Semantic Principles of Aquinas

MEDIEVAL PHILOSOPHY AND THEOLOGY
Volume 5, Number I, March 1996

Medieval Philosophy and Theology 5 (1996),87-141. Printed in the United States of America.
Copyright 1996 Cambridge University Press 1057-0608/96 $7.50 + .10
Contents
Augustine on Theological Fatalism: The Argument of
De Libero Arbitrio 3.1-4
DAVID P. HUNT
On a Sophisma of Richard Kilvington and a Problem of Analysis
BERNARD D. KATZ
Expositions of the Text: Aquinas's Aristotelian Commentaries
JOHN JENKINS
1
3
1
39
The Semantic Principles underlying Saint
Thomas Aquinas's Metaphysics of Being
GYULAKLIMA
Department of Philosophy
University of Notre Dame and
Institute of Philosophy of the Hungarian Academy
I. INTRODUCTION: SEMANTICS AND METAPHYSICS
The Semantic Principles Underlying St. Thomas Aquinas's
Metaphysics of Being
GYULA KLIMA 87
Nicholas ofAutrecourt and William of Ockham on Atomism,
Nominalisnl, and the Ontology of Motion
BLAKE D. DUTTON 63
CAMBRIDGE
UNIVERSITY PRESS
II II I
1057-0608(199603)5:1 ;1-5
As I hope the title clearly indicates, this article is not intended to contribute
its ounces to the tons ofliterature on Aquinas's metaphysics of being. On the
contrary, its primary motivation is the perhaps deplorable, but certainly not
negligible, fact that the very form of discourse within which the substantive
claims of that literature, as well as Aquinas's own claims, are formulated is
radically different from that of contemporary philosophical discussions.
1
Research for this paper was completed during my Morse Fellowship, which
exempted me from teaching duties at the Philosophy Department of Yale University
in the academic year 1994-95. I thank Desmond Paul Henry, John Jenkins, Heikki
Kirjavainen, Eleonore Stump, andJack Zupko, who read and commented on earlier
drafts of this article. I also thank my new colleagues at the University of Notre Dame
for a very lively and thorough discussion of the same material during my first visit
here. My special thanks go to Scott MacDonald for his detailed and perceptive
comments, as well as for his general suggestions for improving the presentation of
the material. My original plan was to provide a nontechnical exposition of the
semantic principles underlying Aquinas's metaphysics of being and goodness. I had
to realize, however, both that if I do not want to compromise precision I cannot
completely abandon technicalities (though I tried to relegate them mostly to the
footnotes) and that the further considerations that link the notion of being to the
notion of goodness would already exceed the limits of a single research paper. A
full account of these further considerations, as well as the technicalities only out-
lined here, will be provided in my book under preparation: Meaning, Nature,
Concept. Nevertheless, I do hope that the foregoing considerations will already prove
useful in interpreting the passage that most clearly relates the notion of being to
the notion of goodness, while also explains the difference of their predication
simpliciter and secundum quid: sn q. 5, 1.
1. For pregnant expressions of the keen awareness of isolation in many con-
temporary Thomists, see the numerous essays devoted to this problem in D. W.
Hudson and D.W. Moran, eds., The Future of Thomism, American Maritain Associa-
tion Publications (Notre Dame, Ind.: University of Notre Dame Press, 1992). Of
course, one cannot ignore the tremendous amount of good work done by philoso-
87
88 GYULA KLIMA
Facedwith this different form ofdiscourse, modern readers are either willing
(and able) to 'Join in," in which case they may become "players" of the rele-
vant "language game," or they are unwilling (or unable) to do so, in which
case theywill be left ultimately "intellectuallyintact" by these claims. In either
case, without careful reflection on the general principles governing "the
game," this willingness (and ability) on the part of the modern reader will be
determined mostly by vague intuitions and more or less articulated sympa-
thies or antipathies, rather than by serious philosophical considerations.
As we know, metaphysics studies the first principles of all knowledge.
2
But even the metaphysical investigation of first principles presupposes that
we understand what is meant by these principles and their terms; that is,
even metaphysical principles presuppose certain semantic principles. As
Aristotle advises in the fourth book of the Metaphysics, in all disputations
about first principles, the ultimate appeal should be to what both we and
our opponents mean, indeed, to what both of us should mean by our phrases
if we are to maintain mutual understanding in the framework of rational
discourse.
3
However, such an appeal is far from compelling or even unam-
phers, apparently also of some analytic background (e.g., Bochenski, Henry, Geach,
Kenny, Kretzmann, McInerny, Stump, Veatch, Weidemann, just to name a few,
without aiming at completeness), to overcome this "language barrier." However, as
far as I know, no comprehensive attempt has been made to date to state those
formal semantic principles, which as such, regardless of the metaphysical contents
of their particular instances, by reason of their formality, constitute the very form
of discourse presupposed in Aquinas's (and I would add, also his contemporaries')
metaphysical discussions. I hope the statement and discussion of these principles
below will also shed some more light on exactly how I conceive of the separation of
metaphysical from formal semantic principles.
2. For this point, which of course does not define the proper subject matter of
metaphysics but is a consequence of the fact that metaphysics is the study of being qua
being, see In meta. lb. 4, le. 5. References to Aristotle are given by referring to
Aquinas's commentaries. For St. Thomas's works I used the supplemental volumes
to R. Busa (ed.), Index Thomisticus, S. Thomae Aquinatis Opera Omnia (Stuttgart-Bad
Cannstatt: Frommann-Holzboog, 1980).
3. Cf. In meta. lb. 4, le. 7. It is important to realize in this connection that the
main thrust of Aristotle's arguments is that those who deny the first principle (i.e.,
the principle of noncontradiction) cannot possibly mean what they say. It is also
important in this regard to consider what Aristotle and St. Thomas, in his commen-
tary on the Posterior Analytics (hereafter PA) and in several other places, say about
the order of questions to be answered by a demonstrative science. The question of
what a thing is (quid est?) is preceded by the question of whether the thing is (an
est?), but even this question presupposes that we know what is meant by the name
of the thing in question (quid significaturper nomen). Cf. "antequam sciatur de aliquo
an sit, non potest sciri proprie de eo quid est: non entium enim non sunt definitio-
nes. Unde quaestio, an est, praecedit quaestionem, quid est. Sed non potest ostendi
de aliquo an sit, nisi prius intelligatur quid significatur per nomen. Propter quod
etiam Philosophus in iv Metaphysicae, in disputatione contra negantes principia
docet incipere a significatione nominum" (In PA lb. 1, le. 2, n. 5). For the signifi-
cance of this order in the proofs for God's existence, see STl, q. 2, a. 2; ScC lb. 1,
c. 12.
SEMANTIC PRINCIPLES OF AQUINAS 89
biguous, especially today, after the "linguistic turn" in philosophy, when
everyone seems to have his or her own "philosophy of language" and
"theory of meaning and/or reference."4 The primary aim of this article,
therefore, is to spell out with clarity and precision the underlying semantic
principles of the very form of discourse presupposed by Aquinas's substan-
tive metaphysical claims concerning being, clearly distinguishing the for-
mer from the latter. I hope thereby to ensure that the contemporary
reader's willingness (and ability) to participate in the language game of
competently evaluating the substantive metaphysical arguments for and
against the relevant metaphysical claims will depend not on unexamined
intuitions but rather on a careful consideration of these underlying princi-
ples themselves.
To be sure, in this article I cannot undertake a presentation of these
underlying semantic principles as ones that we absolutely have to accept;
I shall try merely to articulate as clearly as possible what it is that we
have to accept for a competent evaluation of the substantive metaphysical
arguments. However, of course, even within the framework of this more
modest enterprise, I will have to show that these semantic principles are
at least acceptable, that is, that they are consistent in themselves and do
not commit anyone willing to maintain them to some manifest falsity or
nonsense.
In the next section, therefore, I will begin the discussion with the
introduction of the basic concepts we need for the clear formulation of the
relevant semantic principles. I start with Aquinas's concept of meaning,
or-using a transliteration of medieval terminology to distinguish it from
contemporary conceptions-signijication. However, as we shall see, there
are difficulties even at the very beginning, for Aquinas's Aristotelian con-
cept of signification apparently commits him to some "mysterious," nonex-
istent objects of signification. So in section III I point out how Aquinas's
distinctions between different senses of 'being' might be used to eliminate
our misgivings concerning such objects.
s
However, this solution will leave
us with a set of further, rather disturbing questions concerning the concep-
tual apparatus used in this discussion. So in section IV I provide a systematic
account of this conceptual apparatus as it functions in Aquinas's theory of
signification and predication, addressing various ontological and epistemo-
logical concerns contemporary philosophers may have regarding these
semantic theories. Having thus placed Aquinas's notions of signification
4. Cf. "the double indexical definition of meaning" provided by William Lycan:
"meaning = df whatever aspect of linguistic activity happens to interest me now";
quoted by M. Devitt, "The Methodology of Naturalistic Semantics," Journal of Phi-
losophy, 91 (1994): 545-72, at 548.
5. Fregean connotations aside, throughout this article, by a sense of an analogi-
cal term I simply mean one of its several but related significations. The clarification
of the notion of "signification," however, is one of the tasks of the subsequent
discussion.
90 GYULA KLIMA
and existence into their proper theoretical context, in section V I finish the
introduction of the basic conceptual apparatus by a brief discussion of
Aquinas's account of the relationship between the signification and reference,
or-again using the transliteration of medieval terminology-supposition of
concrete, as well as of abstract, common terms.
6
Section VI will present the
formulation of the relevant semantic principles themselves, providing a
brief comment on each. On this basis, in section VII I shall show how we
can interpret in this framework all predications about substance as predica-
tions of some act of being either absolutely (simpliciter) or with qualification
(secundum quid). Finally, in section VIII I shall argue that in this theoretical
setting the metaphysical idea of an objective hierarchy of being not only
makes good sense but also is even quite plausible, even if we may not be
able to determine offhand the exact place of any given entity in this overall
hierarchy.
II. SIGNIFICATION AND EXISTENCE
As is well known, what set the stage for all semantic considerations in the
Middle Ages was Aristotle's "semantic triangle," the conception sketched at
the beginning of his On Interpretation. According to this conception, words
immediately signify the concepts of the mind, and it is by the mediation of
these concepts that they signify things. Aquinas comments on the relevant
passage as follows:
Therefore 'passions of the soul' must be understood here as concep-
tions of the intellect, and names, verbs, and speech signify these con-
ceptions of the intellect immediately according to the teaching of
Aristotle. They cannot immediately signify things, as is clear from the
mode of signifying, for the name 'man' signifies human nature in
abstraction from singulars; hence it is impossible that it immediately
signify a singular man. The Platonists for this reason held that it
signified the separated idea of man. But because in Aristotle's teaching
man in the abstract does not really subsist, but is only in the mind, it
was necessary for Aristotle to say that vocal sounds signify the concep-
tions of the intellect immediately and things by means of them.
7
6. Thus, in this article I am not going to deal per se with the medieval theory
of supposition in general or Aquinas's version of it in particular. I am going to touch
on Aquinas's (and his contemporaries') conception of the relationship between
signification and supposition only to the extent that an account of this relationship
is needed for understanding Aquinas's treatment of the notion of being. Cf. n. 54
below.
7. Aristotle: On Interpretation: Commentary lJy St. Thomas and Cajetan, trans. J. T.
Oesterle (Milwaukee: Marquette University Press, 1962), p. 25. Unless otherwise
indicated (as I have done here), translations in this article are mine.
So signification is dependent on acts of human thought: by our words we
signify whatever we can think of, whether it actually exists or not. For there is
no doubt that we can think of something that does not exist, and so in this
conception, it follows that our words can also signify something that does not
exist. Thus it is no wonder that at another place Aquinas writes as follows:
In response we have to say that there is a three-fold diversity among
things signified by names. For some are in total, complete being out-
side the soul; and such are complete beings, such as a man or a stone.
Some have nothing outside the soul, such as dreams or the imagination
of a chimera. And some have some foundation in reality outside the
soul, but their formal account [ratio] is completed by the operation of
the soul, as is clear in the case of universals. For humanity is something
in reality, but there it is not universal, for there is no humanity outside
the soul common to many. But as it is conceived by the intellect, a
further concept [intentio] is adjoined to it by the intellect's operation,
on account of which it is called a species. 8
However, this conception of signification immediately gives rise to a
swarm of rather disturbing questions. For even granting that we can think of
and thus speak about these mysterious, nonexistent objects of signification,
the first question that immediately arises is this: what, then, are these objects
that we are thus speaking about, if they are none of the things in the world?
And, in any case,-howexactly is our ability to signify these objects by our words
related to our ability to speak about, that is, to refer to, them? What is
Aquinas's conception of the relationship between meaning (signification)
and reference (suPPosition)? Also, what can we make of Aquinas's threefold
distinction among objects of signification? Ifchimeras and humanity as such
are equally nonexistent, whatever they are in themselves, what is that 'foun-
dation in reality' the having or lacking of which is supposed to distinguish
them from one another? And, after all, what does this whole tangled issue of
nonexistent objects of signification have to do with Aquinas's metaphysics of
being, the semantic principles ofwhich we are elucidating here?
III. ENS REALE VS. ENS RATIONIS AS ENS SIMPLICITER
VS. ENS SECUNDUM QUID
Given the possibility of signifying objects of thought that do not exist (on
the basis of their thinkability and the relationship between thought and
signification), it is natural to ask about the nature of, and the relationships
between, the members of the resulting classifications. In fact, Aquinas's
famous (Aristotelian) distinctions of the various senses of 'being', which
8. SM d.19, q.5, a.I.
92 GYULAKLIMA
serve as the semantic foundation of his metaphysics of being, are part of the
response to such questions.
The primary distinction concerns the notions of what in the scholastic
tradition came to be known as real beings (entia realia) and beings of reason
(entia rationis). Many shorter characterizations are scattered throughout
Aquinas's works,9 but the fullest account of this distinction is provided by
the following passage:
By way of answer we have to say that the Philosopher shows that 'being'
is predicated in many ways. For in one sense 'being' is predicated as it
is divided by the ten genera. And in this sense 'being' signifies some-
thing existing in the nature of things, whether it is a substance, such as
a man, or an accident, such as a color. In another sense 'being' signifies
the truth of a proposition; as when it is said that an affirmation is true
when it signifies to be what is, and a negation is true when it signifies
not to be what is not; and this 'being' signifies composition produced
by the judgment-forming intellect. So whatever is said to be a being in
the first sense is a being also in the second sense: for whatever has
natural existence in the nature of things can be signified to be by an
affirmative proposition, e.g. when it is said that a color is, or a man is.
But not everything which is a being in the second sense is a being also
in the first sense: for of a privation, such as blindness, we can form an
affirmative proposition, saying: 'Blindness is'; but blindness is not
something in the nature of things, but it is rather a removal of a being:
and so even privations and negations are said to be beings in the second
sense, but not in the first. And 'being' is predicated in different man-
ners according to these two senses: for taken in the first sense it is a
substantial predicate, and it pertains to the question 'What is it?' [quid
est?], but taken in the second sense it is an accidental predicate ... and
it pertains to the question 'Is there [such and such a thing]?' [an est?].10
This distinction derives primarily from Aristotle's discussion of the concept
of being in the fifth book of his Metaphysics, where St. Thomas starts his
comments on the relevant passages with the following remarks:
[Aristotle] first divides being that is outside the soul, which is perfect
being, by the ten categories. Secondly he considers another mode of
9. Cf. 2SNd. 34, q. 1, a. 1; ISNd. 19, q. 5, a. 1, ad 1, d. 33, q. 1, a. 1, ad 1.; 2SN
d. 37, q. 1, a. 2, ad 1 and 3; De ente c. 1; QDP q. 7, a. 2, ad 1; QDM q. 1, a. 1, ad 19.;
QDL 9, q. 2, a. 2; In meta. lb. 4, le. 1, lb. 5, le. 9, lb. 6, le. 2, lb. 6, le. 4, lb. 9, le. 11,
lb. 11, le. 8; S71 q. 3, a. 4, ad 2, q. 16, a. 3, ad 2, q. 48, a. 2, ad 2; S71 2, q. 36, a. 1;
ScC lb. 1, c. 12, lb. 1, c. 58, lb. 3, c. 9, lb 3, c. 8, n. 13. Cf. also T. de Vio Cajetan,
Commentary on Being and Essence, trans. L. J. Kendzierski and F. C. Wade (Milwaukee:
Marquette University Press, 1964), c. 1; C. Alamannus, Summa Philosophiae (Paris: P.
Lethielleux, 1888), Tom.l. sect. 2, 5, 1; R. W. Schmidt, The Domain ofLogic according
to Saint Thomas Aquinas (The Hague: Martinus Nijhoff, 1966), part 2, chap. 4, and
part 3, chap. 8.
10. 2SNd. 34, q. 1, a. 1.
being, according to which it is only in the mind.... Thirdly he divides
being by potentiality and actuality: and being thus divided is more
common than perfect being. For a being in potentiality is only a being
with qualification [secundum quid] and is imperfect.
ll
But the same distinction is considered even earlier by St. Thomas, in
his comments on book 4 of the Metaphysics, where he embeds this distinc-
tion into the broader context of his famous doctrine of the analogy of
being,12 Having discussed the ways in which the various analogous senses of
'healthy' (according to which it can be predicated even of food and of
urine) are related to one another, namely, the various ways in which its
secondary senses are related to its primary sense (which signifies the health
of an animal), Aquinas says that 'being' is analogous in the same way in that
all of its secondary senses are related to a primary sense, namely, the sense
in which 'being' is predicated of substance. He summarizes his discussion
in the following passage:
We should know that the above-mentioned modes of being can be
reduced to four. For one of them, which is the weakest, is only in
reason, namely, negation and privation, which we say are in reason
because reason considers them as if they were some beings, when it
affirms or denies something of them.... Another [mode of being],
which is the closest to this one in weakness, is that according to which
generation and corruption and motions are said to be beings. For they
have some privation and negation mixed with them. For motion is
imperfect actuality, as is said in book 3 of the Physics. [A being] in the
third sense has nothing of non-being mixed with it, but it has a weak
existence [habet esse debiZe], for [it has existence] not by itself, but in
something else, as [do] qualities, quantities and properties of sub-
11. In meta. lb. 5, le. 9.
12. Of course, Aquinas's doctrine of analogy in general is in itself a huge and
hotly debated topic. Nevertheless, the details of that doctrine, describing exactly
how the various analogous senses of an analogous term are related to one another
and to the things named analogously, need not be considered here. As we shall see
below, all we need in the present context is to realize that a secondary sense of an
analogous term may be expressed by adding some diminishing qualification to the
same term in its primary sense, that is, that a secondary sense of an analogous term
is the result of some modification of the primary sense. However, we need not
consider in detail exactly how such modifications can occur; that is, we need
not consider what are the different modes of analogy, which is the focus of the
debates. For a masterly exposition of the issues involved, see R. McInerny, The Logic
of Analogy (The Hague: Martinus Nijhoff, 1961). Recently Professor McInerny has
also kindly given me the opportunity to consult the manuscript of his new book
under preparation (Aquinas and Analogy), which presents a detailed, although also
debatable, criticism of Cajetan's interpretation of analogy. Cf. also E. J. Ashworth,
"Analogical Concepts: The Fourteenth-Century Background to Cajetan," Dialogue,
31 (1992): 399, and "Analogy and Equivocation in Thirteenth-eentury Logic: Aqui-
nas in Context," Mediaeval Studies 54 (1992): 94.
94 GYULA KLIMA
stance. The fourth kind is that which is the most perfect, namely, that
which has existence in nature without any admixture of privation, and
has firm and solid existence, as it exists by itself, as do substances. And
it is to this [last one], as primary and principal, that all the others are
related. For qualities and quantities are said to be insofar as they are in
a substance; motions and generations are said to be insofar as they tend
to substance or to some other of the above-said [beings]; and privations
and negations are said to be insofar as they remove some of the
above-said three)3
Thus, as we can see, the distinction between ens reale and ens rationzs IS
regarded by St. Thomas as forming part of the division of the common term
ens, that is, 'being', into its various, though related, analogical senses.
Concerning this kind of division in general, as opposed to the division of a
genus into its species, St. Thomas writes the following:
There are two ways in which something common can be divided into
what are under it, just as there are two ways in which something is
common. For there is the division of a univocal [term] into its species
by differences by which the nature of the genus is equally participated
in the species, as animal is divided into man and horse, and the like.
Another division is that of something common by analogy, which is
predicated according to its perfect concept [ratio] of one of those that
divide it, and of the other[s] imperfectly and with qualification [secun-
dum quid], as being is divided into substance and accident, and into
being in actuality and in potentiality)4
Accordingly, however obscure the details of the divisions of the notion
of being may appear at first, the first thing to realize about the distinction
between real beings and beings of reason is that it does not constitute a
division of a class into its subclasses as, for example, the distinction between
rational and nonrational animals constitutes the division of the class of
animals into two kinds of animals, namely, humans and brutes. Real beings
and beings of reason do not in the same way constitute two subclasses or
two kinds of beings. Indeed, they no more constitute two kinds of beings
than real money and forged money constitute two kinds of money. Just as
it is only real money that is money simpliciter, that is, without qualification,
so it is only a real being that is a being simpliciter, without qualification. As
medieval logicians (and Aquinas himself) would say, the qualification 'of
reason' as added to 'being' or 'forged' as added to 'money' is an example
of a "diminishing" qualification or determination (determinatio diminuens) ,
whereas the qualification 'real' in both cases is a "nondiminishing" qualifi-
cation (determinatio non diminuens). The origin of the theory of these two
kinds of determination goes back to Aristotle's Sophistical Refutations, in
13. In meta. lb. 4, le. 1, n. 15.
14. In Peri lb. 2, le. 2, n. 3.
particular to his treatment of the fallacy referred to by the schoolmen as the
fallacy secundum quid et simpliciter.
15
According to the medieval analysis, the fallacy consists in dropping a
diminishing qualification in an inference from a proposition in which a
predicate is applied to a subject with some qualification (secundum quid) to
a proposition in which the same predicate is applied to the same subject
absolutely, without qualification (simpliciter). In his De Fallaciis,16 Aquinas
characterizes this fallacy in the following manner: .
Next, about the fallacy secundum quid et simpliciter. In this context, that
is [said to be] predicated simpliciter whicp. is predicated without any
modification [modus] added to it, as when we say: 'Socrates is white' or
'Socrates runs'; and that is [said to be] predicated secundum quid which
is predicated with the addition of something to it, as in 'He runs well'
or 'Socrates is white with respect to his teeth'. What is added may be
related in two ways to that to which it is added: for sometimes it does
not diminish the concept [ratio] of that to which it is added, and then
it is possible to proceed from what is secundum quid to what is simpliciter,
as when we say: 'He runs fast; therefore, he runs', for speed does not
take away anything from the concept [ratio] of running.... Sometimes,
however, what is added takes away something from the concept [ratio]
of that to which it is added, as when it is said: 'Ablack man is white with
respect to his teeth'. For the qualification 'with respect to his teeth'
takes something away from the concept ofwhat is said to be white, since
nobody can be said to be white, except who is totally white, or with
respect to most or the principal parts. Therefo"re, if someone were to
conclude: 'A black man is white with respect to his teeth; therefore, he
is white', then this is a sophistic argument, or an instance of the fallacy
secundum quid et simpliciter, and the deception derives from taking what
is predicated secundum quid as if it were predicated simpliciter.
17
Thus, when the qualification is diminishing, its omission yields an invalid
inference. But when it is nondiminishing, that is, when the qualification
added to the predicate does not specify any conditions for the applicability
of the predicate other than those included in the meaning of the predicate
in itself, then its omission preserves the truth of the original proposition.
As Aquinas explains somewhat later, when discussing the several modes in
which this fallacy can occur,
15. For a more comprehensive discussion of, and further references to, the
medieval literature on this fallacy (in connection with St. Thomas's use of the
related theoretical apparatus in his theology of the Incarnation), see G. Klima,
"Libellus pro Sapiente: ACriticism of Allan Back's Argument against St. Thomas
Aquinas' Theory of the Incarnation," The New Scholasticism, 58 (1984): 207-19.
16. A minor work whose authenticity recently has been vindicated. See Father
~ u s s n?te at the end of the list ofworks included in the supplements (very conven-
Iently pnnted at the end ofeach volume) ofhis monumental Index Thomisticus.
17. De Fallaciis c. 13.
96 GYULA KLIMA
We should know, however, that if a whole is aptly denominated from its
part, then there is no fallacy, as is clear in this case: 'He is curly with
respect to his hair; therefore, he is curly'. This correctly follows, for a
man is denominated curly with respect to his hair. And this mode covers
also other [sorts of] parts, namely, [parts] of place, of time, or of other
[sorts of] wholes. If, however, something is added to a whole in place by
the mediation of a part in place, from which part it is not aptly denomi-
nated, the fallacy occurs in such inferences, as [in the following]: 'This
diet is good in unhealthy places, so it is good'. This does not follow, for
'in unhealthy places' signifies a part in place. The case is similar with a
whole and part in time, as here: 'Drinking wine is bad for the sick;
therefore, it is bad'. And the same goes for all similar cases.
I8
Therefore, on the one hand, since 'real' as added to 'being' is a
nondiminishing qualification, a real being is simply a being. On the other
hand, since 'of reason' as added to 'being' is a diminishing qualification, a
'being of reason' is not necessarily a being, except secundum quid. 19 But then
we can see that our question about the nature of nonexistent objects of
signification (at the end of the previous section) was based on a category
mistake. For the question about the nature of anything presupposes that it
has a nature. But this, again, presupposes that it exists, that is, that it is a
being simpliciter, a real being:
Since there is no quiddity or essence of a non-being, nobody can know,
with regard to that which is not, what it is; but one can know the
signification of the name, or the description composed of several
names; as for example one can know what the name 'tragelaphus' (or
'goatstag', which is the same thing) signifies since it signifies an animal
composed of a goat and a stag. But it is impossible to know what a
goatstag is, for nothing is like this in the nature of things.
20
Therefore, to ask, 'What are nonexistent beings of reason?', where that
question is understood as asking about the nature of a curious kind of
being, namely, beings that do not exist, is just as misguided as asking about
the actual currency exchange rate between, say, yen and forged dollars. But
again, this does not mean that we cannot understand what we mean by the
phrase 'beings of reason'. On the contrary, as just explained, such beings
18. De Fallaciis c. 13.
19. Cf. "Unum enim eodem modo dicitur aliquid sicut et ens; unde sicut ipsum
non ens, non quidem simpliciter, sed secundum quid, idest secundum rationem, ut
patet in 40 Metaphysicae, ita etiam negatio est unum secundum quid, scilicet
secundum rationem" (In Peri lb. 2, le. 2, n. 3).
20. In PA lb. 2, le. 6, n. 2. Cf. In PA lb. 1, le. 2, n. 5: " antequam sciatur de aliquo
an sit, non potest sciri proprie de eo quid est: non entium enim non sunt definitio-
nes. Unde quaestio, an est, praecedit quaestionem, quid est. Sed non potest ostendi
de aliquo an sit, nisi prius intelligatur quid significatur per nomen. Propter quod
etiam Philosophus in iv Metaphysicae, in disputatione contra negantes principia
docet incipere a significatione nominum."
of reason are what certain expressions signify in virtue of the fact that we
can think of them when we understand the expressions in question,
whether they actually exist in rerum natura or not.
So far, so good. But even if this response dulls somewhat the critical edge
of this disturbing question, I hardly think any contemporary philosopher
would be absolutely satisfied by the explanations given so far, and apparently
for good reasons. For were not these explanations just a series of obscura per
obscuriora, vain attempts to elucidate some obscure distinctions by other, even
more obscure ones? After all, we started out at the end of the previous section
by answering our disturbing questions about the distinction between existent
and nonexistent objects of signification by presenting St. Thomas's quite ob-
scure distinction between the two senses of 'being'. This turned out to bejust
a part of a more comprehensive distinction among the various analogous
senses of the same word. In its turn, the explanation of this latter distinction
led to further distinctions, namely, those between predication simpliciterand
predication secundum quid and between determinatio diminuens and determina-
tio non-diminuenr-and all of thisjust to pave the wayfor dismissing one ofour
initial questions as mistaken. Thus, the dubious result of eliminating this
question by this procedure was achieved at the expense of introducing dis-
tinctions that are bound to give rise to even further and even more disturbing
questions. For how are we supposed to understand St. Thomas's claims in his
distinction of the two notions of being? Why would blindness be a nonbeing
or a being only in the second sense? And even if it is, why would this rather
obscure fact be expressed by the copula of an affirmative proposition? And
what should we make of the claim that the concept ofbeing that is applicable
to blindness is a result ofsomehow "diminishing" the concept ofbeing that is
applicable to sight, which is in turn a concept of being that results from an-
other "diminishing" of the concept of being that is applicable to substances,
that is, to ordinary things such as animals that can be sighted or blind? In-
deed, what does the fact that an animal is sighted or blind have to do with the
existence (in any sense) of such spurious entities as blindness or sight?
l\Z SIGNIFICATION AND THE INHERENCE
THEORY OF PREDICATION
To provide acceptable replies to these questions as well as to the previous
ones, we have to start with a systematic account of Aquinas's conception of
signification. As we have seen, for Aquinas our words have two sorts of
signification, immediate and ultimate. What a word immediately signifies is
a human concept, the possession and actual exercise of which makes a
human being aware of something.
Note here that by this claim Aquinas is not committed to any particular
"philosophy of mind and language" (although, of course, he has one) or to
98 GYULA KLIMA
any particular psychological or psycholinguistic theory. For such theories
would certainly have to have an answer to the question of what human
concepts in themselves are. That is, they would have to characterize the
nature of these acts of human awareness and the nature of their relation-
ship to their subject, the human mind, which in turn would demand a
characterization of the nature of the human mind itself, and so forth.
Aquinas's claim in itself, however, determines nothing concerning such
issues. Instead, it merely introduces a certain terminology for speaking in a
fairly general fashion about things familiar to anyone. For of course anyone
who speaks any language at all is familiar with the fact that the intelligent
use and understanding of a meaningful word necessarily involves the ability
to think or be aware of what the word is used for by the users of that
language, that is, the ability to think or be aware of what other users of the
same language think or are aware of when they use the word with under-
standing. It is only the same familiar fact that is expressed otherwise by
saying that words-certain meaningful utterances or inscriptions of a lan-
guage-immediately signify concepts of the users of the language. That is,
they are meaningful to users of the language precisely because they are
conventionally attached to concepts-eertain acts of human awareness or
understanding-of the users of the language. To know this much, however,
we do not have to know what thinking or awareness or understanding is or
what these acts of awareness or concepts are. Whatever it is that accounts
for a human being's ability to think of what a word is used for in a linguistic
community, that is, whatever it is that accounts for a human being's ability
to use a meaningful word with understanding-be it a brain process, a
spiritual modification, or whatever else a philosopher can think up-that is
what we call here a concept immediately signified by the word in question.
Thus, whatever it is that makes me (or you) aware of human beings in
general when I (or you) use the word 'man' in English (say, in the sentence
'I'm looking for a good man'), that is what is here called the concept of
human beings signified immediately by the word 'man' in the minds of
those who understand English (or at least this much of English).
Of course, by saying that words immediately signify concepts of the
human mind Aquinas was not claiming that words are simply the signs
of these concepts. Words are ultimately the signs of the objects of these
concepts; that is, (categorematic) words signify ultimately what the con-
cepts they immediately signify make us aware of.
21
It is a concept of this
21. Of course, syncategorematic terms do not have such ultimate significata in
themselves, as their function is not to signifY but only to cosignifY (consignificare).
See n. 68 of E. J. Ashworth, "Signification and Modes of SignifYing in Thirteenth-
Century Logic: A Preface to Aquinas on Analogy," Medieval Philosophy and Theology
1 (1991): 39-67. For a formal treatment of syncategorematic terms along these
lines, see Essay V of G. Klima, Ars Artium: Essays in Philosophical Semantics, Medieval
and Modern (Budapest: Institute of Philosophy of the Hungarian Academy of
Sciences, 1988), but we are not concerned here with syncategorematic terms.
sort that St. Thomas calls the ratio of the things it makes us aware of.
But, perhaps curiously to the modern reader, according to St. Thomas
this ratio is not only in the intellect but somehow also in the thing
thought of:
The ratio of any thing is what its name signifies, as the ratio of a stone is
what its name signifies. But names are the signs of intellectual concep-
tions, whence the ratio of any thing signified by a name is the concep-
tion of the intellect that the name signifies. And this conception of the
intellect is in the intellect as in its subject, but it is in the thing thought
of as in that which is represented: for the conceptions of the intellect
are certain similitudes of the things thought of. But if the conception
of the intellect were not assimilated to the thing, then the conception
would be false of that thing; for example, if the intellect would think
something that is not a stone to be a stone. So the ratio of the stone is
in the intellect as in its subject, but it is in the stone as in that which
causes truth in the conception of the intellect thinking the stone to be
such and such.
22
In another passage, discussing how the attributes we predicate of God may
apply to him, Aquinas explains in more detail in what sense we can say that
the ratio is in the thing:
We should know that a ratio, as taken here, is nothing else, but what the
intellect apprehends from the signification of some name, and this in
the case of those things that have definition is the definition of the
thing itself, in accordance with what the Philosopher says: 'the ratio
signified by the name is the definition' .23
But some things, which are not defined, are [also] said to have a ratio
in this way, e.g., things such as quantity or quality, which are not defined
because they are most general genera. Nevertheless, the ratio of quality
is what is signified by the name of quality; and it is that from which
quality has it that it is quality. Thus it makes no difference whether
those things that are said to have a ratio have a definition. And so it is
clear that the ratio of wisdom predicated of God is what is conceived of
in the signification of this name, even if divine wisdom itself cannot be
defined. Nevertheless, the name 'ratio' does not signifY this conception
itself, because that is signified by the name ofwisdom, or by some other
name of the thing, but it signifies the intention of this conception, just
as the name 'definition' does and other names of second imposition
do. And thus the second point, namely, the one concerning how the
ratio is said to be in the thing, is also clear. For this does not mean that
22. Resp. ad !ect. Vercell. de art. 108, q. 1. Cf. "Ratio enim quam significat nomen,
est conceptio intellectus de re significata per nomen" (Sn q. 13, a. 4).
23. In meta. lb. 4, le. 16, n. 733.
100 GYULA KLIMA
the intention itself which is signified by the name 'ratio 'would be in the
thing, nor even that the conception to which this intention applies
would be in the thing outside the soul, for it is in the soul as in its
subject; but it is said to be in the thing insofar as there is something in
the thing outside the soul that corresponds to the conception of the
soul, as what is signified [corresponds] to the sign.
24
There are two points in this passage that must be clearly understood
for the subsequent discussion. The first is that the name 'ratio' is a name of
second imposition. The second is that to know the ratio of a thing, and
hence the meaning of its name, we need not know the essential definition
of whatever is conceived of in that ratio. That is, we can clearly know what a
name signifies without knowing what the thing signified by the name is in
itself. 25
24. ISNd. 2, q. 1, a. 3. in corpore
25. Or, using medieval terminology, we can know the quid nominis without
knowing the quid rei. The distinction itself is most aptly characterized by Aquinas's
famous commentator, Cajetan, in the following way:
Just as the quid rei is the thing's quiddity, so the quid nominis is the
quiddity of the name: but a name, as it is the sign of the passions that
are objectively in the soul (from bk. 1. of Aristotle's Perihermeneias) ,
does not have any other quiddity but this, namely, that it is a sign of a
thing understood or thought of. But a sign, as such, is relative to what
is signified: so to know the quid nominis is nothing, but to know what
the name, as a sign, is related to, as what is signified. Such a knowledge,
however, can be acquired by the accidental properties of what is signi-
fied, as well as by its common, or by its essential properties, or simply
by a gesture, or whatever else you like. For example, if we ask a Greek
about the meaning of anthropos, if he points to a man, at once we know
the quid nominis, and similarly in other cases. But to those asking about
the quid rei, it is necessary to give what belongs to the thing adequately,
in the first mode of perseity [i.e., in virtue of its essence-for the four
modes of "perseity," i.e., of predication per se; see In PA lb. 1, le. 10].
This is the essential difference between the quid nominis and the quid
rei: namely, that the quid nominis is the relation of the name to what it
signifies; but the quid rei is the essence of the thing related or signified.
And from this difference follow all the rest that are usually enumerated:
namely, that the quid nominis is of nonentities, complexes, by acciden-
tal, common, and external properties; while the quid rei is of incomplex
entities [grasped] by their proper, essential properties. For a word's
relation can be terminated to non-existents, and complexes, and it can
be clarified by accidental and similar properties, but the thing's essence
can be known only by proper, essential properties of incomplex
things. [Sicut quid rei est quidditas rei, ita quid nominis est quidditas
nominis: nomen autem, cum sit nota earum quae sunt obiective in
anima passionum (ex primo Perihermeneias), non habet aliam quiddi-
tatem nisi hanc, quod est signum alicuius rei intellectae seu cogitatae.
Signum autem ut sic, relativum est ad signatum: unde cognoscere quid
nominis nihil est aliud, quam cognoscere ad quid tale nomen habet
As to the first point, we should know that St. Thomas applies a com-
mon medieval distinction here, roughly comparable to the contemporary
distinction between expressions of an object language and expressions of
a metalanguage. However, for the medievals this distinction did not con-
cern two languages distinct and complete in themselves, one of which is
designed to speak about things other than its own expressions, the other
of which is designed to speak about the expressions of the former. On the
contrary, here we have expressions of the same language distinguished
with respect to their function in speaking about different types of objects,
namely, about either concepts or the objects of these concepts. As we have
seen, for Aquinas a word is meaningful because it immediately signifies
some concept of the human mind. But the word is imposed to signify
ultimately the object of this concept, that is, what the actual exercise of
this concept makes a human being aware of. Some concepts, however,
make us aware not of objects of external reality, but of concepts of the
mind. So just as, say, the concept that makes us aware of human beings
in general is the concept of human beings, so the concept that makes us
aware, say, of concepts that make us aware of things that differ only nu-
merically but not by essential differences (as does the concept of human
beings or the concept of horses or the concept of oak trees) is the concept
of species. On the other hand, concepts that have as their objects things
other than concepts are called first intentions (intentiones primae), and,
correspondingly, their names are called names of first imposition (nomina
primae impositionis). On the one hand, concepts that have concepts as such
as their objects are called second intentions (intentiones secundae) , and,
correspondingly, their names are called names of second imposition (no-
mina secundae impositionis).
The qualification 'as such' is needed here because there are con-
cepts that have both concepts and things other than concepts as their
relationem ut signum ad signatum. Talis autem cognitio potest acquiri
per accidentalia illius signati, per communia, per essentialia, per nutus,
et quibusvis aliis modis. Sicut a Graeco quaerentibus nobis quid nomi-
nis anthropos, si digito ostendatur homo iam percipimus quid nominis,
et similiter de aliis. Interrogantibus vero quid rei oportet assignare id
quod convenit rei significatae in primo modo perseitatis adaequate. Et
haec est essentialis differentia inter quid nominis et quid rei, scilicet
quod quid nominis est relatio nominis ad signatum; quid rei vero est
rei relatae seu significatae essentia. Et ex hac differentia sequuntur
omnes aliae quae dici solent: puta quod quid nominis sit non entium,
complexorum, per accidentalia, per communia, per extranea; quid rei
vero est entium incomplexorum per propria et essentialia. Relatio
enim vocis potest terminari ad non entia in rerum natura, et complexa,
et declarari per accidentalia et huiusmodi, essentia autem rei non nisi
per propria essentialia habetur de entibus incomplexis.]
T. de Vio Cajetanus, "Super Librum De Ente et Essentia Sancti Thomae," in
Dpuscula Omnia (Bergomi, It.:Typis Comini Venturae, 1590), p. 299.
102 GYULA KLIMA SEMANTIC PRINCIPLES OF AQUINAS 103
objects. For example, the concepts of 'being', 'thing', and the like are
clearly of this sort. Still, these concepts are not second intentions since,
even if they do represent concepts as well as other things, they do not
represent concepts qua concepts-that is, qua acts (whatever they are in
themselves) that make humans aware of something-but qua beings or
qua things in their own right (regardless of whatever their function in
human cognition may be). Indeed, even if there were intentions that
have only concepts as their objects but do not represent them qua concepts,
those intentions would not be second intentions. For example, even if it
turned out that what are here called concepts are brain processes, this
stunning discovery would not turn the concept of a brain process into a
second intention, for it does not characterize whatever makes a human
aware of something in terms of the relationship of this act of awareness
to the object of this act but otherwise, as some natural phenomenon in
its own right, certain bioelectric changes in a particular sort of grayish
stuff. 26
However, the concept of species, for example, as interpreted by the
medievals, is a concept of a concept qua concept because it characterizes or
represents concepts as such, that is, in their function of making humans
aware of certain sorts of things in a particular manner.
27
In the same way, as
Aquinas explains in the passage quoted above, the concept of ratio is a
second intention, and hence its name is a name of second imposition
because it represents concepts of other things qua concepts, that is, insofar
as these concepts are related to their objects, naturally representing them
in a certain manner. Hence, what the name 'ratio'immediately signifies is
the intention of ratio, a second intention, which represents concepts in their
function in human cognition. But in that case the intention of ratio applies
to concepts that represent their own objects, and hence the name 'ratio'
refers to these concepts.
As Aquinas goes on to explain, however, we can also say that the ratio
is in the thing. But this is not intended in the sense that the concept by
which the thing is conceived of would be in the thing, for the concept is in
26. This is basically what accounts for the futility of all reductionist attempts to
explain away talk about mental acts in terms of a purely physicalistic language,
regardless of whether what we speak about as mental acts are in fact only brain
processes, that is, modifications of matter, or something else, distinct from such
material modifications.
27. In any case, this is how Aquinas, along with many other medieval authors,
would interpret the standard Porphyrean definition: "species est praedicabile in quid de
pluribus solo numero differentibus," insofar as the relation of predicability of many
things is what definitively characterizes a universal as such, whereas a universal as
such is nothing but an abstract, universal concept which, again, as such, is an act of
human awareness that makes humans aware of the nature of particulars in abstrac-
tion from the individuating conditions with which it only can exist in particulars. Cf.
De entee. 4.
the human mind as in its subject.
28
What this means is that there is some-
thing in the thing corresponding to the ratio-what Aquinas calls the form
or nature of the thing-on account of which the concept applies to this
thing at all.
29
For example, if the concept in question is a universal concept
of many particulars, as the concept of man
30
is a specific concept that
represents humans in abstraction from their accidental, individuating fea-
tures, then the ratio in the mind is this concept itself. But the ratio in the
thing is what is represented by this concept in the thing, and that is what is
referred to as human nature, or humanity, on account of which the thing
represented is a human being. Indeed, what accounts for the intellect's
ability to think of humans in this manner is its capacity to think or be aware
of humans precisely qua humans, that is, to be aware of them only with
respect to what makes them humans, their human nature, disregarding
whatever other features they may have, such as color, gender, weight,
height, virtues or vices, and so on. It is precisely this capacity, the capacity
to abstract the natures or forms of things from their individualizing condi-
tions, that accounts for the very possibility of humans' ability to have
universal concepts. As St. Thomas explains,
When we speak about an abstract universal, we imply two things,
namely the nature of the thing itself, and abstraction or universality. So
the nature itself to which it is accidental that it is thought of, or that it
is abstracted, or that the intention of universality applies to it, exists
only in the singulars, but the nature's being abstracted or its being
thought of or the intention of universality is in the intellect. And we
can see this by the similar situation in perception. For sight sees the
color of an apple without its smell. Thus, if it is asked: Where is the
color that is seen without the smell?-it is obvious that it is nowhere
else, but in the apple. But that it is perceived without the smell happens
to apply to it [accidit ei] on account of sight, insofar as in sight there is
a similitude of color but not of smell. Similarly, humanity that
is thought of exists only in this or in that man; but that humanity is
28. For a discussion and formal semantic reconstruction of the details of
Aquinas's rather intricate doctrine of concepts (which, again, we need not consider
here), especiallyWith respect to their relationships to their subjects, to their objects,
and to what St. Thomas calls natura absolute considerata, see G. Klima, '" Socrates est
species': Logic, Metaphysics and Psychology in St. Thomas Aquinas's Treatment of a
Paralogism," in Argumentationstheorie: Scholastische Forschungen zu den logischen und
semantischen Regeln korrekten Folgerns, ed. K Jacobi (Leiden, Neth.: Brill, 1993),
pp. 489-504.
29. This is why the form of the thing that corresponds to the concept, or ratio,
of ~ e mind .is also called a ratio: "Forma vero quae et ratio nominatur, quia ex ipsa
sumltur. ratIO speciei, dicitur substantia quasi ens aliquid actu, et quasi ens
separabl1e secundum rationem a materia." (In meta. lb. 8, le. 1, n. 1678).
. ~ In the sense of human being-in general, the medieval concept of homo, as
slgmfYing the species of humans, was not regarded as gender specific.
apprehended without its individuating conditions (which is nothing
but for it to be being abstracted, which confers on it the attribute of
universality) is an accidental feature of humanity [accidit humanitati] in
virtue of its being perceived by the intellect, in which there is a simili-
tude of the nature of the species but not of the individuating princi-
ples.
31
But then, concerning the second point mentioned above, we should
clearly realize that even if by having such a concept in our minds we
necessarily are aware of human nature without its individuating conditions,
this does not mean that we necessarily have to be able to say what this
human nature is. For to have this concept means only to be able to think
or to be aware of humans in this abstract and hence universal manner.
But the ability to think of humans in this way is nothing but the ability to
think of humans only qua humans, only in terms of what it is necessary
for something to have if it is to be a human (regardless of any other
features it may have); and this is precisely what is called a human nature,
or a humanity, namely, what makes the thing in question a human being,
regardless of whether we are able to spell out what this nature is in terms
of other concepts by providing an essential definition of human beings.
Therefore, by having this concept of human beings we are necessarily
aware of human nature as such, in abstraction from its individualizing
conditions in particular humans, whether we are able to give any other
characterization of this nature in terms of other concepts or not. Of
course, if we do know the essential definition of the thing in question,
then we are also in a position to give an answer in terms of other concepts
to the question that asks about the nature of the thing: we can tell what
the thing is, what it is that it has to have if it is to be the kind of thing it
is. But, as S1. Thomas insisted in the case of divine wisdom, this is not
required of us in order to have a ratio, or concept, of the thing, an act of
simple apprehension by which we can be aware of the thing with respect
to the form or nature represented by this ratio. Indeed, since this ability
to think of particular things in this abstract, universal manner is what it
is for us to have the abstract concepts, or rationes, of particular things, and
since our words signify immediately our concepts (on account of which
they signify ultimately what we conceive of in these concepts), it is precisely
this human ability that accounts for the presence of universal terms in
human languages, signifying things in the same universal manner, with
respect to the natures of particular things conceived of in their concepts
without their individualizing features. But since this applies to all sorts of
universal terms (in all sorts of human languages), we can state in general
that a universal term immediately signifies an abstract and hence universal
concept of the human mind, on account of which it ultimately signifies
31. S11 q. 85, a. 2, ad 2.
whatever is conceived of by this concept in the particulars (without con-
ceiving their individualizing features), namely, the individualized natures
or forms of particulars, which render them such as to fall under this
concept. For despite our ability to think of particulars in this abstract,
universal manner, and hence our common terms' ability to signify them
in the same way, the form or nature signified by a general term in this
particular thing is, of course, a numerically distinct entity from the form
or nature signified by the same term in that particular thing. As St. Thomas
writes,
It is not necessary that if this is a man and that is a man, then they both
have numerically the same humanity, just as two white things do not
have numerically the same whiteness; but [it is necessary] that the one
be similar to the other in that it has humanityjust as the other: whence
the intellect, considering humanity not as belonging to this thing, but
as humanity, forms an intention that is common to all.
32
Thus, the abstractive consideration of the intellect is able to conceive
of these particular, individualized forms in a universal manner (consider-
ing, say, this whiteness or that whiteness only as whiteness but not consid-
ering this whiteness as belonging to this thing or that whiteness as
belonging to that thing). It is therefore able to confer universal meaning
on certain sounds and inscriptions as parts of a human language. However,
it is only this humanity, say, the humanity of Socrates, or that humanity,
say the humanity of Plato, or this whiteness of this white sheet of paper,
or that whiteness of that white sheet of paper that exists in rerum natura.
For even if, say, this whiteness or that whiteness can be considered without
considering whether it belongs to this thing or to that thing, no whiteness
can be without being the whiteness of this thing or of that thing, that is,
without belonging to this thing or to that thing. And this is necessarily so
because for this whiteness to be is nothing but for this thing to be white
and vice versa, and of course it is only some particular thing that can be
white. Thus, in this semantic conception, for some particular thing to be
white is nothing but for its particular whiteness (the particular quality that
is represented in a universal manner by the human concept that renders
the term 'white' in English meaningful) to be. But then this is why, ac-
cording to St. Thomas, we use the verb of existence even when we want
to express only the simple fact that a thing actually is white: "Since actuality,
what is principally signified by the verb 'is', is in general the actuality of
all forms, whether substantial or accidental actuality, when we want to
signify any form or actuality to be actually in a subject, we signify this by
the verb 'is'. "33
32. 2SNd. 17, q. 1, a. 1.
33. In Peri lb. 1, le. 5, n. 22.
106 GYULAKLIMA SEMANTIC PRINCIPLES OF AQUINAS 107
Of course, this is just one of several of Aquinas's formulations of the
theory of predication that historians of medieval logic duly dubbed the
inherence theory.34 Stated as a general, formal semantic principle, the inher-
ence theory claims,
(1) The predication of a common term F of an individual u is true
if and only if the form ultimately signified by F in u actually is,
that is, exists. 35
Again, I have to stress that despite possible modern worries to the contrary,
this claim, as a formal semantic claim, does not involve the introduction of
any sorts of "mysterious" or spurious entities. This apparently atavistic talk
about "forms" and "natures," clearly of the same breed as Moliere's
mocked talk about dormative powers, in this semantic interpretation need
not commit us to any curious sorts of entities other than those familiar to
anyone nor to any sort of epistemological nonsense, such as having to
have some "metaphysical intuition" into the ordinarily hidden nature of
things.
By saying that a general term ultimately signifies a form in the par-
ticulars that render those particulars such that they actually fall under
the concept immediately signified by the term in question, we do not
need to posit some new sorts of mysterious entities in the thing. Indeed,
we need not do so here any more than we need to posit any sorts of
further, mysterious entities (beyond those anyone would agree to be fa-
miliar with) when we talk about concepts in this semantic context. (Re-
member, talk about concepts in itself neither implies nor excludes the
possibility that the acts of awareness that, in relation to their objects, we
talk about as concepts are in themselves just brain processes. However,
there being such acts of awareness in the intended sense is just such a
familiar fact of human existence that denying their existence would yield
34. Cf., for example, L. M. de Rijk, "Introduction," to P. Abaelard, Dialectica
(Assen, Neth.: 1956), pp. 37-38; D. P. Henry, Medieval Logic and Metaphysics (Lon-
don: 1972), pp. 55-56; P. T. Geach, "Nominalism," God and the Soul (London: 1969).
35. Cf., for example, "Oportet enim veritatem et falsitatem, quae est in oratione
vel opinione, reduci ad dispositionem rei sicut ad causam. Cum autem intellectus
compositionem format, accipit duo, quorum unum se habet ut formale respectu
alterius: unde accipit id ut in alio existens, propter quod praedicata tenentur
formaliter. Et ideo, si talis operatio intellectus ad rem debeat reduci sicut ad
causam, oportet quod in compositis substantiis ipsa compositio formae ad mate-
riam, aut eius quod se habet per modum formae et materiae, vel etiam compositio
accidentis ad subiectum, respondeat quasi fundamentum et causa veritatis, compo-
sitioni, quam intellectus interius format et exprimit voce. Sicut cum dico, Socrates
est homo, veritas huius enuntiationis causatur ex compositione formae humanae
ad materi<l;m individualem, per quam Socrates est hic homo: et cum dico, homo est
albus, causa veritatis est compositio albedinis ad subiectum: et similiter est in aliis"
(In meta. lb. 9, le. 11, n. 1898).
a self-defeating claim) .36 Thus, just as we could start talking about con-
cepts without having to commit ourselves to any particular theory about
what concepts in themselves are, so we can start talking about the 'forms'
of things as the ultimate significata of our words without having to take
a metaphysical stance about what these 'forms' in themselves are. Of
course, talking about these ultimate significata as forms, rather than any-
thing else, indicates the close historical relationship between this semantic
conception and a hylomorphist metaphysics. This, however, does not
mean that this semantic approach in itself logically implies a hylomorphist
metaphysics in genera13
7
or any specific version of it in particular. In fact,
such an implication would leave no room for metaphysical disagreements
to be disputed within the same semantic framework. But of course, many
of the great metaphysical disputes of high scholasticism can be construed
as revolving precisely around the distinctness (or identity) of the semantic
values commonly assigned to several expressions within basically the same
semantic framework.
This is most obvious in the famous debate concerning the unicity and
plurality of substantial forms. Recapitulated in semantic terms, the point of
the debate was to decide whether the ultimate significata of the substantial
predicates of the same individual are the same or distinct. This question is
left open by the semantic principles of signification and predication, but it
36. For more on this subject see the comments on rule (l) in section VI.
37. In fact, that the ultimate significata of common terms need not necessarily
be regarded metaphysically as forms in all cases was a commonplace also for
thinkers who otherwise were committed to a hylomorphist metaphysics. As St.
Thomas wrote, "Dicendum est quod illud a quo aliquid denominatur non oportet
quod sit semper forma secundum rei naturam, sed sufficit quod significetur per
modum formae, grammatice loquendo. Denominatur enim homo ab actione et ab
indumento, et ab aliis huiusmodi, quae realiter non sunt formae" (QDPq. 7, a. 10,
ad 8). Cf. also, for example, ''Verum ne fallaris cum audis denominativum a forma
denominante oriri, et credas propter formae vocabulum quod res denominans
debet esse forma eius quod denominatur, scito quod formae nomine in hac materia
intelligimus omne illud a quo aliquid dicitur tale, sive illud sit secundum rem
accidens, sive substantia, sive materia, sive forma." Thomas de Vio Cajetan, Scripta
Philosophica: Commentaria in Praedicamenta Aristotelis, ed. M. H. Laurent (Rome:
Angelicum, 1939), p. 18. In general, it is precisely this point that lies at the bottom
of the distinction between extrinsic and intrinsic denomination. Cf. also "Nam, sicut
dicit Commentator, duodecimo Metaphysicae, grammaticus videt in multis differre
dispositionem et dispositum, et sic movetur ad imponendum eis nomina diversa, ut
'albedo' et 'album'; et quia non est ejus inquirere an in omnibus vel in quibus sic
differant dispositio et dispositum, ipse secundum similitudinem ad illa in quibus
manifeste differunt imponit etiam aliis nomina per modum dispositionis et dispo-
siti, seu determinationis et determinabilis, vel etiam determinati, derivando ab
abstracto concretum vel e converso, relinquens metaphysico considerationem an
illa nomina supponant pro eodem vel pro diversis, propter quem diversum modum
significandi grammaticalem illa nomina habent diversos modos praedicandi. (Jean
Buridan, Lectura de Summa Logicae: De Praedicabilibus c. 7, n. 4, H. Hubien's unpub-
lished edition).
has far-reaching ramifications in both metaphysics and theology.38 Again,
the same semantic framework leaves undetermined the question ofwhether
the significata of such substantial predicates should be regarded as identical
with or distinct from the significatum of the predicate 'is' in the same
individual. Hence the need for Aquinas to deploy several metaphysical argu-
ments for his famous thesis of the real distinction in creatures between
essence and existence, that is, between the significata of substantial predi-
cates and the significata of the verb 'is', while, of course, the same semantic
principles allow him to state the identity of these significata in the case of
God.
39
But the same semantic principles do not determine even whether
the significatum of a predicate in an individual is distinct from this individ-
ual itself. So again, anyone having ontological, that is, metaphysical qualms
about "multiplying entities" is free to construe the significata of predicates
with respect to the particulars as being the particulars themselves, at least
as far as the semantics is concerned.
4o
However, of course, such a decision
will again have its metaphysical consequences, so anyone wishing to do so
38. An excellent historical summary of the debate with ample further refer-
ences is provided by D. A. Callus, "Forms, Unicity and Plurality of," in New Catholic
Encyclopedia, Catholic University of America (New York: McGraw-Hill, 1967-79).
39. See, for example, De ente c. 5.
40. Without intending to go too much into technicalities, let me just give
here a brief indication of how this claim can be made exact in a formal semantic
system. If P is a common predicate, then the ultimate significate of P in respect
of an individual u at a time t may be assigned by the semantic function SGT in
a model in the following way: SGT(P) (u)(t) EWU{O}, where W is the domain of
the model, t is some time, and 0 is a zero entity (OEtW). The case SGT(P)(u)(t)
= 0 represents the situation that P signifies nothing in u, as, for example, the
predicate 'red' signifies nothing in respect of the number 2. It is also reasonable
to stipulate that SGT(P) (0) (t) = O. Then, if the set of things that are actual at a
certain time t, A( t), is a subset of W, in accordance with the inherence theory of
predication we can say that the predication of P of an individual u at a certain
time t in a present-tense sentence (i.e., the copula of which consignifies the time
t of the utterance) is true if and only if SGT(P)(u)(t)EA(t). Now this much of
semantics can certainly stay in place whether in a particular model SGT(P) (u) (t)
= u or SGT(P)(u)(t) = u. But the former case represents precisely the simple
ontology in which the significate of P in u at t is not an entity distinct from u but
is u itself. In such an ontology, for example, the predicate 'round' in respect of
a round thing, say, a billiard ball, would signify the ball itself, but in respect of,
say, a cube, it would signify nothing (SGT('round') (ball) (t) = ball,
SGT('round') (cube) (t) = 0). Nevertheless, in general, I do not intend to go into
technicalities here. I will only indicate in the notes how the semantic principles
could be given exact formulations in a model-theoretical framework. For a com-
plete semantic system constructed along these lines to represent St. Thomas's
ontology, see G. Klima, "On Being and Essence in St. Thomas Aquinas's Meta-
physics and Philosophy of Science," Knowledge and the Sciences in Medieval Philosophy:
Proceedings of the Eighth International Congress of Medieval Philosophy, ed. S. Knuuttila,
R.Tyorinoja, and S. Ebbesen, vol. 2, series B19 (Helsinki: Luther-Agricola Society,
1990),210-221. For the issue of the independence of this semantic construction
from that particular ontology, see Essay V of my Ars Artium.
will have to face, among others, Aquinas's metaphysical arguments to the
effect that in material beings such an identification is impossible, whereas
in the case of God it is more than justified.
41
Finally, not only do these
formal semantic principles, namely, the inherence theory of predication
and the corresponding theory of signification, leave undetermined the
questions of the identity or distinctness of the ultimate significata and
supposita of common predicates, but they also leave undetermined the
question of whether in the thing there really is in actual being something
that corresponds to a predicate-an actually inherent form-even when
the predicate is true of this thing. For certain predicates, namely, negative
or privative predicates, which in their very concept involve the lack of what
their opposite signifies, are true of something precisely because the oppo-
site significatum is lacking in this thing. So for the former to be is for the
latter not to be. But then the being of the significatum of the negative or
privative predicate is nothing but the nonbeing of the significatum of the
opposite positive predicate. Therefore, since nothing can be both being
and nonbeing in the same sense, the significatum of the privative or nega-
tive predicate cannot be said to be in the same sense as the significatum of
the opposite positive predicate. Indeed, this is precisely what seems to be
the primary motivation for the Aristotelian distinction between the two
senses of 'being' with which we started our discussion. As, in a different
context, Aquinas writes,
As was said above, 'being' is predicated in two ways. In one way it
signifies the essence of a thing existing outside of the soul; and in this
way the deformity of sin cannot be said to be a being, for it is a certain
privation, but privations do not have an essence in the nature of
41. See, for example, S11 q. 3, a. 3; Qy,odl. 2, q. 2, a. 2[4]. However, I think
it is also interesting to note here that the possibility of identifying everything with
substance even in the case of material substances was considered already by a
medieval philosopher, John of Mirecourt, who denied all sorts of metaphysical
distinctions corresponding to semantic distinctions in rebus, even the distinction
between substance and accident. In fact, it is somewhat ironical that it took an
"arch-nominalist" like Jean Buridan, someone who himself reduced the number
of distinct ontological categories to three (namely, to substance, quantity, and
quality), to deploy several metaphysical arguments against this "obscure and dan-
gerous" doctrine. Cf. M. M. Adams, 'Things versus 'Hows,'" in How Things Are,
ed.]. Bogen and]. E. McGuire (Dordrecht, Neth.: Reidel, 1985), pp. 175-88, esp.
pp. 179-80. Of course, there are substantial differences between Aquinas's and
Buridan's (and probably also Mirecourt's) approach to semantics. The point, how-
ever, is that for both of them, semantics does not dictate to ontology: however
differently we may pick out things for our consideration, whether we do pick out
distinct things or just the same thing differently is not determined by semantics,
and so this has to be determined by careful metaphysical considerations. For more
on this issue and the particular differences between Aquinas's and Buridan's
approach to semantics, see G. Klima, "Ontological Alternatives vs. Alternative Se-
mantics in Medieval Philosophy," in Logical Semiotics, S-EuropeanJournal for Semiotic
Studies, 3:4 (1991): 587-618.
110 GYULA KLIMA SEMANTIC PRINCIPLES OF AQUINAS III
things. In the second way ['being'] signifies the truth of a proposition;
and in this way a deformity is said to be, not that it has being in the
thing, but because the intellect compounds a privation with its subject
as if it were some form. Therefore, just as from the composition of a
form with a subject or with matter there results some substantial or
accidental being; so does the intellect signify the composition of a
privation with a subject by some being. But this being is only being of
reason, for in the thing it is rather non being.
42
Thus, whatever common predicate we substitute for F in (l) above, (l) will
hold; only the sense of 'is' or 'exists' in it may vary, depending on the kind
of significatum F has in u. In fact, on the basis of this point one may already
surmise how the different senses of 'being' distinguished by Aquinas are
related to the different kinds of predicates that can occur in actual predica-
tions. However, before considering this relationship we have to consider the
function of common terms in the other necessary component of an actual
predication in a proposition, namely, the subject term, which supplies the
referent, or suppositum, for the predication.
v: SUPPOSITION AND SIGNIFICATION OF ABSTRACT
AND CONCRETE TERMS
So far I have been gathering and discussing Aquinas's semantic ideas con-
cerning common terms as they are predicable of individuals. As we saw,
according to this conception, a common term in an actual predication, that
is, as the predicate of a categorical proposition, ultimately signifies some
individualized property of the individual (or individuals) of which it is
predicated, that is, of the individual (or individuals) that is (or are) referred
to by the subject of the proposition. But common terms can themselves
function as subjects, and then, according to medieval logicians, they also
have the function to refer to (or to use the modern transliteration of
medieval terminology: supposit for) the individuals that fall under them.
43
In
this referring function, then, despite the fact that they signify what we called
the forms of the particulars (that is, their ultimate significata in the particu-
lars, whatever those ultimate significata are in themselves), they normally
refer to, or supposit for, the particulars themselves. As St. Thomas says,
In respect of any name, we have to consider two things, namely, that
from which the name is imposed (which is called the quality of the
42. 25Nd. 37, q. 1, a. 2.
43. For formal reconstructions of the relevant medieval ideas and ample
references to and discussion of the enormous amount of contemporary literature
they generated, see Essays II-IV of my An Artium.
name) and that on which the name is imposed (which is called the
substance of the name). And the name, properly speaking, is said to
signify the form, or quality, from which the name is imposed, and is said
to supposit for the thing on which it is imposed.
44
So, for example, the term 'man' signifies human nature in abstraction
from the singulars immediately and signifies individual human natures
ultimately, but normally-say, in the proposition 'A man is white'-it sup-
posits for the things that actually bear the nature it signifies, namely,
individual humans.
45
But then this proposition is true if and only if the form
ultimately signified by its predicate term in at least one of the supposita of
the subject term actually exists, that is, if and only if the individual whiteness
of at least one of the individuals that actually has humanity-the whiteness
of a human being-actually exists.
44. 35Nd. 6, q. 1, a. 3. It is important to note here that although St. Thomas
sometimes also contrasts id a quo with id ad quod nomen imponitur to distinguish the
etymology ofa name from its proper signification (the stock example being "lapis";
d., e.g., 511. q. 13, a. 2, ad 2. ), we must not confuse the two distinctions. As St.
Thomas himself pointed out: "nomen dicitur ab aliquo imponi dupliciter: aut ex
parte imponentis nomen, aut ex parte rei cui imponitur. Ex parte autem rei nomen
dicitur ab illo imponi per quod completur ratio rei quam nomen significat; et haec
est differentia specifica illius rei. Et hoc est quod principaliter significatur per
nomen. Sed quia differentiae essentiales sunt nobis ignotae, quandoque utimur
accidentibus vel effectibus loco earum, ut VII Metaphys. dicitur; et secundum hoc
nominamus rem; et sic illud quod loco differentiae essentialis sumitur, est a quo
imponitur nomen ex parte imponentis, sicut lapis imponitur ab effectu, qui est
laedere pedem. Et hoc non oportet esse principaliter significatum per nomen, sed
illud loco cuius hoc ponitur" (QDV q. 4, a. 1, ad 8). Thusd a qd ndomen lapidis
imponitur, ex parte rei, is the ratio seu natura lapidis, that is, the nature of stones,
sometimes also referred to as lapideitas, namely, what is signified by the term lapis,
whatever it is, and whether we know it in terms of an essential definition or not;
whereas id a quo nomen lapidis imponitur, ex parte imponentis, is the accidental property
of stones that they tend to hurt the foot (laedere pedem) , which may have provided
the motivation for so naming them. (Whether Isidore of Seville's etymology is
actually correct or not is irrelevant here.) For a more detailed discussion of the two
distinctions and references to their earlier history, see Ashworth, "Signification,"
pp. 39-67, esp. pp. 47-50.
45. I say "normally" because it is only in some special context that a term is
made to refer to what it normally signifies. As Aquinas says, "Quia enim forma
significata per hoc nomen 'homo', idest humanitas, realiter dividitur in diversis
suppositis, per se supponit pro persona; etiamsi nihil addatur quod determinet
ipsum ad personam, quae est suppositum distinctum. Unitas autem sive communi-
tas humanae naturae non est secundum rem, sed solum secundum considera-
tionem, unde iste terminus 'homo' non supponit pro natura communi, nisi propter
exigentiam alicuius additi, ut cum dicitur, 'Homo est secies'" (511 q. 39, a. 4; d. 513
q. 16, a. 7). Of course, those familiar with medieval logic will immediately recognize
here a case of simple supposition, as opposed to personal supposition. However, as I said
before, in this article I do not intend to go into the technicalities of supposition
theory. Following Aquinas's practice, when I will speak about a term's supposita
without any qualification, I will intend the term's personal supposita.
But here we should note again that this semantic distinction between
suppositum and significatum, as such, need not posit any real distinction
between what is signified and what is supposited for by a common term in
a proposition. Even if, according to Aquinas's metaphysical views, a real
distinction corresponds to this semantic distinction in the case of material
beings (and thus, for example, a human being is not his or her humanity,
and a wise person is not his or her wisdom), nevertheless in the case of God
there is no corresponding real distinction between what is signified and
what is supposited for, although the semantic distinction is still in force (for
example, the predicate 'wise' is still related to divine wisdom as to its
significatum and to God as to its suppositum since, as a consequence of
divine simplicity, God is identical with divine wisdom).46 Thus, at least as far
as common concrete terms are concerned, their mode of signification
(modus significandi) is such that they signify their ultimate significata in
particular things (whether these significata are actual or merely potential).
But as subjects of categorical propositions, they supposit for those particular
things in which their ultimate significata actually exist.
47
Nevertheless, what
they ultimately signify and what they supposit for mayor may not be
identical, depending on the nature of the things signified and supposited
for. Thus, the question of the identity or distinction of their supposita and
significata is a metaphysical question, not determined by the semantic
distinction. However, the abstract counterparts of these concrete terms
differ from them precisely in their mode of signification, insofar as abstract
terms both signify and supposit for what their concrete counterparts signify.
But they do not supposit for the supposita of their concrete counterparts,
except when these supposita are identical with the significata. This is why
we can refer to the significata of concrete terms by their abstract counter-
parts and say, for example, that the term 'man' signifies humanity (whereas
it supposits for humans) and that a man is not identical with his humanity,
46. Cf. "Respondeo dicendum, quod natura et suppositum naturae in quibus-
dam differunt re et ratione, sicut in compositis; in quibusdam autem ratione et non
re, sicut in divinis" (3SN d. 11, q. 1, a. 4, in corp). Cf. also "propter divinam
simplicitatem, consideratur duplex realis identitas in divinis eorum quae differunt
in rebus creatis. Quia enim divina simplicitas exc1udit compositionem formae et
materiae, sequitur quod in divinis idem est abstractum et concretum, ut deitas et
deus. Quia vero divina simplicitas exc1udit compositionem subiecti et accidentis,
sequitur quod quidquid attribuitur deo, est eius essentia, et propter hoc sapientia
et virtus idem sunt in deo, quia ambo sunt in divina essentia" (S71 q. 40, a. 1, ad 1).
47. We should also add, "relative to the time and modality of the copula." But
we need not consider these complications here. Cf., however, the passage referred
to in n. 53 below and the related discussion in the text. In any case, using the
notation introduced in n. 40, we could state this rule concerning assertoric (i.e.,
nonmodal, nonampliative) propositions in a formal system as follows:
SUP(P) (t) E{u: SGT(P) (u) (t) EA(t)}, provided {u: SGT(P) (u) (t) EA(t)} # 0, other-
wise SUP(P) (t) = 0, where t is the time connoted by the copula of the proposition.
From this and from the rules of n. 40, it can be seen that an affirmative proposition
whose subject supposits for nothing will have to be false.
whereas God (the suppositum of the term 'God') is identical with deity, i.e.,
the significatum of the term 'God', and the suppositum (and significatum)
of the term 'deity'.48
Note here that in the previous sentence the term 'humanity' is used to
refer to what the term 'man' signifies. Indeed, in general, the function of
abstract terms seems to be precisely this, namely, to afford us the linguistic
means to refer in a proposition to the significata of their concrete counter-
parts, since the concrete terms do not refer to their significata but to the
things actually having their significata. Thus, whenever we need to refer not
to the supposita but to the significata of a concrete term, we need an
abstract term, corresponding to the concrete term. We can therefore see
how this semantic framework generated in scholastic philosophy and sci-
ence the need for abstract terms even in cases in which ordinary usage did
not supply any. It was this systematic need that yielded all the 'barbarisms'
(such as lapideitas, asininitas, or even Sorteitas, and the Scotists' famous
haecceitas and their like) that later on were to evoke the contempt of an
emerging, new, nonscholastic intelligentsia, which originally refused and a
couple of centuries later simply forgot to think in terms of the semantic
principles governing scholastic discussions. But, of course, this is already a
different and enormously complicated story, which we cannot go into here.
Still, it is important to keep in mind that it is only as a result of this
'historical amnesia' that we are now in a situation in which it takes this
meticulous gathering and arranging of bits and pieces ofAquinas's thought
to piece together his semantic conception, consisting of principles that
constitute a certain form of discourse, which was common in his time but
for various reasons was gradually abandoned. In any case, we have now
gathered enough of these bits and pieces to state at least some of the most
general semantic principles that underly Aquinas's thought.
VI. AQUINAS'S SEMANTIC PRINCIPLES
What follows will not be an exhaustive statement of what might be dubbed
"Aquinas's semantics." Here I only summarize and briefly comment on
some of the most general principles we could glean from the foregoing
48. Accordingly, if [P] is the abstract counterpart of P, then we can say that
SGT( [P] )(u)(t) = SGT(P)(u)(t), and SUP( [P]} (t) = SGT(P)(SUP(P)(t) )(t), if
SGT(P)(SUP(P)(t) )(t) EA(t), otherwise SUP( [P]} (t) = 0. Thus, in a model repre-
senting Aquinas's metaphysics, the following will hold: SGT('God')
(SUP('God')(t) )(t) = SUP('deity')(t) = SUP('God') (t); whereas SGT('man')
(SUP('man') (t) )(t) = SUP(,humanity') (t) # SUP('man')(t). But of course other
models may represent different metaphysics. In particular, the same semantic
system permits even the case when SGT('man') (SUP('man') (t (t) = SUP('hu-
manity')(t) = SUP('man') (t).
114 GYULA KLIMA
discussion. On this basis I will be able to answer all the questions and doubts
that have emerged in the course of this discussion.
(1) Common terms signify human concepts immediately, and the
forms of particulars represented by these concepts ultimately.
The point of this principle is that the signification of a common term is
something it has because of being associated with some act of human
understanding. An utterance or inscription becomes a term of a language
precisely because of this association with, or subordination to, such a mental
act. Once this conventional relation of subordination or immediate signifi-
cation is established by an act of imposition or namegiving, the utterance
or inscription becomes a term. Such a term, then, ultimately signifies the
features, or forms, of the particular things represented by the concept to
which it is subordinated.
49
Note that what I am here calling concepts are just certain acts of the
human understanding on account of which a human being is able to use
the terms of a language with understanding, whatever those acts are in
themselves. Again, what I am here calling forms are just the features of the
things represented by these acts of understanding, whatever those features
are in themselves. Thus, concepts and forms, as just the immediate and
ultimate significata (respectively) of our terms, are to be understood here
in such a noncommittal way that any denial of their existence (in the
intended sense) should be self-defeating. In this way, to claim, for example,
that "I'm sure I have no such concepts in my head" in the intended sense of
'concept' would amount to the admission that "I can't use a single word
with understanding, not even the words I'm now uttering." Again, if some-
one were to say, "Things just don't have such forms," then in the intended
sense of 'form' this claim would imply the admission that "Nothing is
characterized by anything, and so neither is the proposition I just uttered
characterized by truth."
However, admitting such concepts and forms as (respectively) the
immediate and ultimate significata of our terms need not involve any
knowledge of what these significata are in themselves. In general, we can
think of and hence signify and refer to things without knowing their nature.
In fact, for most of the things we talk about we do not know what they are,
whereas, of course, we know what we signify by their names. We are merely
unable to give an essential definition of what is thus signified. For example,
49. Formally, if CON(m,t) (P) is the concept to which P is subordinated in a
mind m at a certain time t-that is, the immediate significate of P in m at t-and
CON(m,t) (P) (u) (t) is what is represented by this concept in respect of the particu-
lar u at time t, then the ultimate significate of P in u at t can be defined as follows:
SGT(P)(u)(t) = CON(m,t)(P)(u) (t). Notice how this formulation allows for con-
ceptual change and hence for change of meaning ofP. For details in this regard see
Essay V of my Ars Artium.
we can know what we signify by the term 'diamond' in English, without
knowing that for something to be a diamond is for it to be a tetrahedrally
crystalized allotrope of carbon. To know the latter is to know that the form
signified by the term 'diamond' is the same as the form signified by the
phrase 'tetrahedrally crystalized allotrope of carbon' (provided this is in-
deed what it is for something to be a diamond, but that is a question to be
answered by natural scientists). But whether or not we are in possession of
this piece of knowledge, by the term 'diamond' we signify this form, which
we are certainly entitled to baptize by coining an abstract term, say,
'diamondhood', to be able to refer to it in further discussion. Of course,
later on, it may turn out that what we referred to as 'diamondhood' is
nothing but the tetrahedral arrangement of carbon atoms held together by
covalent bonds, which we may further identify as the carbon atoms them-
selves thus and so arranged, and so on. But none of these later results
should affect the validity of the original, semantic principle, which specifies
how our terms are related to what they signify but says nothing about what
these significata are in themselves. That is, it says nothing about what
ontological category they belong to, or what combination of other terms
would enable us to pick out precisely the same significata.
(2) A concrete common term P is true of a particular thing u iff the
form (ultimately) signified by P is actual in u.
50
This principle is just a reformulation of the inherence theory: 'is actual' is
meant to be taken in the same sense as 'is' or 'exists', the actual sense of
which is determined by the signification of the actual substituents of P. For
example, if P signifies a privation, then, of course, its significata can be
actual only in the sense in which privations can be actua1.
51
They do not
posit any entity in reality but rather indicate the lack of some real being in
actual reality. Thus, for example, if 'bald' is construed as a privative term
that signifies the lack of hair where there should be hair by nature, then the
predication of 'bald' of a bald man is true because of the lack of hair on the
bald man's head. It is the actuality of his baldness, the actual lack of hair on
his head, that verifies the predicate 'bald'. But, of course, the actuality of
this lack of hair does not make his baldness a real being superadded to the
other, real entities of the world. Such a superadded real being would be the
hair (or a wig, for that matter). But since it is precisely the actual lack of this
superadded entity that is the actuality of baldness, the latter cannot be
regarded as another such being, adding to the number of entities of the
same kind. Hence, even if baldness is actual, it is in a sense different from
the sense in which those real entities are. Indeed, even among real entities,
50. Cf. n. 40 above.
51. Cf. "et sicut in rebus, quae extra animam sunt, dicitur aliquid in actu et
aliquid in potentia, ita in actibus animae et privationibus, quae sunt res rationis
tantum" (In meta. lb. 5, le. 9, n. 13).
116 GYULA KLIMA
which are not privations or just other beings of reason, there are further
differences in their modes of being, reflected in the analogous senses of
'being' distinguished by Aristotle and Aquinas. I will have more to say about
these in the next section.
(3) A concrete common term P supposits for (refers to) a particu-
lar thing u as the subject of a proposition iff the form signified by
Pin u is actual (relative to the time and modality of the copula of
the proposition and to the ampliation of the predicate).52
The actuality of the significata of a common term is what makes it true
of particular things, and so, if the term in question occurs as the predicate
term of an affirmative categorical proposition, this will render the propo-
sition true of the individuals referred to in it. But if the term in question
is the subject term of a proposition, then its function is not to state some-
thing of the things referred to in the proposition but to refer, to supply
the referents or supposita for the predicate term to be true of. However,
even in this referring function, what determines whether a certain par-
ticular is actually supposited for by the common term as the subject term
of a proposition is whether the form signified by the term in the particular
is actual at the time consignified5
3
by the copula (or the verb implying the
copula) of the proposition. For example, in the sentence 'A dinosaur is
running', the subject term 'dinosaur' should supposit for actual dino-
saurs-that is, things that actually have the nature of a dinosaur-for the
sentence would be true only if the things actually having the significate of
the subject would also actually have the significate of the predicate at the
present time of the utterance of the proposition. But since by all prob-
ability there is no such thing in actuality (nothing has a dinosaur nature
in actuality; i.e., nothing is a dinosaur), no such thing has the significate
of the predicate in actuality, whence the sentence is actually false. However,
in the proposition 'A dinosaur was running', the time consignified by the
copula is some past time relative to the present time of the utterance of
the proposition. So this proposition is true if the things which have or had
the significate of the subject in actuality also had the significate of the
predicate in actuality at some past time relative to the utterance of the
proposition. But in that case, since some things did have dinosaur nature
in the past (i.e., there were dinosaurs), and by all probability some of them
were actually running at some past time (i.e., they had the significate of
the predicate, the act of running, in actuality), this proposition is probably
actually true.
Similar considerations would apply to other tenses, modalities, and
special predicates habentes vim ampliandi, as medieval logicians would put
52. Cf. n. 47 above.
53. Cf., for example, In Peri lb. 1. le. 5.
it. These are predicates that have the force of extending the range of
reference of their subject terms beyond the set of things that actually exist
at the time of the utterance of the proposition, such as verbs and participles
that signify mental acts (say, 'think', 'imagine', etc.). But we need not go
into those details here.
54
All we need to note here is that while reference
depends on signification-for whether a particular thing has or does not
have the significate of a common term ultimately determines whether or
not it is supposited for in a proposition-nevertheless, this determination
takes place in a propositional context, which further determines how these
significata contribute to the determination of supposition. 55
(4) The significata of the abstract counterpart [P] of a concrete
common term P are the same as those of P, but as the subject of a
proposition [P] supposits for the ultimate significata of P (which
are also the significata of [P], of course), provided they are actual
relative to the time and modality of the copula of the proposition.56
The ultimate significata of abstract and concrete terms (and, hence, also
their significations) are the same. Both 'white' and 'whiteness' signify the
same forms, the individual whitenesses of particular things, whether they
are actually or merely potentially white.
57
(There are, of course, things
which cannot possibly be white, say, numbers, which are extra genus colons,
54. Especially since St. Thomas never addressed ex professo the issue of
the medieval theory of reference. However, that he knew well and applIed con-
sciously the theoretical apparatus of the sophistae, the teachers of logic of the Faculty
of Arts, is obvious from his references to their doctrine in the following places: 1SN
d. 21, q. 1, a. la, ad 2-um; STI q. 31, a. 3, ad 3-um; STI q. 39, a. 4, obj. 1; STI q. 39,
a. 5, ad 5-um. For a discussion of Aquinas's doctrine in relation to the logical theory
of his contemporaries, see Ashworth, "Signification," pp. 39-67. Also, for a discus-
sion of ampliatio in particular, see G. Klima, "Old Directions in Free Logic: Existence
and Reference in Medieval Logic," in New Directions in Free Logic, ed. K Lambert
(Sankt Augustin bei Bonn: Akademia Verlag, forthcoming). . . .
55. Although Peter of Spain, for example, regards natural supposztwn as the kmd
of supposition a term has absolutely (per se), apparently even outside the context
of a proposition. But even according to him, within a propositional context the
actual (what he calls accidental) supposition of a term is determined by the context.
See Peter of Spain, Tractatus, ed. L. M. de Rijk (Assen, Neth.: Van Gorcum, 1972),
p. 81. For some discussion and further references see Klima, "Old Directions in Free
Logic."
56. Cf. n. 48.
57. Cf., for example, the detailed discussion of denomination by Cajetan: "non
debet denominativum differre a nomine formae denominantis in significatione.
Hoc enim esset differre non solo casu sed significatione [interpunction mine]. Et
consequenter diffinitione et essentia idem album et quum (ut
infra docet Aristoteles) album puram qualitatem sIgmficat [d. WIth thIS In met.a.
5, le. 9, n. 894]. Differentia autem in modo significandi inventa inter denommatI-
vum et denominans non excluditur per ly solo casu, quoniam talis differentia comes
est differentiae secundum casum" (ScriptaPhilosophica, pp. 16-17).
62. Of course verbs in this theory are to be analyzed into a copula and a
participle. In fact, even in languages like Russian or Hungarian, in which no copula
is needed for the formation of present-tense categoricals (although the two lan-
guages belong to different families oflanguages), the well-formedness of sentences
in the past and future tenses requires the addition of the appropriate forms of the
verb corresponding to the verb 'be'. This indicates that on the conceptual level a
copula is present also in the present-tense sentences, even if it is unmarked in their
surface syntax.
63. Cf. "Deinde cum dicit amplius autem ponit alium modum entis, secundum
quod esse et est, significant compositionem propositionis, quam facit intellectus
componens et dividens. Unde dicit, quod esse significat veritatem rei. Vel sicut alia
translatio melius habet quod esse significat quia aliquod dictum est verum. Unde
veritas propositionis potest dici veritas rei per causam. Nam ex eo quod res est vel
non est, oratio vera vel falsa est. Cum enim dicimus aliquid esse, significamus
propositionem esse veram. Et cum dicimus non esse, significamus non esse veram;
et hoc sive in affirmando, sive in negando. In affirmando quidem, sicut dicimus
quod Socrates est albus, quia hoc verum est. In negando vero, ut Socrates non est
albus, quia hoc est verum, scilicet ipsum esse non album. Et similiter dicimus, quod
non est diameter incommensurabilis lateri quadrati, quia hoc est falsum, scilicet
non esse ipsum non commensurabilem" (In meta. lb. 5, le. 9, n. 895).
As we have seen, the theory of truth and predication outlined above in itself
seems to require a systematic correspondence between the different kinds
of significata of several sorts of predicates and the various analogical senses
of 'being' distinguished by Aristotle and Saint Thomas. We have also seen,
however, that according to St. Thomas's theory of the copula, the copula of
an affirmative categorical proposition invariably signifies the truth of the
proposition in which it occurs.
To see in more detail the relationships between the sense of the copula
and the various senses in which the significata of several predicates can be
said to be, let us take a more careful look at the various formulations of the
inherence theory of predication provided above. All of these formulations
are equivalences; the left-hand side states the truth of a predication, while
the right-hand side states the actual existence of the significata of the
predicate in the supposita of the subject of the predication. But a predica-
tion itself is performed by a categorical proposition, the formal element of
which is the copula, which joins the terms to form a proposition. (Whether
it is expressed in the surface structure of a language or not, the copula is
necessary if a sequence of terms is to form a proposition.
62
) This is why we
can say that what is invariably signified by the copula is the truth of the
proposition in which it occurs. This is precisely the reason St. Thomas also
often speaks about the mode of being that is signified by the copula as ens
ut verum, "being as truth. "63 But stating the truth of the proposition is
nothing but stating the obtainment of the state of affairs expressed by the
proposition as a whole. In English the state of affairs expressed by a propo-
sition can be referred to by a that clause, and in Latin it can be referred to
118 GYULA KLIMA
that is, are things for which color predicates are simply not interpreted. In
such things color predicates signify nothing.
58
) However, while concrete
terms supposit for the things that have their significata in actuality, abstract
terms supposit for their significata in actuality (relative to the conditions
required by the propositional context). So whereas 'white' supposits for
white things, 'whiteness' supposits for whitenesses, and so on. Of course,
this does not tell us what kind of entity a whiteness is or, indeed, whether it
is something distinct from a white thing.
59
(5) An affirmative categorical proposition is true iff its predicate is
true of (all or some, depending on the quantity of the proposition)
of the supposita of its subject.
The truth of an affirmative categorical proposition-a proposition in which
two common terms are joined by an affirmative copula-is ultimately deter-
mined by whether its predicate is true of the thing(s) supposited for by its
subject term. Note that the terms here may be of any complexity, so the
theory of categorical propositions need not be thought of as covering only
simple cases like 'A man is an animal'. But here we need not deal with these
complexities.
60
Moreover, the truth of a categorical is determined by its
quantity, that is, the determiner of its subject term, which determines how
many of the supposita of the subject term should be considered in deter-
mining the truth of the proposition. In fact, this analysis of the categoricals
allows for a uniform treatment of all sorts of determiners in the framework
of a logical theory the expressive capacity of which matches that of gener-
alized quantification theory. But again, such technicalities need not be
considered here.
61
The only really important point in the present discus-
sion is that in accordance with this principle, the truth of a categorical
proposition ultimately depends on whether its predicate is true of the
supposita of its subject. But whether the predicate is true of these supposita
depends on the actual inherence of its significata in these supposita. And
so, since the actual inherence of these significata is nothing but their
actuality, that is, their actual being, this point directly takes us back to the
starting point of our discussion: the various analogical senses of 'being'
distinguished by St. Thomas in accordance with the various modes of
predication.
58. For this point, see again n. 40.
59. See again nn. 47 and 48.
60. For a detailed discussion of these complexities, see G. Klima, "Latin as a
Formal Language: Outlines of a Buridanian Semantics," Cahiers de l1nstitut du
Moyen-Age Gree et Latin, 61 (1991): 78-:106.
61. Nevertheless, see Essay IV of my Ars Artium; or G. Klima and G. Sandu,
"Numerical Quantifiers in Game-Theoretical Semantics," Theoria56 (1990): 173-92.
SEMANTIC PRINCIPLES OF AQUINAS
VII. BEING SIMPLICITER AND SECUNDUM QUID
119
120 GYULA KLIMA
either by a clause beginning with a corresponding quod or quia or by an
accusative with infinitive construction.
64
Consider the following, equivalent
sentences:
(1) S isP
(2) 'S is P' is true
(3) It is true that S is P
(4) That S is P is true
(5) That S is P is
(6) That S is P is a being.
Aquinas's point seems to be precisely that all occurrences of 'is' (or rather
the corresponding occurrences of 'est' in Latin), as well as the occurrence
of 'being' in (6) above, express the same sense of 'being'. This sense,
however, is not the primary sense but somehow analogically related to it.
That is, all occurrences of 'is' above, both as a copula and as an absolute
predicate, express the same sense in which something can be said to be.
Indeed, they express the same sense as that expressed by the predicate
'being in (6), as well as in sentences such as 'A blindness is a being.65 In fact,
any of the above formulations could stand on the left-hand side of an
64. For the early medieval theory of such sentential nominalizations, called
appellatio dicti, see L. M. deRijk, ed., Logica Modernorum, II. 1. 2 (Assen, Neth.: Van
Gorcum, 1967).
65. But how is it possible, one might ask, for both the copula and the predicate
of 'A blindness is a being to express the same sense? How could any absolute
predicate have the same sense as the copula? Well, using 'is
2
' to distinguish the
usage of 'is' when it is used in the ens rationis sense, and using 'being
2
' to distinguish
the usage of 'being' when it is used in the ens rationis sense, we can say first, in
general, that 'is
2
' signifies the same whether it is used as a copula of a proposition
or as the absolute predicate of a significate of the predicate of the same proposition:
SGT('is
2
') (SGT(P (SUP(S) (t (t) = SGT('is
2
') (SGT(P) (SUP(S) (t (t (t); that
is, given that, in accordance with rule (4), SGT(P)(SUP(S) (t(t) = SUP([P]) (t),
SGT('is
2
') (SGT(P) )(SUP(S)(t (t) = SGT('is
2
') (SUP([P]) (t (t), which is pre-
cisely to say that 'is
2
' both in 'An S iS
2
P' and in 'A [P] is
2
' signifies the same, or has
the same sense. But then, further, we can say that SGT('is
2
')(SUP([P])(t(t) =
SGT('being
2
') (SUP( [P]) (t (t), which is nothing but to claim that 'is
2
' and 'being
2
'
as absolute predicates of the supposita of [P] signiry the same. In general,
SGT('is
2
') (SGT(P (SUP(S) (t (t) = SGT('being
2
') (SUP(S) (t (t). Whence, in
particular, SGT('is
2
') (SUP ('blindness') (t (t) = SGT (' is
2
')(SGT('being
2
'
(SUP('blindness') (t (t) = SGT('being
2
') (SUP('blindness') (t (t). So since 'blind-
ness' supposits for a privation, we say that blindness is and it is a being, but it is and
it is a being precisely and only in this sense, namely, in the sense that is also
expressed by the copula of an affirmative proposition. For the different case of
'sight' see n. 69. Indeed, even further, SGT('is
2
') (SUP('blindness') (t (t) =
SGT('is
2
')(SGT('blind' (SUP(S)(t (t); that is, when we say that a blindness is, the
predicate of this sentence signifies the same as what is signified by the copula of the
sentence by which we assert that an S is blind, precisely as Aquinas says (see In meta.
lb. 5, le. 9, n. 896, quoted just below).
equivalence, stating the inherence theory of predication. However, on the
right-hand side we would have to have formulations of the following sort
(where [P] is the abstract form of P):
(7) A significate of P in a suppositum of S is (is actuallexists)
(8) A [P] of an Sis (isactuallexists)
(9) A [P] of an S is a being.
Note that here it is only the italicized copulae of these schemata that express
the same sense as the occurrences of 'is' in (1)-(6). The other occurrences
of 'is', as well as occurrences of 'actual', 'exists', and 'being', express
different senses of 'being', depending on what sort of entity is signified by
P in a suppositum of S. For example, with 'blind' as the privative opposite
of 'sighted', and 'blindness' and 'sight' as their respective abstract forms,
consider the following instances of a scheme that expresses the inherence
theory of predication:
(10) A man is sighted if and only if the sight of a man is (is a being)
(11) A man is blind if and only if the blindness of a man is (is a
being).
St. Thomas's point seems to be that here the italicized occurrences of 'is'
(and 'being') signify being in the same sense:
We should know that this second mode is compared to the first as effect
to cause. For it is from there being something in the nature of things
that the truth or falsity of a proposition follows, which the intellect
signifies by the verb 'is' as it is a verbal copula. But, since something
which in itself is not a being, as a negation and the like, is considered
by the intellect as some being, sometimes that it is is said of something
in this second way, but not in the first. For it is said that blindness is, in
the second way, because the proposition by which something is said to
be blind is true; but it is not said that this is true in the first way. For
blindness does not have some being in the things, but it is rather a
privation of being.
66
However, the nonitalicized occurrence of ,is' (as well as that of 'being')
in (10) signifies being in the sense contrasted with the ens-rationis sense, as
expressing real being, ens reale, which itself comprises both the primary
sense and several other secondary senses:
We have to say that 'being' is predicated in two ways.... In one way as it
is a verbal copula signifYing the composition of any proposition that the
soul forms: whence this being is not something in the nature of things,
66. In meta. lb. 5, le. 9, n. 896.
122 GYULAKLIMA SEMANTIC PRINCIPLES OF AQUINAS 123
but only in the act of the judgment-forming intellect. And in this way
being is attributed to everything ofwhich a proposition can be formed,
whether it is a being or a privation of being: for we say that blindness is.
In the other way being [esse] is said to be the act of being [ens] insofar as
it is a being, that is, that bywhich something is denominated as a being in
the nature of things. And being in this way is attributed only to the things
themselves which are contained in the ten categories, whence 'being'
[ens] predicated on account of such [an act of] being [esse] is divided by
the ten categories. But this [act of] being [esse] is attributed to something
in two ways. In one way as to that which [quodJ properly and truly has
being, or is. And thus it is attributed only to a per se subsisting substance;
whence, in bk.1. ofthe Physics, that which truly is is said to be a substance.
All those [things], however, which do not subsist per se, but in others and
with others, whether they are accidents or substantial forms or any sorts
of parts, do not have being [esse] so that they themselves would truly be,
but being [esse] is attributed to them in another way, namely, as to
something lJy which [quo] something is; as whiteness is said to be, not that
it itself would subsist in itself, but because it is by it that something has it
that it is white. Being, therefore, is properly and truly attributed only to
a per se subsisting thing. To this, however, two kinds of being are attrib-
uted. The one is what results from those from which its unity is inte-
grated, which is the proper substantial being of a suppositum. Another
being is attributed to a suppositum besides those that integrate it, which
is an additional being, namely, accidental being; as being white is attrib-
uted to Socrates when it is said: Socrates is white.
67
So, if in
(i) An Sis P
P signifies a privation (or negation or relation of reason
68
), then in the
corresponding
(ii) A [P] of an S is
'is' signifies being in the secondary, ens-rationis sense, inasmuch as the two
propositions are equivalent. Ofcourse, nothing prevents anyone from assert-
ing, say, that [a] blindness is, or [an] evil exists, intending to signify that
blindness and evil are real beings, that is, taking 'is' or 'exists' in the sense in
67. QJJL 9, q. 2, a. 2, in corp.
68. Although here I take only privations as my examples of beings of reason, it
is not only privations that were regarded as such by St. Thomas. Indeed, according
to him, the proper subject matter of logic consists of beings of reason of a different
sort: relations of reason. For a thoroughgoing discussion of St. Thomas's concep-
tions of the subject of logic, see R. W. Schmidt, The Domain ofLogic according to Saint
Thomas Aquinas (The Hague: Martinus Nijhoff, 1966). For a discussion of various
sorts of entia rationis and their systematic role in medieval semantics, see G. Klima,
"The Changing Role of Entia Rationis in Medieval Philosophy: A Comparative Study
with a Reconstruction," Synthese96:1 (1993): 25-59.
which they signify the actuality of real being. But if the terms 'blindness' and
'evil' are genuinely privative, in the way they were interpreted by St. Thomas,
these assertions simply will not be true. However, if P signifies a real being,
then, while the copula of (i) will still express being in the ens-rationis sense, 'is'
in (ii) will signify some real being. For example, if the proposition 'A man is
sighted' is true, then 'A man's sight is' will also be true, so that while the
occurrence of 'is' as the copula of the first proposition signifies being in the
ensrationis sense, the occurrence of ,is' as the predicate of the second proposi-
tion signifies being in an ens reale sense.
69
Again, of course, one may wish to
claim that a man's sight is, or exists, or that there is such a thing as a man's
sight, in the ens rationis sense, in which sense the claim thus made will also be
true since whatever is a real being also exists in this weaker, ens rationis sense
according to St. Thomas. But the point is that, unlike the case of 'A man is
blind' and 'A man's blindness is', if 'A man is sighted' is true, then 'A man's
sight is' will also be true even when 'is' is taken in the latter case in an ens reale
sense. However, as we have seen, given that a man's sight is not a substance,
even this is not the primary sense of 'is' but only one of the several senses
analogically related to its primary sense, namely, the sense in which we can
say of a substance that it is.
Perhaps, in viewofwhat has been said so far, we can nowsee howAquinas
is able to base this distinction between the primary and the secondary senses
in which a real being can be said to be on the distinction between that whichis,
quod est, and that by which something is, quo aliquid est, which is based in turn
on a distinction between what he regards as the predication of 'being' or 'is'
simpliciter and secundumquid. The clue, ofcourse, lies in the inherence theory
ofpredication and in the related theories ofsignification and supposition. In
accordance with these, when a substance u is said to be, say, white, this is
equivalent to saying that the whiteness ofu exists. We can say, then, that for
u's whiteness to be is nothing other than for u to be white, which is further for u
69. Of course, the absolute predicate 'is', expressing an ens-reale sense, could
be analyzed further into 'is a being'. In this case 'is' would express the ens-rationis
sense and 'being' an ens-reale sense. Using 'iS
1
/
2
' and 'being
1
/
2
usage of 'is' and 'being', respectively, in which they express the sense in which they
are truly predicated of really inherent accidental forms but not of mere beings of
reason, we can say that SGT('isl/
2
') (SUP('sight') (t(t) = SGT('being
1
/
2
')
(SUP('sight')(t (t), but SGT('being
1
/
2
') (SUP('sight')(t (t) =F SGT('is
2
')
(SGT('being
1
/
2
') )(SUP('sight')(t) )(t), whereas, of course, SGT('is
2
') (SGT('be-
ing
1
/
2
' (SUP('sight') (t (t) = SGT('is
2
') (SUP('sight') (t (t). So, what 'is' signifies
in a secondary ens reale sense in a suppositum of 'sight' at the time of uttering 'A
sight is' is the same as what 'being' signifies in the same sense in respect of the same
at the same time in the utterance 'A sight is a being', but this is not identical with
what the copula of this latter utterance signifies at the same time in respect of the
signification of the predicate ('being') and the suppositum of 'sight' at the same
time, which is nevertheless nothing, but what would be signified by 'is
2
' in the
sentence 'A sight is
2
', claiming the existence of sight in the ens rationis sense. Cf.
with this the analysis of 'A blindness is a being in n. 65 above.
124 GYULA KLIMA
to benot absolutely (simpliciter) but with respect to something (secundum
quid), namely, with respect to its whiteness. As St. Thomas says,
Substance is the first being, and is being absolutely [simpliciter], and not
being with respect to something, i.e., with qualification [secundum
quid], as is the case with accidents. For to be white is not to be absolutely
[simpliciter esse], but with qualification [secundum quid]. And this is clear
from the fact that when something begins to be white, we don't say that
it begins to be, absolutely, but that it begins to be white. For when
Socrates begins to be a man, we say absolutely that he begins to be.
Whence it is clear that to be a man signifies to be, absolutely. But to be
white signifies to be with qualification.
70
Thus we can see exactly how Aquinas can interpret all predications as
predications of being, either with or without qualification. Just as in "ordi-
nary predications" we can attach various qualifications to the predicate, so
these "ordinary predications" themselves may be regarded as various quali-
fications of the predication of being. According to this analysis, therefore,
when we say, 'Aman is blind', this is equivalent to saying, 'Aman's blindness
is' which in turn, is equivalent to saying, 'A man is with respect to his
blindness' .
It is this last formulation which shows explicitly why we can say that the
attribution of blindness to a man is an attribution of being to him (insofar as
we say that he is with respect to something), although not simpliciter but
secundum quid (insofar as we say that he is with respect to something). Moreover,
it shows that this attribution is at the same time an attribution of being to his
blindness (insofar as what is signified by the complex predicate 'is with
respect to blindness' in him is identical with what is signified by 'is' in his
blindness, and which in turn is identical with what is signified in him by 'is
blind').71 However, given the privative character of the predicate 'blind', the
being thus attributed is, and can only be, being in the ens rationis sense, which
is nothing but the lack of real being being conceived by the intellect as a form
signified by the predicate 'blind'. So we can see how we should understand
that the being of a privation (or of any other ens rationis, for that matter) is
being in the intellect, having some foundation in reality. Being in the intel-
70. In meta. lb. 7, le. 1, n. 1256. Cf. De Principiis Naturae c. 1.
71. Using 'is
2
' again in the same way as above, and using 'is
1
usage of 'is' in which it is truly predicable in its primary sense of substances, we can
state this formally as follows: SGT('is
2
blind') (SUP('man') (t = SGT('is
2
')
(SGT('blind' (SUP('man') (t = SGT('is
1
with-respect-to blindness')
(SUP('man') (t (t) = SGT('is
2
') (SUP('blindness') ) (t (t). Here, of course, the sig-
nification of the complex, qualified predicate can (and, to comply with the require-
ment of compositionality, should) be analyzed further as follows: SGT ('is
with-respect-to blindness') = SGT('with-respect-to') (SGT('is
1
'
ness') (t. (The phrase 'with-respect-to' need not be analyzed because it functions as
a single modifier. Indeed, in Latin the one-word phrase respectu, as in homo est respectu
caecitatis, or secundum, as in homo est secundum caecitatem, would do the samejob.)
lect is nothing other than being conceived by the human intellect, which we
can express by predicates that signify acts of human awareness, as when we
say that a chimera is thought of. In fact, even this predication can be con-
strued as a predication of being secundum quid (which may be expressed by
saying that a chimera iswith respect to its being thought of), but the sense of
being that is predicated in this case is not even the ens rationis sense, which
Aquinas says is the sense in which we can answer a question that asks whether
there is such and such a thing. For, ofcourse, to the question ofwhether there
are chimeras, it would not be an appropriate answer to say that yes, there are,
insofar as they are thought of However, to the question of whether there is
blindness, it is an appropriate answer to say that yes, there is, insofar as some
animals that should have sight by nature actually lack sight, which is precisely
what is conceived of in the concept of blindness.
Note here that for the being of blindness, both conditions are required:
both the lack ofsome sight and the activity of the human intellect to conceive
of this lack of sight by forming the concept of blindness. Were there no
humans to form the concept ofblindness, there would be no blindness, even
if there were animals that lacked sight, for the actuality of blindness consists
in the actual lack of sight conceived by humans by applying the concept of
negation to the concept of sight (in forming the concept of blindness as an
animal's lacking, i.e., not having sight).72 In contrast, the actuality of sight
does not involve any such activity of the human mind, for if there were
sighted animals, there would be sights in actual reality, even if there were no
72. At this point one might object that if there are animals that lack sight, then
there must be blindnesses also, even if there are no minds to conceive of them, given
that for a blindness to be is nothing but for a sight not to be. (In fact, I received this
objection from Scott MacDonald.) Well, it is true that for a blindness to be is for a
sight not to be, in the sense privations and negations can be; that is, SGT('is
2
')
(SUP('blindness') (t (t) = SGT('not') (SGT('is
2
' (SUP('sight') (t (t). Here, of
course, SGT('not') (SGT(P (SUP(S) (t (t) EA(t), if SGT(P) (SUP(S) (t (t) EW-
A(t), SGT('not') (SGT(P (SUP(S) (t (t)EW-A(t), if SGT(P) (SUP(S) (t (t)
EA(t), and SGT('not') (SGT(P (SUP(S) (t (t) = 0 ifSGT(P) (SUP(S) (t (t) = 0
(Note how these rules provide the conditions for an internal negation, the truth of
which presupposes that the negated predicate is interpreted for the subject of which
it is denied. This is how we can do justice to the intuition that, say, the number 2 is
neither sighted nor blind.) Still, even if there are no blindnesses, that is,
SGT('is
2
') (SUP('blindness') (t (t) it is quite possible thatthere are no sights
either: SGT('is
1
/
2
') (SUP('sight') (t) )(t) So even if there are no sights in the
sense sights could be because, say, all animals lack sight, still it is quite possible that
there are no blindnesses either in the sense blindnesses could be (that is, in the
sense that they are conceived and have the foundation in reality required by their
concept) because there are no minds to form the concept of negation and hence
to be able to conceive of them. In fact, we can say that in the case when there are
no minds to form the concept of negation SGT('is
2
') (SUP('blindness') (t (t) =
SGT('not') (SGT('is
2
' (SUP('sight') (t (t) = O. But of course with this it is com-
patible that SGT('is
1
/
2
')(SUP('sight')(t and even that SGT('is
2
')
(SUP('sight') (t (t) In brief, whenever there is no sight, there is no being of
a sight either, but this does not mean that then there also has to be the being of the
nonbeing of sight conceived by an intellect.
126 GYULAKUMA
humans to conceive of them. The difference is thatwhat we conceive of by
the concept ofblindness involves, in its very concept, some mental act (in this
case negation), whereby it can be said to be only if the relevant mental act
exists. However, even if the concept ofsight is a mental act in itself, it does not
involve any mental act in what is conceived ofby it, whence what is conceived
of by it can exist whether there are any mental acts in reality or not. Thus, the
difference between a real being and a (mere) being of reason is the mind-de-
pendence of the latter. However, the difference between a being of reason
and a mere object of thought (and hence an object of signification and
reference) is precisely that the former has some foundation in reality,
namely, the way real beings are, as conceived ofin the concept of the being of
reason in question. So we can say that for a mere object of thought to be is
nothing but for it to be thought of. But for a being of reason to be, in the
sense in which, say, a privation is, is for it to be thought of and for something
that is conceived ofin its concept also to obtain in the realm of real beings. In
the case of a privation, for it to be is for it to be thought of and for its real
opposite not to be.
73
Thus, Aquinas's "universe of discourse" consists primarily of objects of
thought, that is, whatever human beings can think of and hence signify (by
external, spoken or written words) or refer to in the context of a proposi-
tion. Therefore, since a chimera can be thought of,74 and hence since the
proposition 'A chimera is thought of, in which reference is made to a
chimera, can be true, a chimera can be said to be one of the things that can
be thought of, signified, or referred to. Still, of course, it is not true to say
that what is thought of when a chimera is thought of, or what is referred to
in the proposition 'A chimera is thought of, is a chimera. For nothing is a
chimera. Thus a chimera cannot be said to be at all, except perhaps in the
very thin sense of possibly being thought of, in the sense of 'being' Walter
Burleigh called 'ens maxime transcendens'.75
73. Thus, the actualityofa being ofreason is conditioned both on the side ofreal-
ity and on the side ofthe activity ofthe intellect. So ifeither ofthese two fails to obtain,
a being ofreason will not be in actuality, just as it was technically detailed in n. 72.
74. Of course, insofar as we construe 'chimera' as referring to a possible
imaginary animal, and not an implicitly contradictory term, in the way fourteenth-
century logicians used this term. For more on this issue see E. J. Ashworth, "Exis-
tential Assumptions in Late Medieval Logic," American Philosophical Qy,arterly 10
(1973): 141-47. Cf. Klima, "Old Directions in Free Logic."
75. "... ens potest accipi tripliciter. Uno modo ut est maxime transcendens et
commune omni intelligibili. Et sic est adaequatum obiectumintellectus. Et sic non se-
quitur: Hoc est ens, ergo hoc est. Secundo modo accipitur pro ente, cui non est esse
prohibitum, et sic omne possibile est ens. Et sic etiamnon sequitur: Hoc est ens, ergo
hoc est. Tertio modo accipitur pro ente actualiter existente, et sic est participium de-
scendens ab hoc verbo 'est'. Ens primo modo dictum dicitur ens in intellectu, quia est
obiectum intellectus; et ita est in intellectu obiective. Ens secundo modo dictum dici-
tur ens in suis causis vel ens quod est in sua causa. Sed ens tertio modo dictum dicitur
esse ens in se." W. Burleigh, De Puritate Artis Logicae Tractatus Longior, ed. P. Boehner
(St. Bonaventure, N.Y: Franciscan Institute, 1955), pp. 58-59.
But this is not the sense of 'being' in which St. Thomas calls something
a being of reason, the sense in which to say of something that it is answers
the question of whether there is such a thing. Thus the domain of beings
of reason forms a subdomain ofAquinas's universe of discourse, comprising
both real beings and mere beings of reason but excluding mere objects of
thought. A being of reason can properly be said to be in the sense in which
a being of reason can be at all, that is, in the sense that is signified by the
copula of an affirmative proposition. But we have seen that the being that
is signified by a copula is not signified only by a copula, for being is signified
in the same sense by 'is' as the predicate, say, in the sentence 'A blindness
is' (or by 'being' in 'A blindness is a being'), as this sentence is true.76 This
sense of 'is' or 'being' is analogically related to the sense in which really
existing substances are said to be. But then, just as any analogical sense of
a common term is derived from its primary sense by some determination
or qualification of its primary sense, so this (as well as any other, analogical
sense of 'being') is derived from its primary sense by some qualification of
it. In general, the predication of being (by means of predicating 'is' or
'being' or 'is P') in any of its analogical, secondary senses is the predication
of being in its primary sense with some determination or qualification. And
since in view of the inherence theory of predication every predication is a
predication of being, either in its primary or in one of its analogical senses,
every predication is a predication of being either without or with some
qualification, either simpliciter or secundum quid. 77
To see this relationship between predication and the predication of the
several analogous senses of being in general, consider the following
scheme:
(EP) (i) An S iS
2
P iff (ii) An S's [P] iSx iff (iii) An S is} with respect
to [P] 78
In this scheme the subscripts to 'is' are designed to distinguish the various
senses in which being is signified in the given propositional context, insofar
as the proposition is or can be true (for, of course, anyone may intend to use
'is' in a different sense in a given context, but then, in that use, his or her
76. See nn. 65 and 69 above.
77. Cf. n. 71. Note that when I am talking about 'the predication of being' in
a certain 'sense of being', I mean the act of predicating either the absolute predi-
cate 'is' or the predicate 'being' or 'is a being' or the complex 'is P' or 'is with
respect to P' in the sense of 'is' or 'being' required by the context for the truth of
the predication.
78. In such a "chain" of equivalences, the logically squeamish would certainly
look for parentheses (as in 'p iff (q iff p)'), but in vain. Such "chains" of equiva-
lences are intended only to abbreviate a series of proper equivalences. Thus, '(i) iff
(ii) iff (iii)' abbreviates '(i) iff (ii) and (ii) iff (iii) '; that is, this is just a brief
expression of the idea that each member of the chain is equivalent to any other
member. Of course the same applies to such "chains" of any number of members.
128 GYULA KLIMA
proposition may be necessarily false or even ill formed). Accordingly, in (i)
'is
2
' signifies being in the ens rationis sense, the sense uniformly expressed
by the copula, as well as by 'is' or 'exists' as absolute predicates, insofar as
they are true of beings of reason; in (ii) 'is
x
' signifies being in any of the
analogical senses of being (including the ens rationis sense); and, finally, in
(iii) 'is
l
' signifies being in the primary sense, the sense in which it can truly
be stated of substances as an absolute predicate, to signify their actuality
absolutely speaking, as well as a predicate with some qualification, to signify
their actuality in some determinate respect.
Let us see, then, exactly how the various kinds of substitutions of P
affect what sense of being is expressed by 'is
x
' in (ii) insofar as it is true. If
we substitute for P a term that signifies an ens rationis (which is not a real
being), such as the term 'blind', then the equivalences of (EP) will hold if
we take 'iSx' to be 'is
2
', that is, if we take 'iSx' in (ii) to express the ens rationis
sense of being as it was contradistinguished by St. Thomas from the ens reale
sense. Thus, substituting 'blind' for P and 'animal' for S, we get the follow-
ing valid instance of (EP):
(E
2
P) (i) An animal iS
2
blind iff (ii) The blindness of an animal iS
2
iff (iii) An animal iS
l
with respect to blindness
Note that in (ii) 'is
2
' could be analyzed further as 'is
2
a being2" which in
turn is equivalent to 'is
2
a beingl of reason'. 79 This is how the secondary, ens
rationis sense of being can be shown to be derivable from the primary sense
by the addition of the appropriate (in this case, diminishing) qualification.
Of course, this is why the inference from 'A blindness of an animal is
2
', that
is, 'A blindness of an animal iS
2
a being
l
of reason', to 'A blindness of an
animal is
l
', that is, 'A blindness of an animal iS
2
a beingl" would be invalid.
Thus, Aquinas can justifiably claim that a (mere) being of reason is not a
being, except secundum quid.
But then one might ask why the inference secundum quid ad simpliciter
in the case of (iii) would be valid. For we can certainly infer from (iii) thaJ
an animal is, indeed, absolutely and in the primary sense of being. o r d ~
ing to the equivalences given above, (iii) is equivalent to (i), and (i)
certainly implies that an animal is, for no animal can be blind unless it
exists. But if the above equivalences also show how this secondary sense of
being is derivable from the primary sense by adding the appropriate,
diminishing qualification, then the act of being signified by the predicate
of (ii) is the same as the act of being signified by the qualified predicate of
(iii). Indeed, it is precisely this qualification that modifies the sense of the
predicate in such a way that it signifies an act of being in this secondary
sense (in this case, the actuality of the blindness of an animal) instead of
79. In fact, SGT('being} of reason') = SGT('is
2
') = SGT('being
2
'). Cf. nn. 65
and 69. .
what it would signify without this qualification (an act of being in the
primary sense). However, in the case of (ii) we could not conclude from the
actuality of the act of being signified by the predicate of (ii) to the actuality
of the act of being signified by 'is
l
' absolutely. So why can we do so in (iii)?
The answer is straightforward if we consider these inferences them-
selves.
so
In the case of (ii) the conclusion stated in itself would state the
existence of a blindness of an animal in the primary sense of being, in which
sense this claim could not be true. However, in the case of (iii) the conclu-
sion in itself states the actual being of an animal in the primary sense, which
of course is implied by the existence (in the secondary sense) of its blind-
ness. So although it is indeed the same act of being that is signified both by
the predicate of (ii) and by the qualified predicate of (iii) (a very important
point to which I shall return), the subjects of the conclusions of these
inferences secundum quid ad simpliciter are not the same. Therefore, the two
inferences have nothing to do with each other.
Still, one might insist that the validity of the inference secundum quid ad
simpliciterin the case of (iii) seems to invalidate the general rule concerning
such inferences, namely, that if the qualification added to the predicate is
diminishing, then the inference is invalid, whereas the equivalences above
show precisely that the qualification in (iii) is diminishing.
This objection rests on a misunderstanding of what a formal rule of
inference is. Even if an inference is not valid in its form, nothing prevents
it from being valid on the basis of the actual meaning of its terms. For
example, even if an inference of the form 'an S is M, and an M is P;
therefore an S is P' is formally invalid (as can be shown by substituting
'man', 'animal,' and 'donkey' for S, M, and P, respectively), an inference of
that form can be valid because of the meaning of the particular terms (for
example, if we substitute 'man', 'animal,' and 'living being' for S, M, and P,
respectively). But then, in the same way, although an inference of the form
'An S iS
2
P with respect to [Q]; therefore an S iS
2
P' is formally invalid if 'with
respect to [Q]' is a diminishing qualification of P-but formally valid if
'with respect to [Q]' is a nondiminishing qualification of P-nothing pre-
vents an inference of this form from being valid because of the meaning of
its terms even if it is formally invalid, that is, even if the qualification of the
predicate is diminishing. Indeed, the particular reason why in the case of
(iii) such an inference (that is, the one resulting from substituting 'is
l
' or
'is
2
a being
l
' for 'is
2
p', as well as, 'blindness' for [Q], namely, 'An animal
iS
l
with respect to blindness; therefore an animal is
l
') would be valid,
80. Namely, 'The blindness of an animal is
2
; therefore, a blindness is}', and 'An
animal is} with respect to blindness; therefore an animal is}'. The question is why
the first inference is invalid while the second is valid, given that SGT('is
2
')
(SGT('blind') (SUP('animal') (t (t (t) = SGT('is} with-respect-to blindness')
(SUP('animal') (t) )(t), where, of course, SGT('is} with-respect-to blindness') =
SGT('with-respect-to') (SGT('is}' (SGT('blind' = SGT('is
2
') (SGT('blind'. Cf.
n.71.
130 GYULA KLIMA
despite the fact that the qualification is diminishing, is that a privation is a
lack of a property in a determinate subject, so if the privation is actual, then
the subject has to be actual too.
81
That the inference is still formally invalid
if the qualification is diminishing is shown by the fact that by substituting
'is1' or 'is
2
a being
1
' for 'is
2
P' and, say, 'being thought of for [Q] in this
rule of inference, the resulting inference, 'An animal iS
1
with respect to
being thought of; therefore an animal is
1
', is invalid.
Thus, if in (EP) we substitute for P a predicate that signifies a being of
reason, the qualification added to 'is
1
' in (iii) modifies its sense in such a
way that the resulting complex, qualified predicate will signify the same act
of being as that signified by 'is
2
' as an absolute predicate in (ii) or by the
equivalent phrases 'is a being in the secondary, ens rationis sense', 'is a being
of reason', 'is a being in thought having some foundation in reality', and so
on, all signifying being in the same, diminished, qualified sense. So the
qualification in this case yields a sense of being in which it is predicable even
of objects of thought that exist neither in the primary sense of subsisting,
nor in the secondary sense of being some really inherent property of a
subsistent being, but only in the sense of being conceived by the human
mind and having for their actuality some foundation in reality, some par-
ticular way real beings actually are, as was explained above.
On this basis we can perhaps better understand what it is for a quali-
fication to be diminishing and how the addition of diminishing qualifi-
cations is related to analogical predication in general. The first interesting
thing to note about a diminishing qualification is that, to put it in con-
temporary terms, while it is intensionally diminishing, it is extensionally "en-
larging." The point is that a diminishing qualification is one that takes
away some of the conditions of the strict applicability of the term to which
the qualification is added. Hence the term with the qualification added
becomes applicable even to things to which in its strict usage, without
the diminishing qualification, it could not apply. For example, if to the
predicate 'white' we add the qualification 'with respect to its one-half,
then the qualified predicate will apply even to things which strictly speak-
ing and without this qualification could not be called white, since properly
and strictly speaking only things that are white with respect to their whole
surface can be said to be white without qualification. Indeed, this also
shows immediately that the qualification 'with respect to its whole surface'
is a nondiminishing qualification of 'white',82 whence in a predication it
81. Cf. n. 72.
82. That is, SGT('white} with-respect-to a-part-of-surface) = SGT('white ') =I:-
SGT('white} ') = SGT('white} with-respect-to whole-surface'). Here 'white},2
indi
-
cates the strict usage of 'white', in which it is true only of those things that are totally
w h ~ t w h r ~ 'white
2
' indicates a broader, less stringent usage of 'white', say, in
whIch the whIte pages of the phone book can be called 'white' (as opposed to the
yellow pages), despite the black letters that cover a part of their surface. Of course,
in a complete semantics the hyphenated qualifications could (and should) be
analyzed further, but that is irrelevant here.
can be added or dropped without affecting the truth conditions of the
predication. It follows from this that an inference secundum quid ad sim-
pliciter that drops this qualification from 'white' will be valid. (It should
be noted, however, that a qualification that is nondiminishing of one
predicate may be diminishing of another; for example, 'with respect to
its whole surface', is a diminishing qualification of 'hot', for something
can be called strictly and properly hot only if it is hot in its every part,
not only on its surface.) However, in ordinary usage we do not always
bother adding even diminishing qualifications. Instead, we apply the same
term without qualification, intending it to have some qualification which
should be clear from the actual context.
83
If such a usage is quite clearly
related to the primary usage and is generally received, we get exactly an
analogical, extended usage of the same term-that is, a usage of the term
in which it applies to things to which it would not apply in this strict
primary usage-but because of the implied qualification that relates it to
its primary usage, it is still properly applicable in this secondary usage
even without explicitly adding the relevant qualification. This, then, is
the reason St. Thomas said that a division of a common term with respect
to its analogata-that is, distinguishing a term's usage in its primary sense
from its usage in its secondary, analogical senses-yields a distinction be-
tween the sense in which the term can be predicated simpliciter, and several
other senses in which it can be predicated only secundum quid.
84
Also,
this is how, according to St. Thomas, 'being' is predicated simpliciter, in
its primary sense, of substance (that is, of that which is simpliciter, a sub-
83. Indeed, most things we normally call white, even if they are white allover
their whole surface, could not be called absolutely and totally white in the strictest
sense; namely, in the sense in which the definition of whiteness would be satisfied
absolutely and 100 percent by the property of the thing called white. On the basis
of contemporary physical optics we could reasonably define that property, white-
ness, as the reflective capacity of a surface measurable by what is called albedo
(Latin for whiteness) in modern science, that is, the percentage of the reflected
light relative to the incident light (which is, by the way, just one of the nicest
illustrations, pace Moliere, of modern science's unabashed usage of "scholastic
barbarisms" when they come in handy). But then we can also quite reasonably say
that only that thing can be called absolutely white that has 100 percent albedo. So
any object with a lesser albedo is only white to a certain degree but not strictly and
absolutely white, without any qualification, even if we normally would not add any
qualification in ordinary usage. This example also shows very nicely that it is
precisely this phenomenon, namely, the omission of several sorts of implied quali-
fications, that accounts for much of the vagueness or fuzziness of everyday usage,
represented in contemporary "fuzzy logics" by assigning fuzzy or diminished truth
values to predications, that is, truth values between (and including) 0 and 1,
instead of the classical 0 and 1. However, in treating this fuzziness, instead of
assigning fuzzy, qualified truth to unqualified, absolute predicates, medieval Aris-
totelian logicians assigned absolute, unqualified truth to either explicitly or im-
plicitly qualified, fuzzy predicates. But vagueness is an issue beyond the scope of
this paper.
84. See n. 14.
132 GYULA KLIMA
sistent being),85 and secundum quid, in one of its secondary senses, of
substantial forms (that is, of beings by which a substance is), accidents
(that is, of beings by which a substance is somehow), beings of reason, beings
in potentiality, and perhaps even mere objects of thought.
86
It should also be clear, then, why St. Thomas would claim that there
is a distinction between the attribution of being to something as to that
which is and the attribution of being to something as to that by which
something else (namely, a substance) is, even if in both cases we are
talking about real beings and not about mere beings of reason.
87
As was
pointed out above, the act of being signified by 'is
x
' in (ii) in (EP) is
the very same act of being as that signified by the qualified predication
of being in (iii). But if in (EP) for P we substitute an accidental predicate,
'is
x
' will be true of the supposita of [P] only in a qualified sense, namely,
in the same sense which is expressed by the qualified predication of being
in (iii), but not in the primary sense, signifying the subsistence of a sub-
stance. The latter would be expressed by an unqualified predication of
being in (iii) .88
But if it is only the absolute predication of 'is' of a substance that can
express the primary sense of being, it might appear that 'is
x
' in (ii), insofar
as it is true, can never be 'is}'; that is, it cannot express the primary sense
of being for any substitution of P, whether P is an accidental or a substantial
predicate. For the subject term of (ii) supposits for the significata of P in
the supposita of S by means of the abstract counterpart of P, namely [P].
However, according to St. Thomas, those that are signified in abstracto are
signified as something by which something else is, not as that which is.89
However, it is only that which is by itself, namely, what subsists (a per se
existing substance) that is said to be in the primary sense. So even if the act
of being signified by 'iSx' in the significatum of a substantial predicate of a
substance is the same act of being that is signified by 'is}' in the same
substance, 'is
x
' cannot be the same as 'is}'. Indeed, St. Thomas includes
even substantial forms, that is, the significata of substantial predicates,
among the things that can be said to be only in the qualified sense, as that
by which something is. But since every predicate that signifies some real
being is either substantial or accidental, it seems that nothing that is signi-
fied in abstracto can be said to be in the primary sense. Hence, for no
substitution of P would 'is
x
' in (ii) be 'is}', that is, 'is' signifying the primary
sense of being, insofar as (ii) is true.
85. "Non enim ens dicitur proprie et per se, nisi de substantia, cuius est
subsistere. Accidentia enim non dicuntur entia quasi ipsa sint, sed inquantum eis
subest aliquid, ut postea dicetur" (In De hebd le. 2).
86. Cf. In meta. lb. 4, le.l, n. 15.
87. See QDL 9, q. 2, a. 2, in corp., quoted above.
88. Cf. In meta. lb. 7, le. 1, n. 1256, quoted above.
89. ISNd. 33, q. 1, a. 2, in corp. Cf. In meta. Ib.7, le. 1, nn. 1252-1256.
But St. Thomas would say that this reasoning confuses modi significandi
with modi essendi, that is, confuses semantic with ontological distinctions.
9o
Although it is true that abstract terms are imposed to refer to what concrete
terms signify, and in the case of material beings what are thus signified are
inherent forms of material substances-for which to be is only for them to
inform the matter of these substances (that is, to render these material
substances actual in some respect)-this does not exclude the possibility of
there being forms for which to be is not for them to make something else
actual in some respect, but just to be, absolutely, in actuality. In the case of
such a subsistent form, therefore, there is no real distinction between what
is supposited for by the concrete and by the abstract term. Thus, 'is' or
'exists' or 'being' can equally and truly be predicated of both in the same,
primary sense of being, that is, signifying the subsistence of a substance.
9
}
Thus, whereas the semantic distinction (the difference between the mode of
signification of an abstract term and that ofa concrete term) is in force even
concerning such an immaterial form,92 the lack of a corresponding ontologi-
cal, or real, distinction is expressed precisely both by the fact that the
abstract and concrete terms that signify and supposit for such a subsistent
form are truly predicable of each other and by the fact that existence can
be attributed to the thing thus signified and supposited for in the primary
sense, whether the thing is referred to in concreto or in abstracto.
93
Indeed, this also shows why, adapting D. P. Henry's happy phrase,
the "aloofness" of Aquinas's semantic theory from ontology is so impor-
tant, not only from the point of view of its acceptability by others of a
different metaphysical persuasion, but also from the point of view of his
own metaphysics.
94
It is precisely this "aloofness" of the semantic theory,
not prescribing anything concerning the ontological categories of the se-
mantic values of our phrases, that makes it possible for Aquinas to admit
subsistent forms into his ontology, that is, forms which exist not only as
90. For St. Thomas's principled insistence on keeping the two apart, see both
his numerous discussions of what he took to be Plato's fundamental error concern-
ing universals and of how the modi significandi of the names we attribute to God do
not correspond to some matching modi essendi in Him.
91. 3SNds. 11, q. 1, a. 4; S71 q. 13, a. 1, ad 2-um. Cf. S71 q. 32, a. 2.
92. QDPq. 8, a. 3. Cf. ISNd. 22, q. 1, a. 2.
93. Cf. Sec lb. 1, c. 30; ISNd. 22, q. 1, a. 2; De ente c. 5. See also n. 41.
94. For D. P. Henry's own Principle of Logical Aloofness, see That Most Subtle
Qy,estion-(Qy,aestio Subtilissima): The Metaphysical Bearin[f of l\!edieval and Contempo-
rary Linguistic Disciplines (Manchester: Manchester UmversIty Press, 1984): (See
entry "aloofness" in the index.) Again, we note here the of
whether there are such subsistent forms or not IS a metaphysIcal quesuon, not
determined by these semantic considerations. All these semantic considerations
determine is what it means to claim that there are such subsistent forms. But, of
course, it is only on the basis of the proper understanding of this claim that one can
set about determining its truth.
something by which something else (namely, a substance) is but also as
that which is.
Thus, on the other hand, if in the scheme above we substitute 'God'
for P (and also, for the sake of simplicity, for S) and if we take 'divinity' (or
'godhead', if you prefer a Saxon word in the abstract form) as the abstract
counterpart of 'God', then we get the following:
(ElP) (i) God iS
2
God iff (ii) God's divinity iS
l
iff (iii) God iS
l
with
respect to divinity
On the other hand, if we substitute, say, 'tree' for P (and, again, for S, too)
and 'is[l]' for 'is
x
' to distinguish the sense in which 'is' can be truly predi-
cated of nonsubsistent substantial forms, and if we baptize the significata of
'tree' in individual trees 'arboreity' (or 'treeness', if you prefer), then we
get
(E[l]P) (i) A tree iS
2
a tree iff (ii) A tree's arboreity is[l] iff (iii) A
tree iS
l
with respect to arboreity
The point of the comparison is that even if we suppose that 'tree' is a
substantial predicate of trees as 'God' is of God, if the form signified by
'tree' in trees (that is, the substantial form of trees) is not a subsistent
form, then (ii) can be true only in a diminished, qualified sense, even if
for a tree to be, absolutely and in the primary sense, is nothing but to
have this substantial form in actuality (that is, to be in actuality with respect
to that form). Nevertheless, it is only the substance that has this actuality
as that which is (quod est), while the form, if it is not a substance itself,
exists only insofar as that by which the substance of which it is the form
has being (quo aliquid est), even if the form's act of being (provided the
form is a substantial form) is the very same act of being that the substance
itself has.
95
However, should it turn out that arboreity, the substantial form
of trees, is not only an inherent form but also a substance that for some
reason should be regarded as having subsistent being, this would mean
that, instead of (ii) in (E[l]P), 'A tree's arboreity is
l
' would be true. In
fact, showing that the rational soul is precisely such a substantial form
because of the fact that it has an operation that is not the actuality of the
body is the most crucial step in Aquinas's proof of the immortality of the
95. Formally, the point is that if P is a substantial predicate of a substance u,
then SGT('is}')(u)(t) = SGT('is[}] ')(SGT(P) (u)(t) )(t); still, of course, SGT('is}') :I:
SGT('is[l]'), for ifP signifies a nonsubsistent substantial form, then, ifu is actually
P(i.e., SGT(P)(u)(t)EA(t, then SGT('is[l]')(SGT(P)(u)(t (t)EA(t); neverthe-
less, SGT('is}')(SGT(P) (u) (t) )(t) = 0, and also SGT('is[}] ')(u)(t) = O. For the
general definition of what it is for a predicate P to be substantial to an individual
substance u, see n. 98 below.
human sou1.
96
But again, from our present, semantic point of view, the
important point is that whether this claim is true or false is not determined
by the semantic theory itself, whereas it is these semantic considerations
that tell us what such a claim means and so what evidence can be regarded
as counting for or against it.
In any case, these comparisons between the various senses in which
being may be attributed to different kinds of things should also show that
it is precisely these different senses of being, once carefully distinguished,
that can be used to express the basic ontological differences between
various kinds of beings. Accidents are the kind of beings for which to be is
to make substances actual in some respect, that is, for which to be is for their
subject to be actual in some respect-actual by an act of being that does not
make them actually be, simpliciter. So the act of being truly attributable to an
accident is just an act of being secundum quid of the substance whose
accident it is. Thus, the sense in which accidents can be said to be is only a
sense of being secundum quid, analogically related to the primary sense, in
which something is called a being simpliciter. Again, substantial forms are the
kind of beings for which to be is to make a substance actual in respect of
what it is. Hence, the act of being of a substantial form is the act of being
of the substance itself. This is why the actuality of the substantial form
makes the substance actual simpliciter and not only secundum quid. Still,
unless the substantial form is itself a substance, either being the whole
substance itself or being a substantial part of a composite substance, this act
of being cannot be attributed to it as to that which is but only as to that by
which the substance is. Finally, it is only substances that can be said to be
absolutely and without qualification, in the primary sense of being.
In these comparisons we have already seen that even among sub,stances
there are important differences with respect to their being. Even if all sorts
96. Cf. QDA q. 14. Indeed, this is precisely the point that Siger of Brabant,
unable to separate modi essendi from modi significandi in the way Thomas did, found
unacceptable in Thomas's conception of the intellective soul:
Praeterea, alia est ratio essendi formae materialis et compositi seu
formae per se subsistentis. Ratio enim essendi formae materialis est
secundum quam est aliquid aliud, ut ratio compositionis est secundum
quam habet esse compositum, et ratio figurae secundum quam habet
esse figuratum unde ratio essendi formae materialis est quod sit unita
alii. Ratio autem essendi compositi vel formae liberatae a materia est
quod sit ens per se et separate, non unum ens cum alio.... Et sunt istae
rationes essendi, qua aliquid habet esse unite ad materiam et qua aliquid
habet rationem subsistentis per se et separate, oppositae adeo ut eidem
inesse non possunt. Dnde anima intellectiva non potest habere rationem
per se subsistentis et, cum hoc, unum facere cum materia et corpore in
essendo.
Siger of Brabant, De anima intellectiva, in B. Bazan, Siger de Brabant (Paris: Louvain,
1972), pp. 79-80. Cf. also St. Thomas's De unitate intellectus, nn. 37-38.
136 GYULA KLIMA
of substances can be said to be without qualification, in the primary sense
of being, we could also see that the being of these various sorts of substances
was at the same time the being of their substantial forms, which, insofar as
they make them be what they are, are what distinguish these various sorts
from each other, determining their specific and generic kinds. Since their
actuality simpliciter is nothing but their actuality in respect of their substan-
tial form, their being, too, is determined to that form:
97
they can only be
actual inasmuch as their form permits them to be, for this (namely, to be
in respect of their substantial forms) is precisely what it is for them just to
be, simpliciter. But then these determinations of being in the determinate
kinds of substances are also limitations on the actuality in principle available
to a substance of a given, determinate kind. And if this is true, then
different determinations yield different limitations, which in turn imply
different degrees.
It is in this way, then, that from the analysis of the various analogical
senses of being in their relation to the primary sense, in which it is predica-
ble of substance, we can arrive at an idea of the different degrees of being
that is not only comprehensible but even plausible, even for different kinds
of substances themselves. (It makes no difference whether we are actually
able to correctly identify-let alone measure-these various degrees.) In
any case, the semantic question, again, is not whether what is said in
particular in arranging different kinds of substances according to different
degrees of being is true (that is, whether humans, say, are indeed a higher
form 6f being than worms, for example), for that is a question of metaphys-
ics; the semantic question is whether what is thus said makes good sense,
and if so, then what sense it makes. This has to be understood before we
inquire into truth, under pain of ignoratio elenchi.
VIII. THE GREAT CHAIN OF BEING
We have seen how the different analogous senses of being are all related to
the primary sense (signifying the subsistence of a substance), namely, how
these secondary senses are interpretable, in the framework of the inherence
theory of predication and the corresponding theories of signification and
supposition, as signifying being in some qualified sense that derives from
the primary sense by the addition of some diminishing qualification. How-
ever, we have also seen that even within the range of application of this
primary sense of being it is possible to distinguish different degrees of
being, insofar as different kinds of substances are different precisely be-
cause of their different substantial forms. These substantial forms may be
regarded in turn as just further (not properly diminishing but in a certain
97. In De hebd le. 2, n. 34.
way limiting) qualifications on some more central, more absolute sense of
being. These limiting qualifications yield differences that we may charac-
terize as constituting different degrees of being.
Let us take a closer look at this idea, and let us see how we can say that,
despite the fact that the actuality of a substantial form is an act of being
attributable to the substance that has this form in the primary, absolute
sense, the form itself imposes some limitation on this absolute being,
determining it to a certain degree in comparison to others. Our starting
point, again, should be a comparison of the various instances of the scheme
(EP):
(EP) (i) An S iS
2
P iff (ii) An S's [P] iSx iff (iii) An S is} with respect
to [P]
We have seen that the act of being signified by the qualified predicate in
(iii) is the same act of being as that signified by 'is
x
' in (ii). We have also
seen that if the subject term (i.e., the phrase S's [P]) of (ii) supposits for an
inherent, as opposed to a subsistent, form, then 'is
x
' can be truly predicated
of it only if it is not 'is}'. In this case, being can be attributed to the form
supposited for by this term only as to that by which something else [namely,
the substance supposited for by S in (iii)] is and not as to that which is.
However, if P is a substantial predicate, then the act of being signified by
'is
x
' in (ii) is the very same act as that by which the substance itself is,
absolutely and in the primary sense, for this is precisely what it is for P to
be a substantial predicate of the supposita of S.98 Therefore, provided P is
a substantial predicate of the supposita of S, we can supplement (EP) as
follows:
(EP*) (i) An S iS
2
P iff (ii) An S's [P] is
x
iff (iii) An S is} with respect
to [P] iff (iv) An Sis}
since the act of being signified by the predicate of (ii) is the same as the act
of being signified by the predicate of (iii), which in turn is the same act of
being signified by 'is}' as an absolute predicate of S in (iv) , we can see that
even the absolute predication of 'is}' in (iv) involves an implied determina-
tion or qualification, namely, the qualification of being provided by the
substantial form signified by P in the supposita of S. So if the predication of
'is}' in (iv) is true at all, it can be true only in the sense in which the
predication of 'is} with respect to [P]' is true in (iii) .99 This means that an
S can be only in the way things that have the substantial form signified by P
98. For, in general, a predicate P is substantial to a substance u iff
SGT('is
x
') (SGT(P) (u) (t)) (t) = SGT('is}') (u) (t); otherwise P is accidental to u. Cf.
n.95.
99. That is, ifP is a substantial predicate ofu, then SGT('is}') (u) (t) = SGT('is}
with-respect-to [P]')(u)(t).
138 GYULA KLIMA
can be, but not in any other way. What is, say, a diamond (provided 'diamond'
is a substantial predicate of diamonds) can be only in the way a diamond
can be, localized in space and time and characterized by certain capacities
and incapacities, such as being capable of scratching other solid bodies, but,
say, not being capable of self-propagation. In contrast, what is a tree (pro-
vided 'tree' is a substantial predicate of trees) can be only in the way trees
are-also localized in space and time but characterized by different capaci-
ties and incapacities, and so on.
It seems, then, that whatever substantial predicate we substitute for P
in the above scheme, the substantial form signified by that predicate will
always impose some determination and thereby a certain limitation on the
act of being signified by 'is
1
' in (iv). However, what if we can find a predicate
that signifies a substantial form which, not being distinct from this act of
being itself, does not impose any determination or limitation on the kind
of being signified by 'is
1
' in (iv)? Suppose we find such a predicate, and we
find that there is a being whose substantial form signified by this predicate
is the very act of being that is signified in it by the predicate 'is
1
' .100 Such
an act of being then would truly be called infinite, in the sense of not being
determined to the capacity of some form distinct from it. Therefore it is
only an entity that has such an act of being that could be said truly and
genuinely absolutely to be, without any restriction or limitation (whether
explicit or implicit) imposed on what is signified in it by the verb 'is' or by
the predicate 'being'. It is only to such an entity that these terms would most
properly, absolutely, and maximally apply, without any limitation of a par-
ticular kind or degree. Aquinas's metaphysical proofs are designed to show
precisely that there is such an entity, God, because God's nature, the "form"
signified by the term 'God', is identical to the act of being sigified by the
term 'being' in Him. But then this is why Aquinas can claim that 'He Who
Is' (qui est) is the most proper name of God, expressing most clearly the
unlimited character of Divine Being, and that even among substances,
which all are beings in the primary sense, it is God to whom this term most
properly and maximally applies.
101
But such a being, the nature of which is nothing but its act of being,
can be only one, for there could be two such beings only if the one would
be different from the other. Therefore, one of them would have to have
something, some form, that somehow qualifies and thereby restricts its
being. But then the nature of that being would not be the pure, unlimited,
unqualified act of being, so that other being would not be the kind of being
of which we assumed that we had two.
102
Different kinds of beings, which
are different because of their different forms, have their acts of being
qualified and limited differently by their forms, and thus these acts of being
100. That is, suppose we find a P such that for a substance u, SGT(P) (u) (t) =
SGT('is
1
') (u) (t).
101. sn q. 13, a. 11.
102. In De hebd le. 2, n. 33. SeC lb. 1, c. 42, n. 10.
are either more or less removed from the fullness of absolute being, thereby
constituting different degrees of being)03
However, even this overall hierarchy of being that is necessarily gener-
ated by the different substantial forms of different kinds of substances
(which are different precisely because they are nothing but more or less
restrictive qualifications of the absolute, unqualified being
104
) allows for
further, even finer degrees of being, on the level of individuals, in which the
same substantial nature may be more or less perfectly realized in the
actualization of the individuals' accidents.
How should we understand this claim? What does the accidental being
of a substance's accidents have to do with the degree of the substantial
being of the substance itself? Even granting that different kinds of sub-
stances can be arranged into a hierarchy with regard to the perfection of
their substantial being (even if we may not be able to determine the exact
position of any given kind in this hierarchy), is it not the case that any given
individual of any given kind either simply is or simply is not, at the degree
determined by its kind?
The answer to this last question, as one may expect, is yes, absolutely
speaking, but no with qualification. Absolutely speaking, any substance of
any given kind either is or is not, but if it is at all, absolutely speaking, it can
still be more or less; that is, it can still be or not be with some qualification
(it can still be to a certain extent or not be to that extent). But again, is this
not just play with words? If a substance already is by its substantial being, it
mayor may not have several sorts of accidental being. But why would the
actuality or nonactuality of any such accidental being have anything to do
with the degree or intensity of its substantial being?
To this we can say, first, that we must not forget that even if an
accidental act of being of a substance is not identified with its substantial
act of being, that accidental act is a certain act of the substance itself, an
act of being secundum quid, because of which it is the substance that is in
actuality in some respect. Indeed, it is only such actualities secundum quid
that manifest the degree of a given substantial being in the first place in
the overall hierarchy of beings determined by the substantial form of the
kind of the substance in question. The degree of a given sort of substantial
being determines, and thereby manifests itself precisely in, the range of
actualities that are in principle available to or attainable by any particular
substance that holds that degree of substantial being by belonging to its
kind, that is, that has the kind of substantial form that determines that
degree of substantial being. For example, supposing with Aquinas that
living is a sort of substantial being that is of a higher degree than sheer
103. SeC lb. 3,c.97,n. 3.
104. For this point see J. F. Wippel, ''Thomas Aquinas and Participation," in
Studies in Medieval Philosophy, ed. J. F. Wippel (Washington, D.C.: The Catholic
University of America Press, 1987).
140 GYULA KLIMA
lifeless existence,105 we can say that this is so because the notion of the
mode of being which is living imposes less limitation on the notion of
absolutely unqualified being than does the notion of a lifeless existence
(perhaps it imposes no restriction at all) ,106 This manifests itself precisely
in the fact that the actualities in principle available to or attainable by a
living being are in principle unavailable to or unattainable by a lifeless
being, just because of the difference between what it is for something to
live and what it is for something merely to be but not to live.
But is there still not some problem here? Even granting that a living
being can do things that a lifeless being cannot, which may be regarded as the
manifestation ofa higher degree ofsubstantial being, is not the converse also
possible, indeed, actually true, namely, that a lifeless being can do things that
a living being cannot? And ifso, is not this latter capacity a manifestation of a
higher degree of substantial being as well? In that case do we not have to
conclude that the same lifeless being has a higher degree ofsubstantial being
than the living being, in contradiction to what wejust conceded?
Of course, it is true that lifeless things can do several things that living
things cannot do; a diamond, for example, can scratch glass, which a tree
or a cat cannot do. (In fact, properly speaking, a diamond cannot scratch
glass by its own activity either, but this is beside the point. The point is that
the body of a tree or a cat lacks the capacity that a diamond has, namely, to
scratch glass when it is rubbed against it, even if this requires the action of
an external force.) But we should realize that the mere difference in the
range of available capacities does not automatically translate into differ-
ences in the degree of substantial being. If a capacity is available precisely
because of the unavailability of some other capacity or a whole range of
other capacities, which in its turn is directly an indication of a higher degree
of substantial being, then the availability of the former is an indication not
of a higher but of a lesser degree of substantial being. This is the case in the
example: a diamond has the capacity to scratch glass precisely because it has
the sort of molecular structure that prevents it from performing all sorts of
activities in which the degree of being we call life manifests itself. Hence,
this capacity is precisely an indication of a more limited form of being,107
From our present point of view such an example, or any other particu-
lar case, is relevant only to the extent it helps illustrate how the general
105. Whether this is indeed so is a metaphysical question, again left undeter-
mined by our formal, semantic considerations, but for the sake of the example it is
a quite plausible assumption in any case.
106. Cf., for example, sn q. 18, a. 3; ScG lb. 1, cc. 97-98.
107. Note that in this argument, we already assumed with Aquinas that life is a
kind of being of a higher degree than lifeless existence. This argument shows only
that the fact that lifeless entities can have certain capacities that living beings do not
have does not contradict this assumption. Of course, it is a further issue whether
and why we are justified in this assumption. For this see Aquinas's texts referred to
in n. 106.
semantic principles concerning the concept of being should be applied in
the particular discussions of these particular cases.
l08
These principles
themselves, however, being presupposed by the discussion of particular cases,
should be regarded as valid regardless of the outcome of any particular
discussion, provided they are consistent in themselves and with one
another.
108. Consider the following, more drastic example. God cannot scratch His
nose. This is not an incapacity, absolutely speaking, for this is an incapacity only
insofar as a higher capacity excludes the presence of the limiting opposite capacity.
My capacity to scratch my nose is a capacity that directly involves the limitation on
the form of being I have, namely, the limitation that it is a sort of bodily existence,
restricted to a portion (a rather small portion at that!) of space and time. So I can
have this particular, limited capacity only and precisely on account of the obviously
restricted mode of being I have. Thus, despite possible appearances to the contrary,
this capacity is obviously an indication of a lower degree of being, whence the lack
of this capacity in, or rather its sheer inapplicability to, Divine Being is just a further
indication of the absolute perfection of that Being. See Thomas on why we cannot
truly and properly predicate of God terms that signify perfections but also neces-
sarily involve limitations of being. Cf. sn q. 13, a. 3; ISNd. 22, q. 1, a. 2.

Semantic Principles of Aquinas

Uploaded by

Copyright:

Available Formats

Semantic Principles of Aquinas

Uploaded by

Document Information

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

Semantic Principles of Aquinas

Uploaded by

Copyright:

Available Formats

MEDIEVAL PHILOSOPHY AND THEOLOGY

Volume 5, Number I, March 1996

You might also like