do not. We must not construe " the wisdom of
Socrates " as " wisdom, which Socrates possesses": apart
from lapsing into the Platonic error of taking " wisdom " as
a singular term, we should run into notorious antinomies
about the relation supposedly meant by " possesses".
" Of" in " the wisdom of Socrates " does not signify a
special relation, as in such phrases as " the statue of
Socrates " or " the shield of Socrates ". The statue is of
Socrates by being related to him in one way, the shield is of
Socrates by being related to him in another way; but if we
start saying " and the wisdom is
of
Socrates . . . " we have
already gone wrong, for logically " the wisdom of Socrates "
does not split up into " the wisdom " and " of Socrates "
(sc. " that wisdom which is of Socrates ") but into " the
wisdom of. . . " and "Socrates ". What refers to a form is
" the wisdom of . . . ", not the whole phrase " the wisdom
of Socrates "; " the wisdom of . . . " needs to be completed
with a name of something that has the form, just as the

predicate " . . . is wise ", which also stands for this form,
needs to be completed by a subject. " Of" is a logically inseparable part of the sign " the wisdom of . . . ", indicating
the need to put a name after this sign; and this need is
what makes the sign suitable to express a form, since a
form, as Aquinas says, is more properly termed entis than
ens. (Ia. q. 45 art. 4.) The linguistic oddity of the division
into " the wisdom of . . . " and " Socrates "-a
division
that cannot be made in Latin at all-is quite trivial and

accidental; in Hebrew, for example, such a division would
be perfectly natural, since "the wisdom of Socrates "
would be rendered by inflecting the word for " wisdom "
and leaving the name " Socrates " uninflected.
It may be asked: How cana form be designated bothby
a logical predicate like " . . . is wise " and by an expression
like " the wisdom of . . " ? These sorts of expressions are
certainly not interchangeable; but I think we can show

that the difference between them is only secundummodum
significandi, not secundumrem significatam. When we want
to mention a form for the sake of expressingthe supposition
258
P. T. GEACH.
that in a given individual that form is found, we refer to it

by an expression which together with the name of that individual forms a statement-i.e., by a logical predicate like
" is wise ". But when we want to talk directly about the
form itself, to get a reference to the form into the subjectplace in our statement, then we need to refer to the form by.
an expression which, together with a reference in obliquoa
to that in which the form is found, will compose a complex
name that can be logical subject-e.g., the expression " the
wisdom of..."
It admittedly looks queer that a form cannot be designated either (say) by " redness " alone or by " the redness
of Socrates's nose ", but only by " the redness of . . . "
(understood to be followed by some name or other). We
may help our understanding by the analogous case of
functions in mathematics: neither the isolated square-root
sign nor (say) " V25 " designates a function, but rather
the circumstance that the square-root sign is followed by
some numeral or other.
This comparison of forms to functions in mathematics
seems to me very useful; it was suggested to me by Frege's
assimilation of his Begriffe to mathematical functions.
(Frege indeed held that Begriffearea sort of functions, which
can take only two values, the " true " value and the " false "
value; I shall not try to explain here this part of his doctrine,
and would emphasize that I am asserting no more than
an analogy between forms and functions.) It may for
instance help us to see how " of " in " the wisdom of
Socrates " does not stand for a special relation of " inherence in " or " belonging to ". Somebody ignorant of
mathematics might take " the square root of 25 " to mean
" that one among square roots which belongs to 25," and
wonder how one number can " belong" to another; but
somebody who understands the term " square root " can
already understand " square root of 25 " and can see that
a relation of " belonging " does not come in. So also somebody who understands the term " wisdom ", and does not
Platonically misunderstand it as a proper name, already
understands the make-up of the phrase " wisdom of
FORM AND EXISTENCE.
259
Socrates " without needing to understand the mythical relation of inherence. The term " wisdom ", like the term
" square root ", of itself demands a genitive to complete its
sense.
Again, in the sense of the word " form " that we have
so far used, what the phrase " the wisdom of Socrates"
designates is not a form simpliciter;but can legitimately be
said to be a form of Socrates, a form occurring in Socrates.
This too can be elucidated by our mathematical analogy.
Consider the square-root function: we cannot say that the
number 5 is that function or any other function: but we
can say that the number 5 is that function of the number 25.
So also the wisdom of Socrates is not a form simpliciterbut
is a form of or in Socrates. What designates a form is not
the whole phrase "the wisdom of Socrates ", but merely
"the wisdom of .
. "-although
without completion this
latter phrase is senseless.

Now at this point I think Aquinas's terminology is
defective. He emphatically rejects the Platonic error of
taking " wisdom " as the proper name of a form; he says,
e.g., that when " Dionysius " speaks of per se vita and per se
sapientiahe is not meaning quasdamsubsistentes
res. (Ia. q. 44
art. 3 ad 4 um). But he often uses the term " form " for
what is referred to by a phrase like sapientiaSocratis;
whereas if we keep to the sense of" form " that we have so
far used, a sense that is also to be found in Aquinas, what
designates a form is rather sapientiatogether with the genitive
inflexion of the following noun. The syntax of Latin impedes
clear statement of this point, and Aquinas could not, of
course, use our mathematical analogy for elucidation.
I am here leaving myself open to the charge of developing
a new doctrine of form and departing from Aquinas's own
doctrine. But if we will not impute to Aquinas a slight and
natural inexactness of language at this point, then his
doctrine of form becomes mere inconsistency and nonsense.
For " the wisdom of Socrates" must designate something
individual, something that is no more multipliable or
expressible by a predicate than Socrates himself is; so if
this is taken to be a typical designation of a form, the whole
260
P. T. GEACH.
distinction between form and individual is undone. To keep

our heads clear, we must distinguish two senses of " form "
in Aquinas-the form that is the reference of the predicate
". . is wise " or of the phrase " the wisdom of . . . ", and the
form that is the reference of a phrase like " the wisdom of
Socrates ". In the latter case I shall speak of individualized
forms; individualized forms will then not be forms
simpliciter(just as a number that is a certain function of
another number is not a function simpliciter)and will always
be referred to by the full style " individualized forms ".
This distinction is needed in order to make Aquinas's
doctrine of subsistent or separate forms logically intelligible.
When Aquinas says things like Deusest ipsasapientia,he is not
meaning that God is that of which the noun " wisdom " is a
proper name; for the Platonists are wrong in thinking that
there is such an object, and Aquinas says that they are wrong.
But we can take it to mean that " God " and " the wisdom
of God " are two names of the same thing; and this interpretation does not make Aquinas guilty of the impossible
and nonsensical attempt to bridge the distinction previously
expounded between form and individual, or find something
intermediate. For we can significantly say that " God"
and " the wisdom of God " and " the power of God " are
three names with the same reference; but "the wisdom
of . . . " and " the power of . . . " have not the same reference,
any more than the predicates "wise" and " powerful" have.
Non dicimusquod attributumpotentiaesit attributumscientiae,
licetdicamusquodscientia(Dei) sit potentia(Dei) (Ia. q. 32 art.
3 ad 3 um).
This account, it may be argued, departs fatally from
Aquinas's mind, in that it makes out that in " the wisdom
of God " and " the power of God " " the wisdom of" and
" the power of " differ in reference from the name " God "
and from one another; for this conflicts with Aquinas's
teaching on the divine simplicity. But what we signify by
" the wisdom of " and " the power of " is really distinct
-otherwise " the wisdom of Socrates " would coincide in
reference with " the power of Socrates "-and on the other
hand what we signify by " the wisdom of " is not, and is
FORM AND EXISTENCE.
261
not signified as being, a part or element of what we signify

by " the wisdom of God " i.e. (according to Aquinas) God
himself. If one designation is part of another, it does not
follow that the things designated are respectively part and
whole. " The square root of 25" is a complex designation
having as parts the designations " the square root of " and
" 25 "; but it does not follow (as Frege oddly inferred) that
there is some sense of " part " in which we may suitably say
that the number 5, which is the square root of 25, has two
heterogeneous parts-the square-root function and the
number 25. Again-to get an analogy to the three designations " God ", " the wisdom of God ", " the power of
God "-the square and the cube are quite distinct functions,
but " 1 " and "the square of I " and the " the cube of 1 " all
designate the same number, and there is no distinction even

in thought between the 1 that is the square and the 1 that
is the cube and the I that is squared and cubed.
The supposition that x itself is that by (or in virtue of)
which x is F is certainly not logically absurd on the face of it
(as would be the supposition that x was identical with the
reference of the predicate "F", an individual with a
form). It will be instructive to consider an example from
modern philosophy in which the assertion of identity
between x and what makes x to be F might seem plausible,
and to see why such identity must be here denied. The
red-patch school of philosophy might well wish to say that
a red patch in a sense-field neither is the red-patchiness of
some other individual (of a suppositum) nor yet has any
individualized forms distinct from itself. That in virtue of
which the patch is red, that in virtue of which it is so big, that
in virtue of which it is square, is nothing other than the patch
itself, which is red, so big, and square.
But now let us observe our red square; after a short time
it begins a gradual change, and the outer region of the
square becomes green, while an inner region remains
red and is circular in shape. Are we then to say " The red
was square and larger, and has become circular and
smaller ", or rather "The square was all red and has
become partly green 5 ? Plainly we may say either-or
262
P. T. GEACH.
rather both. But now the individual red and the individual
square appear as distinct individual things, each with its
own distinct persistingidentity; there is no third individual,
non est ens. I do not
the red square; rubrumet quadratum
mean, of course, that there cannot be a red square object;
but then the persistent identity of this object will not
consist in any simple sensible character like redness or
squareness. It is because the philosophers' red square is
supposed to have no characteristics except a few simple
sensible ones that it falls to pieces under examination; for to
each such characterthere answersa distinct individual reality,
an individualized form-an individual extension, colour,
shape, etc.-and though these are united in one suppositum
x as the redness of x, the squarenessof x, etc., they are not
all one thing, the red patch; the red patch has no identity of
its own, nonest vereunumnecvereens.
These examples of individualizedformsmay be objectionable to some people as factitious philosophical examples.
But a wave, for example, is an identifiable individual that
can move locally (although Professor Prichard knew that
it was nonsense to say so); and this is certainly an
individualized form-it is that by which a certain body of
water is in a certain shape over part of its surface.
II.
"
The expression that by which the individual x is (or:
exists) " is senseless unless there is a sense in which " is "
or " exists " is properly predicable of individuals. Now

many modern philosophers would deny that there is such
a sense : " exists " is not a predicate ! This maxim is often
glibly and thouhgtlesslyused, but it has a serioususe; it is an
attempt to resolve the paradox of reference that arises over
the denial of existence. For we can significantly and truly
deny existence; indeed I should hold that there is no
sense of " exists " for which we do not get significant and
true denials of existence. (Some philosophershave thought
otherwise-that for a certain sense of the verb " to be " "A
is " or " A s are " is a form of statement that is always
true; even dragons are, even round squares are, in this
FORM AND EXISTENCE.
263
sense ! But at any rate to a discussion of Aquinas this

supposed sense is irrelevant; we may stick to the everyday
senses of " to be " or " to exist " for which existence can
be truly denied.) How is such denial possible? It might
look as though "A is not " or "A does not exist " were
never true; for if it were, the subject-term"A" would fail to
have reference, and so no statement would have been made
at all, let alone a true statement.
We can get out of this difficulty by denying that in
"A is not " or "A does not exist " the verb " is " or " exists'"
is a logical predicate. For since" subject " and " predicate "
are correlatives, this is tantamount to denying that the
grammatical subject "A" is a logical subject. And from
this again it follows that the statement "A does not exist " is
not really about what the subject "A" ostensibly stands for;
so in making the statement we do not fall into the absurdity
of using "A" as though it stood for something and then
in effect denying that it does so.
But though saying that " exists " or " is " is not a genuine
logical predicate of individuals would thus resolve the
paradox of non-existence, this does not prove that it is not,
sometimes at least, a genuine logical predicate; for the
paradox might be resoluble some other way. Moreover, it
is not enough to say that in "A does not exist " "A" cannot
be the genuine subject of predication, unless at the same
time we can bring out the real logical role of "A"-the real
logical nature of existential statements.
Now I think it is a great mistake to treat all existential
statements as having the same logical status. I am not
here concerned with the familiar thesis that existence is an
analogical notion-that questions of existence are different
according to what it is whose existence is in question, a
God, an historical character, an animal species, a subatomic particle. . . . Quite apart from these differences,
we have to recognise three different kinds of existential statement even about the same kind of things-e.g., about the
existence of living creatures. The difference perhaps comes
out best when we take negative statements as our examples.
264
P. T. GEACH.
A. There is no such thing as Cerberus; Cerberus

does not exist, is not real.
B. There is no such thing as a dragon; dragons do
not exist.
C. Joseph is not and Simeon is not.
The A statement that I gave as an example might
typically be used to comfort a child who had been frightened
by hearing Greek myths and thought Cerberuswould come
and bite him. " Cerberus ", we might say " doesn't exist (is
not real) like Rover ". Here we are not pointing out any
trait that Cerberus has and Rover lacks; for it would be
nonsense to speak of the trait of being what thereis such a
thing as, and more nonsense to say that some things (e.g.,
Rover) have this trait, while other things (e.g., Cerberus)lack
it, and are thus things that there is no such thing as.
Logically our statement is about a difference not between
two dogs, Cerberus and Rover, but between the uses of two
words " Cerberus " and " Rover ".
The word
''
Rover "
is seriously used to refer to something and does in fact so

refer; the word " Cerberus" is a term that we only make
believe has reference. Now since " Cerberusdoes not exist "
and " Rover does exist " have not (as they seem to have) the
names of two dogs as their logical subjects-since we are
not here using " Cerberus" and " Rover" as names, but
talking about their use-it is appropriate to say that in
A statements " exists " or " is real " is not a predicate, not
even of the word " Cerberus ". To show the real force of
the parent's reassuring " Cerberus does not exist ", and
how it is about the word " Cerberus", we cannot content
ourselves with writing " (The word) ' Cerberus ' does not
exist ", but must completely recast the sentence, say as

follows: " When I said ' Cerberus' in that story, I was
only pretending to use it as a name".
As regards the B statement " there is no such thing as a
dragon " or " dragons do not exist ", it is equally clear that
this cannot be referring to an attribute of beingwhat thereis

sucha thingas, which cows have and dragons lack. But there
is also a great difference between A and B statements. In
FORM AND EXISTENCE.
265
the A statement we have an ostensible use of a proper name;

in the B statement we have a descriptive, predicable
expression like " cow" or " dragon "-what Frege would
have called a Begriffswort(concept-word). The difference
between A and B statements may perhaps come out better
if I shift to another pair of examples. A certain astronomer
claimed to have identified an intra-Mercurianplanet, which
he christened " Vulcan ". His claim was not substantiated;
and a modern astronomer would accordingly say " Vulcan
did not exist " (an A statement) ; he would also say " So
far as we know, an intra-Mercurian planet does not exist"
(a B statement). Now the force of the A statement is to
deprecatethe prematureintroduction of the term " Vulcan "
into astronomical discourse; but in the B statement the
astronomer does not deprecate the use of the term " intraMercurian planet ", but himself uses that term to make a
scientific remark. He is not, however, using the term as a
name, as a subject of predication, but as a logical predicate.
" There is no intra-Mercurian planet " means " nothing at
all is an intra-Mercurianplant ". Similarly, the affirmative
B statement " there is a hairless cat " means " something or
other is a hairlesscat ". Now the use of a logical predicate
in general does not commit you to allowing that there is
something it applies to; it does so commit you if you make
an affirmative assertion with that as the predicate, but not
if e.g. you use the predicate negatively or in the antecedent
or consequent of a hypothetical. So saying " nothing whatever is an intra-Mercurian planet " does not commit you to
allowing that there is after all such a planet.
The importance of B statements is that what Aquinas
calls the question an est ? has to be answered with an affirmative or negative B statement. Aquinas realised the logical
peculiarity of B statements: that the B statement " an F
exists " does not attribute actuality to an F, but F-ness to
something or other; e.g., " there is evil " does not mean
" evil has actual existence " but "some things have
defects ". (Ia. q. 48 art. 2 ad 2 um.) And let us not think
this is so because of some peculiarity of the term " evil ";
Aquinas speaks of the question an est ? quite generally in the
266
P. T. GEACH.
place I have cited, and says that the " existence " involved
in a true affirmative answer to it consists in the truth of an
affirmative predication (compositio). This is exactly right,
for " an F exists " is true if and only if " F " is truly predicable
of something or other. Moreover, the same logical status
is expresslyascribed to " God exists ", or " there is a God"
(Ia. q. 3 art. 4 ad 2 um); and Aquinas expresslydenies that
this statement relates to what he calls God's esse or actus
essendi. (This most important negative indication as to how
we must understandthe term essehas often been overlooked).
In " God exists " we are not predicating something of God,
but predicating the term " God " itself; " God exists"
means " something or other is God ". When we see this,
we can steer our way safely through all the shoals of the
17th-centuryontological argument. (Though it is commonly called by the same name, I think Anselm's argument is
essentially different; I shall not here discuss it.)
It is important that for Aquinas " God " is a descriptive,
predicable, term (nomennaturae-Frege's Begriffswort)and
not a logically proper name. Only because of this can the
question whether there is one God or many make sense;
just as the question whether there is one sun or many
makes sense only if " sun " is used to mean " heavenly
body of such-and-sucha nature ", not if it is a proper name
of this heavenly body. (Ia.q. 13 art. 9) Only because of
this can the heathen say his idol is God and the Christian
contradict him and both be using " God " in the same
sense; if " God " were a proper name, it would be a
logically impossible, not a lying, wicked, act, to predicate
it of stocks and stones. (Ia. q. 13 art. 10).
We now come on to C statements like "Joseph is not
and Simeon is not ". It would be quite absurd to say
that Jacob in uttering these words was not talking about
Joseph and Simeon but about the use of their names. Of
course he was talking about his sons; he was expressing a
fear that something had happened to them, that they were
dead. We have here a sense of " is " or " exists " that seems
to me to be certainly a genuine predicate of individuals;
the sense of " exist" in which one says that an individual
267
FORM AND EXISTENCE.
came to exist, still exists, no longer exists, etc.; the sense

of " to be " in which God says of himself " I am who am "
(i.e. " I am he who is "), or in which Homer spoke of the
Gods who ever are, a\v
Now why should one
'd0vreT.
suspect that this cannot be a genuine predicate of individuals? The fact that in A and B statements the verb
" exist " or " be " is not such a genuine predicate tells us
nothing about C statements. And we cannot argue that
if the C statement " x is not " is true, then the subject
" x)" no longer has anything to refer to and therefore no
statement has been made. As Wittgenstein says (PhilosoI?40): " That is to confound the
phische Untersuchungen
reference (Bedeutung)
of the name with the bearerof the name.
When Mr. N. N. dies, we say that the bearer of the name
dies, not that the reference dies. And it would be nonsensical to say that; for it the name ceased to have reference,
it would make nonsense to say 'Mr. N. N. is dead' ".
To put the same thing another way: The reference of a
name admits of no time qualification; names are tenseless.
Octavian was not known as " Augustus "till quite late in his
career; but once the name was in established use with that
reference, it could be used by anybody at any time, in
relation to any stage of Octavian's career (e.g. in answering
the question " where was Augustus born ?"), and regardless
of whether Octavian was alive or dead. So negative C
statements can raise no paradoxes of reference-and in
showing this we had no need to deny that in them " is
or " exists " is a genuine predicate.
Now it is this sense of " is " or " exists ", the one found
in C statements, that is relevant to Aquinas's term esse.
This interpretation, I maintain, alone makes coherent sense
of all that Aquinas says about esse.
It is worth noticing that as regards living beings " to be"
(in the C sense) has the same reference as " to live", vivere
viventibusest esse (Ia.q. 18 art. 2). This may confirm us
against sophistical attempts to show that the verb " to be "
in this sense is not a genuine predicate of individuals.
" Poor Fred was alive and is dead ", how could one argue
that this is not a genuine predication about poor Fred ?
2F
268
P. T. GEACH.
and what difference does it make if we say instead " poor

Fred was, and is not"?
Some people may not easily see the difference between
"God exists " (sense B) and " God is " (sense C). But in
the contradictories the difference is apparent. "God is
not " (sense C) would have to be construed like "Joseph
is not " ; it would then suitably express the supposition
that perhaps the world was made by an old superannuated
God who has since died (a suggestion of Hume's). This is
quite differentfrom the atheist's" there is no God " (senseB).
One would indeed wish to say that everlasting existence
is part of the concept of a God ; of Hume's senile creator
one would wish to say that since he is dead now he never
was God when he was alive. But saying this does not
commit us to the fallacy of the Ontological Argument. It
belongs to the concept of a phoenix that it should never
die by accident and should ward off old age with a bath of
flames at regular intervals; so a dead phoenix is a contradiction in terms. But this does not mean that there must
be a live phoenix. If there is a God, then he lives for ever;
but we cannot determine from this whether there is a God.
Existence in sense C is very closely connected in Aquinas's
mind with form: quodlibetesse est secundum
formamaliquam.
(Ia. q. 5 art. 5 ad 3 um). For it is in this sense of " exist "
that we say a thing goes on existing; and for a thing to
continue to exist is for it to be the same X over a period of
time, when " X " representssome Begriffswort; and this in
turn means the persistence in an individual of the form
expressed by the predicable expression" X ". Thus, a man
continues to exist in that the baby, the youth, and the
grown man are thesameman; and this means the persistence
in some individual of the form, Begrff, that answers to the
" man
Begriffswort
III
Having explained the construction of the phrase " that
by which. . .", and the relevant sense of " is
",
we consider
what is meant by phrases of the form " that by which x is ",

" x " being replaced by a name of an individual. Now
FORM AND EXISTENCE.
269
it is the fundamental doctrine of Aquinas, repeatedly stated,

that except when x is God, x is never identical with that by
which x is. This doctrine is, I think, intelligible in the light
of our previous enquiry; but it is a most surprisingdoctrine.
Why should we, how can we, distinguish between an individualised form and that by which the individualized form
is? e.g. between the redness of Socrates's nose and that by
which the redness of Socrates'snose is, goes on existing? We
find in Aquinas himself remarks which might suggest the
view that esseis an inseparableand only conceptually distinct
aspect of the individualized form itself. Esse autemper se
convenit
formae, quaeest actus (Ia. q. 75 art. 6); unumquodque
formaliterestperformamsuam(IIIa. q. 2 art. 5 ad 3 um).
All the same, I think there are good reasonsfor accepting
the real distinction between an individualized form and the
corresponding esse. The most important and most general
reason is stated succinctly but clearly by Aquinas himself.
If x is F andy is F, then in respect of F-ness x andy are so
far alike; the F-ness of x will indeed be a differentindividualized form from the F-ness ofy, but they will be, as F-nesses,
alike. But when x is andy also is, the esseof x and the esse
ofy are in general differentas such. (Ia.q. 3 art. 5: Tertio. . .)
Now this marks an unbridgeable distinction between esse
and any form F-ness whatsoever. And the distinction between the references of the expressions " that by whichis " and " that by which-is F " will not vanish even if the
esseof God and (say) the wisdom of God are identical; no
more than " the square of " and " the cube of " have the
same reference because I is both the square and the cube
of 1.
A modern philosopher will often challenge philosophical
assertionswith the question: As opposed to what? This is a
legitimate move; as Aquinas says, knowing the truth of
an assertion is tantamount to knowing the falsehood of its
contradictory opposite (Ia. q. 58 art. 4 ad 2 um). Well
then, as opposed to what does one say that in general the
esseof x and the esseofy are as such distinct ?-In the fairytale, all the human members of the family and the family
cat shared a single life, that is, a single esse (vivereviventibus
270
P. T. GEACH.
est esse); and when the betrothed of the youngest daughlter

took a pot-shot at the cat, its death was the death of the
whole family. In actual families, animality is common to
all the members of the family, including the cat, but esse
is not, and so killing the cat has no such consequence. So,
although for a man or cat to go on existing is precisely the
continued existence of his animality, that is, the persistence
of a certain individualized form in continuously renewed
matter: nevertheless we must recognise a real distinction
between his animality and his esse.
A second reason appeals to the nature of intensive
magnitude. It may be that x is F andy is F, and that they
have the same specific F-ness, but yet the F-ness of x is more
intense than that of y. Moreover, the F-ness of x may become more or less intense; and increase of F-ness plainly
resembles a thing's coming to be F, whereas a decrease of
F-ness resembles a thing's altogether ceasing to be F. Now
difference between x andy as regards intensity of F-ness is
not difference precisely as regards F-ness (especially as we
may suppose x and y to have the same specific F-ness);
it is rather, I wish to say, difference as regards the existence
of F-ness-the F-ness of x exists morethan the F-ness of y.
So also a change in mere intensity is a change as regards
existence; increase in the intensity of x's F-ness resemblesthe
coming to be of x's F-ness, both being additions of existence;
decrease in the intensity of x's F-ness resembles the ceasing
of x's F-ness, both being subtractions of existence. Here
again, there is a real distinction hetween the F-ness of x
and the esse of this individualized form; while the F-ness
as such remains unchanged, its existence may vary in degree.
(I have deliberately stated this argument in a schematic
way, in order to avoid irrelevant controversy about my
choice of an example. It is not so easy as it looks to find an
unexceptionable case of difference or change in mere
intensity, without any difference or change in quality:
Aquinas's favourite example, heat, would land us in many
difficulties. So far as I can see, soundis a good example.
A louder and a softer sound may be qualitatively identical;
and a sudden increase of loudness resembles a sound's
271
FORM AND EXISTENCE.
suddenly starting, a sudden decrease of loudness its suddenly

stopping.)
A third reason is found in considering the nature of
thought. How remarkable that if there can (logically)
be an X, there can also be a thought of an X ! What is this
relation beinga thoughtof, which can have anything whatsoever as its term-even non-existent things like dragons?
And how can there be an activity whose whole nature
consists in its having this relation to something-to " something " possibly non-existent?
Now Aquinas's account of thought denies that its nature
consistswholly in relation to something outside itself. When
Plato thinks of redness, what exists in Plato is not a certain
relationto rednessor red things, but is redness,is an individual
occurrence of the very same form of which anotherindividual
occurrence is the redness of this rose. But how then is it
that this rose is red but not Plato's mind? Because the mode
of occurrence of redness-not the redness that occurs-is
unlike; the redness in Plato's mind is, exists, in a different
way (esse intentionale)from the redness in this rose (which
has essenaturale).We thus understandthe intimate connexion
of thought and its object. There is, as Aquinas often says,
likeness(similitudo)between them. We also now understand
the odd-seeming fact that there can be a thought of anything
that there can be. Existence makes no difference to,
and can impose no restriction on, the nature of that which
exists; if it is possible that there should be F-ness at all, then
it is possible alike that there should be F-ness occurring with
esse naturale(a real live cow, say) and that there should be
F-ness occurring with esseintentionale(a thought of a cow).
The whole basis of this account is that the individual
F-ness is really distinct from its esse,naturaleor intentionale
as
the case may be.
To get this doctrine of Aquinas's properly straightened
out in our minds we must realize that Aquinas is not saying
e.g. that the cow in the meadow leads a double life and has
another essein my mind; or that I, when I think of the cow,
become somehow identical with the cow. These kinds of
talk (derived, I should guess, from taking too seriously the
2G
272
P.
T.
GEACH.
notorious trick expression " in the mind ") are expressly

rejected by Aquinas. Lapis non est in anima, sed species
lapidis.. . similitudo rei intellectaeestforma intellectus. (Ia. q.
85 art. 2). There is one individualized form in the stone,
and another individualized form in the mind of the man
who thinks of it; these individualized forms are both
occurrences of the same form, but differ in their manner of
esse; neither the stone nor its individualized form is to be
found in my mind. Even when one angel thinks of another,
what is in the knower is an individualized form differing
from the individualized form that is in (or rather, for
Aquinas, is identical with) the angel who is known, although
both are occurrencesof the very same form. On Aquinas's
view, the difference (say) between Michael's thought of
Gabriel and Gabriel himself, like the difference between my
thought of a cow and the cow in the meadow, consists in
one and the same form's occurring in Michael (or me)
with esse intentionale,in Gabriel (or the cow) with esse
naturale; and this involves the existence of two individualized forms. (Ia. q. 56 art. 2 ad 3 um). It should be
noticed in regard to the last citation that the difference
between essenaturaleand esseintentionale
can occur even when
there is no question of matter; angels, for Aquinas, have
no sort of matter in their make-up, but the difference still
holds good in their case.
I have had to be brief and leave out many important
topics: in particular, the difference between the nature or
essence and the (substantial) form of a material thing; the
sense in which the human soul is the " form " of the human
body; the conception of materiaprima; the question whether
angels are " forms " (sc. individualizedforms.) Even so I
could not make this paper easy going, nor avoid what
many people would call hairsplitting distinctions. But all
seriousphilosophy is difficult; and as for hairsplitting,
A Hair perhaps divides the False and True,
Yes; and a single Alif were the clueCould you but find it-to the Treasure-house
And peradventureto THE MASTER too.

Geach, Peter (1954) - Form and Existence

Uploaded by

Copyright:

Available Formats

Geach, Peter (1954) - Form and Existence

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geach, Peter (1954) - Form and Existence

Uploaded by

Copyright:

Available Formats

Form and Existence

Meetingof theAristotelianSocietyat 21, BedfordSquare,London,

predicate. (Ia q. 13 art. 12; q. 85 art. 5 ad 3 um). I

FORM AND EXISTENCE.

relation of subject and predicate, a problem that is thus

If this answer were true, there would be objects of which

about a set of ducksunless the question answeredis a question

FORM AND EXISTENCE.

to bridge the gulf; but, as we shall presently see, they do

abstract " proper name " in it (say: " Redness is an eternal

FORM AND EXISTENCE.

of forms, the spurious proper names "wisdom" and

Socrates . . . " we have

with a name of something that has the form, just as the

that cannot be made in Latin at all-is quite trivial and

certainly not interchangeable; but I think we can show

to mention a form for the sake of expressingthe supposition

that in a given individual that form is found, we refer to it

FORM AND EXISTENCE.

without completion this

latter phrase is senseless.

distinction between form and individual is undone. To keep

FORM AND EXISTENCE.

not signified as being, a part or element of what we signify

designate the same number, and there is no distinction even

or " exists " is properly predicable of individuals. Now

FORM AND EXISTENCE.

sense ! But at any rate to a discussion of Aquinas this

A. There is no such thing as Cerberus; Cerberus

is seriously used to refer to something and does in fact so

exist ", but must completely recast the sentence, say as

this cannot be referring to an attribute of beingwhat thereis

FORM AND EXISTENCE.

the A statement we have an ostensible use of a proper name;

FORM AND EXISTENCE.

came to exist, still exists, no longer exists, etc.; the sense

and what difference does it make if we say instead " poor

what is meant by phrases of the form " that by which x is ",

FORM AND EXISTENCE.

it is the fundamental doctrine of Aquinas, repeatedly stated,

est esse); and when the betrothed of the youngest daughlter

FORM AND EXISTENCE.

suddenly starting, a sudden decrease of loudness its suddenly

notorious trick expression " in the mind ") are expressly

You might also like