Philosophia (2013) 41:589602

DOI 10.1007/s11406-013-9481-y

Aquinas on Intellectual Cognition: The Case

of Intelligible Species
Elena Baltuta

Received: 5 February 2013 / Revised: 5 April 2013 / Accepted: 14 April 2013 /

Published online: 24 August 2013
# Springer Science+Business Media Dordrecht 2013

Abstract The paper argues in favour of a direct realist reading of Aquinass theory
of intelligible species, in opposition to the recent representationalist challenges. In
order to secure the direct realist reading, the paper follows three steps: a short
description of Aquinass process of cognition, a survey of the direct realist arguments
and the analysis of the representationalist interpretation. The final step consists of
investigating the representationalist reading as it is suggested by two scholars,
Claude Panaccio in Aquinas on Intellectual Representation and Robert Pasnau in
Theories of Cognition in the Latter Middle Ages. Thus, the paper can be construed as
a reply to these two authors, due to the thorough attention paid to their argumentative
trails. With regard to Panaccios reading the paper focuses on the identity between the
intelligible species and the essence of the extra-mental object and argues that
Panaccio understands identity in a very narrow sense. Concerning Pasnaus line of
reasoning the focus is on the primum cognitum status of the intelligible species, and
the main argument is that intelligible species is understood by Aquinas as the quo and
not the quod of cognition. As the paper shows, neither one, nor the other interpretation poses a threat to the direct realist reading of Aquinass intelligible species.
Keywords Cognitive and causal role . Direct realism . Formal identity . Intelligible
species . Primum cognitum . Representationalism . Thomas Aquinas
This article revolves around one question: what role do the intelligible species play in
Thomas Aquinass theory of human cognition? The question might seem simple, but
the answer causes enough debate among scholars to split them in two major camps:
direct realists and representationalists. The classical view, direct realism, treats the
intelligible species as playing a causal role. Scholars such as Robert Pasnau and
Claude Panaccio have recently challenged this view and suggested that a
This paper was supported by the CNCS-UEFISCDI under the project PN-II-ID-PCE-2011-3-0661.
E. Baltuta (*)
Alexandru Dragomir Institute for Philosophy, Romanian Society for Phenomenology, Bd. Mihail
Kogalniceanu nr. 49, ap. 45, Bucharest, Romania RO-050104
e-mail: elena_baltuta@yahoo.com

590

Philosophia (2013) 41:589602

representationalist answer, according to which the intelligible species have a cognitive role, is more adequate to Aquinass doctrine of intelligible species. I will argue
that the classical view is sound enough to resist the representationalist challenge. In
order to do this, I will follow three steps: I will present a sketch of Aquinass first
operation of the intellect, after which the direct realist interpretation will be introduced, and, in the end, I will analyze the representationalist arguments put forth by
Claude Panaccio and Robert Pasnau. Such an investigation is worth pursuing for at
least two reasons: it provides the necessary tools for solving a conflict of interpretations, but, more importantly, it helps us understand the implications and the complexity of such a simple question, which in turn triggers a new way of understanding
both Aquinass epistemology and the initial question itself.
According to Aquinas, the simple act of looking at an object, which normally does
not take more than a few milliseconds, is made possible by a whole process that
involves complex mechanisms of cognition, enabling the human beings to receive,
sort, store and transform the information about the external object. The cognition of
an external object starts with the external senses. For example, when a cognizer, call
him John, is looking at a piece of amber, his five external senses receive the sensible
species1 of the object. The sensible species contains the forms of the properties of the
external object. According to the realists, they are the means by which (id quo)2 our
senses actually sense the object, and not what (id quod) is sensed. At this particular
moment, John can see the yellow colour of the semi-precious stone, can sense the
texture, and can observe its size and shape, which are called the proper and common
sensible.3 After the external senses are informed, the internal senses4 gather all the
information and form a phantasm, an image of the external object. Now John can
see this particular stone with this particular shade of yellow, with a specific textural
structure, placed in this particular moment in time and under specific spatial coordinates. From the phantasms, to which the intellect will always need to return in order
to get a glimpse of the individual, the agent intellect abstracts the intelligible species.5
John is now able to apprehend in what size, in what shape and in what genus this
object consists. Inside the realist framework, the intelligible species is the thing by
means of which (id quo) we cognize, in virtue of its similarity relation to the essential
form of the object, and not that which (id quod) is cognized. At the end of this
process, John is able to know the quiddity6 or the essence of the external hylomorphic
1

The sensible species are, as the realists would say, the first causal intermediary entities we come into
contact with. They are numerically different in different cognizers, and are the quo of our cognition, the
means by which we know; based on their similitude with the object, they work as causal mediators.
2
See S.c.G. (II, 75, 7; II, 98) (Aquinas 1961); S.Th. (I, q.76, a.2, ad 4; I, q.85, a.2) (Aquinas 1952).
3
The objects of sensation can be the proper objects, namely colour, smell, etc., the common objects, like
movement, size, etc., and the accidental objects, like trees, people, etc., which exceed the power of a single
sense.
4
The internal senses are: the common sense, which acts like the root of all external senses, the cogitative
power, which prepares the images and compares different individual intentions, the memory, which stores
the images and recognizes past experiences, and imagination, which retains and combines the images.
5
According to the realist interpretation, the intelligible species is a causal intermediary entity which, just
like the sensible species, is assimilated by the cognizer. This stands for the interpretation of the intelligible
species as the quo and not the quod of cognition.
6
The quiddity of an object signifies what is common to all natures, through which the various beings are
placed in various genera and species. In the case of human beings, humanity is the essence that can be
signified by a definition, which includes both the common matter (body) and the substantial form (soul).

Philosophia (2013) 41:589602

591

object. The final step of this first operation ends with the intellect actively forming a
concept, which can later be signified by a spoken word, helping John actually to
understand the external object.
Each intermediary entity, be it the sensible or the intelligible species, marks a different
stage in the process of cognition. The manner in which the relation between these
entities and the external object is understood marks the difference between a realist and a
representationalist interpretation of Aquinass theory of cognition. The realist interpretation analyses this relation as one of formal identity. Basically this means that the
intelligible species which inform the intellect have the same formal content as the
objects essential form: when John understands the piece of amber, he has in mind the
essential form of amber, the same one which exists in the external object. This identity is
possible because the essential form of amber is able to exist not only in the matter-form
compound, but also in Johns mind, where it is separated from its material substrate. In
the external object, the form exists materially and naturally, while in the mind of the
cognizer, it has an immaterial and intentional way of existence. This is the so called
thesis of the double existence7 of the same form, and it is crucial for realism, because it is
the assertion on which objectivity and veridicality of the process of cognition are based.
The id quo status of the intermediary entities, the formal identity and the double
existence of a form are the most common themes discussed by the advocates of the
realist interpretation, but there are, also, two further ones: the idea that a form has a
double aspect,8 as a mental accident is a particular, while its content is an universal,
and on a special way of understanding the similitude (similitudo) aspect of the
intermediary entities. The double aspect argument states that, although the intelligible
species has a universal mode of existence in the intellect due to the fact that it is able
to produce cognition, it also has a singular existence there: although cognition is
about universals, that which enables cognition has the existence of a singular in the
intellect. The special approach to similitude is based on the recognition that in most of
the passages where Aquinas speaks about species he describes them as a similitudo of
the objects form. At this point one can suggest similitudo is more in accordance with
a representationalist interpretation, than with a realist one. One can suggest that this
character of a form can lead to its being understood as a sign for the external object, a
sign which needs to be decoded in order to understand what it stands for. For
example, when we say that A is similar to B or A is like B, we can understand that,
though A and B share some features, they do not share all of them, therefore A and B
are not identical. And if one knows A by means of B, which is similar to A only in
some respects, then how come one is able to know A for itself rather than just its
features shared with B? Still, the similitudo character of the species is advocated as
one of the main arguments in favour of the realist interpretation, and this is possible
because similitudo is a technical term (Perler 2000, 115) which stands not for sign,
but for agreement or sharing of a form. To overcome any possible problems related
with similitudo, the realists point to the analysis of one passage from Aquinass
Summa Theologiae (I, q.4, a.3, co) in which it is stated that the term similitudo is to be
understood not in the sense of similarity or likeness, but in the sense of agreement or
7

See S.c.G. (III, 49, 2) (Aquinas 1961), Quod. (VIII, q. 2, a. 2, co.) (Aquinas 1996)
See S.c.G. (I, 46) (Aquinas 1961); II Sent. (d. 17, q. 2, a. 2, co.) (Aquinas 1929); Perler (2000), 114; Spruit
(1994), 169.
8

592

Philosophia (2013) 41:589602

sharing of form, therefore it shouldnt be understood as a mere representation, but as

a formal identity. The fact that the form changes its way of existence, from material to
immaterial can be seen as an argument in favour of casting aside the term identity
(identitas).
To conclude, the realist interpretation rests on: (1) the formal identity between the
intelligible species and the essential form of the extra-mental object, and (2) the
double aspect of the form, as an accident in the mind of the cognizer and as an
universal because it leads to cognition. The first argument has the following consequences: (1.1) the id quo status of the intelligible species that by which and not that
which are cognized; (1.2) between the intelligible species and the essential form of
the extra-mental object there is a similitudo relationship that is understood as agreement in form or as sharing of form; (1.3) the double existence of a form material and
natural in the extra-mental object, immaterial and intentional in the mind of the
cognizer.
Allow me to explain the realist arguments by means of an example. The formula
C10H16O accompanied by the appropriate knowledge of chemistry enables John to
recognize amber. But though, on the one hand, C10H16O can be taken to be the same
thing as the form of amber (formal identity), on the other hand, it exists differently in
the material compound, where it is connected with matter, and in the mind of the
cognizer, where it has no connection whatsoever with any material determinations
(double existence). This form, which resides in the mind, C10H16O, can be seen as the
intelligible species, and thus is not the thing we know, but only a means of knowing
the external object, the id quo of cognition. The similitudo relationship is based on the
sharing of the form. Namely, C10H16O is the same form which informs the matter in
making the natural compound known as amber, and exists in the mind of the
cognizer, apart from the material constituents, enabling cognition of the external
object. Though the intelligible species has a singular existence in the intellect, it also
has a universal mode of existence by being able to produce cognition (double aspect).
In other words, although cognition is about universals, that which enables it, the form
in the mind, has the existence of a singular in the intellect. Using the same amber
example, we can say that C10H16O has a singular and accidental mode of existence in
Johns mind, and, at the same time, it enables his intellect to achieve cognition of the
universal essence of amber.
In an article from 2002,9 Claude Panaccio criticizes the formal identity theory by
advancing the hypothesis that the intelligible species cannot be identical with the
quiddity of the extra-mental object, since the intellects proper object, the quiddity, is
not multipliable, while the intelligible species is multipliable in different cognizers.
This results in the fact that, though the intelligible species is that by which (id quo) we
cognize the external object, it is not being assimilated with it, it does not redirect the
cognition, but it represents the essential form of the external object. Roughly speaking, by combining the attributes of the intelligible species the quo of cognition,
Claude Panaccio, Aquinas on Intellectual Representation, in Chaiers depistmologie 002/2652 (2000).
In his latest article, Mental representation, in The Cambridge History of Medieval Philosophy, vol. 1, eds.
Robert Pasnau and Christina van Dyke, Cambridge, (2010), 34656, Panaccio seems to leave room for a
more moderated view, which takes into account the fact that Aquinass theory of intellectual cognition is
build on two different levels, one of the intelligible species and one of the concept, which interwine. Apart
from this, the line of argumentation does not go astray from the previous one.

Philosophia (2013) 41:589602

593

multipliable in different cognizers and having an accidental mode of existence and

those of the quiddities the quod of cognition, non-multipliable in different cognizers
and having a universal mode of existence , Panaccio builds his way up to the
conclusion that these two entities are not and cannot be treated as identical.
In fact, what Panaccio does, is to reject the formal identity thesis on three grounds:
(A) the object of cognition is the quiddity of the thing, the intelligible species is not
the object of cognition, and therefore the intelligible species is not identical with the
quiddity of the thing; (B) the quiddity is not multipliable in different cognizers, the
intelligible species is multipliable in different cognizers, and therefore the intelligible
species is not identical with the quiddity; (C) the intelligible species has an accidental
mode of existence in the mind, the quiddity is an universal, so it has a universal way
of existence; something universal cannot have an accidental existence, and therefore
the intelligible species is not identical with the quiddity. Let us consider each of these
arguments.
Reformulated, Panaccios first rejection (A) seems to take the following form: how
is it possible for the quo of cognition to be identical with the cognized thing and yet
not to be cognized? The first argument appears to be right, but let us take a look at
how Aquinas describes the intelligible species as being in a certain way the quiddity
and the nature of the thing itself according to intelligible existence, and not according
to natural existence, as it is in things10 (Quod.VIII, q.2, a.2, co.) What Aquinas says
here is that the form can have two different types of existence: a natural and material
existence in the hylomorphic compound, and an immaterial and spiritual one in the
mind of the cognizer. Let us take an example: for a chemist multiple concentric layers
of calcium carbonate (CaCO3) represent a pearl. Though the layers of calcium
carbonate and the pearl are one and the same thing, CaCO3 exists in a material and
natural way, connected with matter, in the pearl, and in an immaterial and intentional
way in the mind of the chemist, where it exists separated from any material determinations. CaCO3 is, according to the realist interpretation, the intelligible species of
pearl, and therefore it is not the one we cognize, the quod of cognition, but that by
which we cognize, the quo. What we actually cognize in a simple cognitive act is the
pearl. Bearing in mind this double existence of the form, Panaccios first argument
seems to lose its robustness because, from the standpoint of formal content, the
intelligible species and the quiddity are in fact identical, meaning that they both refer
to the same thing.
10
() quodammodo ipsa quidditas et natura rei secundum esse intelligibile, non secundum esse natural,
prout est in rebus. In this article I will make use of the following Latin editions and shortcuts: Quaestio
disputata De spiritualibus creaturis (D.S.C.), in Opera Omnia, ed. J. COS, XXIV-2 (Roma & Paris:
Commissio Leonina - Les ditions du Cerf, 2000); Quaestiones disputatae de Potentia (D.P.), ed. P. M.
Pession (Torino & Roma: Marietti 1965); Quaestiones disputatae de Veritate (Q.D.V.), in Opera Omnia, ed.
H. F. Dondaine, XXII (Roma & Paris: Commissio Leonina & Cerf 19701976); Quaestiones quodlibetales
(Quod), in Opera Omnia, ed. R. A. Gauthier, XXV (Roma & Paris: Commisio Leonina & Les ditions du
Cerf, 1996); Scriptum super libros Sententiarum magistri Petri Lombardi episcopi Parisiensis
(Super.Sent.), ed. P. Mandonnet (Paris: 1929); Summa Theologiae (S. Th.), ed. P. Caramello (Torino &
Roma: Marietti 1952); Summa contra Gentiles (S. c. G.), ed. C. Pera (Torino & Roma: Marietti 1961). I also
make use of the following translations: for Summa Theologiae the translation of the Fathers of the English
Dominican Province; for I Sent. (d. 5, q. 1, a. 2, co.), II Sent. (d. 4, q. 1, a. 1, ad 4), D.S.C. (a. 9, ad 6) and
Q.D.V. (q. 1, a. 11, co.) Pasnaus translation from Theories of Cognition in the Latter Middle Ages; for II
Sent. (d. 17, q. 2, a.1, ad 3) and Quod. (VIII, q.2, a.2, co.) my own translations.

594

Philosophia (2013) 41:589602

Panaccios second argument against the identity between the intelligible species
and the quiddity of the extra-mental thing tackles the problem of multiplication of
species in different cognizers. As he puts it: the quiddity is not multipliable in
different cognizers, the intelligible species is multipliable in different cognizers,
and therefore the intelligible species is not identical to the quiddity of the thing. This
second argument looks more convincing than the first one, if we take into account the
transitivity of identity. If all intelligible species are identical with the quiddity, they
will, by the transitivity of identity, be identical with each other; but they are not
identical with each other, therefore the intelligible species are not identical with the
things quiddity. However, let us not rush to any conclusions and instead ask one
question: how should identity be understood in this case of cognition? In various places
Aquinas states that the intelligible species is multipliable in different cognizers.11 This
entails that if I and the abovementioned chemist look at the same miniature we both see
the redness of the paint, but our intelligible species are not the same. So far so good, but
despite the fact that our species are different and, in consequence, they are different from
the quiddity of the thing, how come we are still thinking about the same thing? The only
way in which Panaccios second argument can make sense is if we understand identity as
strictly numerical identity. In this case, our species are indeed different and both are
different from the quiddity of the thing. There is but one crucial flaw in this reasoning:
the issue at stake here is not numerical, but formal identity. In other words, though
intelligible species1 and intelligible species2 are not identical because they are accidents
in the mind of the cognizer, they are identical qua representational content:
To the third it is to be said that, according to Avicenna, the known species can
be considered in two ways, either according to the existence that it has in the
intellect, and in this way it has singular existence; or with respect to its being a
similitude of such a known thing, as it leads to its cognition, and from this
aspect it has universality: because it is not a similitude of this thing according to
its being this thing, but according to the nature in which it agrees with others
individuals of its species.12 II Sent. (d.17, q.2, a.1, ad 3)
As it can be seen the intelligible species can be understood from two perspectives: as
an accident in the mind of the cognizer, and as an universal because it leads to cognition.
Regardless of whether the intelligible species of cinnabar is or not in my intellect, my
intellect will not cease to exist. At the same time, these accidents of the mind enable the
cognition of universal essences, given the fact that they have a universal content, since
the intellects immateriality allows only the cognition of immaterial universal entities,
and, finally, from this perspective the intelligible species are universal. CaCO3 has a
singular accidental mode of existence in my mind and, at the same time, it can be
See for example S.Th. (I, q. 76, a. 8, ad 4) (Aquinas 1952); Comp. Th. (1, 85): Quia si est alius
intellectus in me, alius in te, oportebit quod sit alia species intelligibilis in me, et alia in te, et per
consequens aliud intellectum quod ego intelligo, et aliud quod tu. Erit ergo intentio intellecta multiplicata
secundum numerum individuorum, et ita non erit universalis, sed individualis (Aquinas 1979).
12
Ad tertium dicendum, quod secundum Avicennam species intellecta potest dupliciter considerari: aut
secundum esse quod habet in intellectu, et sic habet esse singulare; aut secundum quod est similitudo talis
rei intellectae, prout ducit in cognitionem ejus; et ex hac parte habet universalitatem: quia non est similitudo
hujus rei secundum quod haec res est, sed secundum naturam in qua cum aliis suae speciei convenit
(Aquinas 1929).
11

Philosophia (2013) 41:589602

595

regarded as an universal because, through its universal content, it enables me to achieve

cognition of the universal nature of pearlness. If the species is considered from this
double perspective, Panaccios second argument seems no longer sound, because
though the intelligible species1 and intelligible species2 are different because they are
singular accidents of two distinct minds, they both share the same representational
content. In other words, between the two intelligible species there is only a formal
identity and no numerical identity as Panaccios argument implies.
Before discarding Panaccios interpretation there is yet one more argument to be
taken into account, namely (C)13: the intelligible species has an accidental mode of
existence in the mind, quiddity is an universal, so it has a universal way of existence;
something universal cannot have an accidental existence, and therefore the intelligible species is not identical with the quiddity of the thing. Despite the fact that both
premises are true, Panaccios conclusion is wrong and the abovementioned double
aspect of the intelligible species is enough to prove this. Though the intelligible
species has an accidental mode of existence in the mind of the cognizer, it also has a
universal way of existence given its representational universal content which makes
possible cognition of universal nature of the extra-mental thing.
Despite its appealing architecture, Panaccios argumentation does not seem convincing. But perhaps we might find more persuasive arguments in favour of representationalism if we follow Pasnaus demonstration centred not on formal identity,
but on the primum cognitum status of the intelligible species. Pasnau (1997) proposes an
act-object theory of Aquinass cognition grounded on three argumentative steps. The
act-object doctrine states that we perceive the world by perceiving the species or, in other
words, the species are received in order to receive something else. Unlike Panaccio,
Pansau does not deny the formal unity,14 on the contrary, he uses it as an argument in his
favour. The three steps for proving the validity of the act-object doctrine are: (1)
pinpointing the passages where Aquinas affirms an act-object doctrine, (2) discovering
that in none of Aquinass later works the act-object doctrine is denied and (3) showing
that the official position regarding species allows an act-object interpretation.
It is within the framework of the act-object doctrine that we should understand
Pasnaus statement that the intelligible species is in some sense the object of
cognition. The textual evidence invoked in the defence of this view consists of some
passages, most of them extracted from works such as I Sent. (d.35, q.1, a.2, co.),
Q.D.V. (q.1, a.11, co.) or D.S.C. (a.9, ad.6). For a better understanding of Pasnaus
reading, let us review the passages on which he builds his interpretation. I shall start
with the passage from the Sentences, because there Aquinas states clearly that first the
intelligible species is cognized:
It should be known, nevertheless, that a thing is said to be intellectively
cognized in two ways, just as a thing seen. For there is a first thing seen, which
is a species of the visible thing existing in the pupil, which is also the
13

Besides this triple attempt to reject the realist interpretation at the level of the intelligible species,
Panaccio also exploits the ambiguity of similitudo and stresses the fact that similarity is different from
identity.
14
Pasnau (1997) refutes the generally accepted thesis that formal identity secures direct realism, but I find
his line of argumentation unconvincing. For an article which goes in the same direction as mine, see Klima
(1996).

596

Philosophia (2013) 41:589602

completion of the one seeing and the source of vision and the intermediary light
of the visible thing. And there is a second thing seen, which is the thing itself
outside the soul. Likewise, the first thing intellectively cognized is the likeness
of the thing itself, which is intellectively cognized through that likeness.15 I
Sent. (d.35, q.1, a.2, co.) (P1)
In the face of this kind of evidence one has two options: either one thinks that
Aquinas is committed to a brand of representationalism, or that this is nothing but a
singular passage from one of his earliest writings, where his doctrine wasnt yet very
clearly articulated. Let us consider that this is not a slip on Aquinass part. If this is the
case, then we need to find some reinforcements for this interpretation, and Pasnau
suggest the following passage:
As regards the apprehension of the senses, it must be noted that there is one
type of apprehensive power, for example, a proper sense, which apprehends a
sensible species in the presence of a sensible thing; but there is also a second
type, the imagination, for example, which apprehends a sensible species when
the thing is absent. So, even though the sense always apprehends a thing as it is,
unless there is an impediment in the organ or in the medium, the imagination
usually apprehends a thing as it is not, since it apprehends it as present though it
is absent. Consequently, the Philosopher says: Imagination, not sense, is the
master of falsity.16 Q.D.V. (q.1, a.11, co.)
Pasnau takes apprehension to have cognitive implications and therefore it is clear
that the act-object doctrine is not far away. But, let us not jump to any conclusions
and see what the consequences of this cognitive reading are. As far as I see things,
this reading can have two consequences: either apprehension has two acts, or it has
two objects. The first consequence is ruled out by Pasnau on the grounds of what
Aquinas states in II Sent. (d.4, q.1, a.1, ad.4):
In the case of cognitive power, there is a single conversion to the things
species and to the thing itself. Hence someone is not said to infer17
Since Aquinas rules out any form of cognitive idealism,18 Pasnau moderates his
views and puts forward a theory of cognition characterized by a single non inferential
Sciendum tamen est, quod intellectum dupliciter dicitur, sicut visum etiam. Est enim primum visum
quod est ipsa species rei visibilis in potentia existens, quae est etiam perfectio videntis, et principium
visionis, et medium lumen respectu visibilis: et est visum secundum, quod est ipsa res extra oculum.
Similiter intellectum primum est ipsa rei similitudo, quae est in intellectu; et est intellectum secundum quod
est ipsa res, quae per similitudinem illam intelligitur (Aquinas 1929). When discussing Pasnaus interpretation I will use his own translations for a better understanding of his decisions.
16
Sed circa apprehensionem sensus sciendum est, quod est quaedam vis apprehensiva, quae apprehendit
speciem sensibilem sensibili re praesente, sicut sensus proprius; quaedam vero quae apprehendit eam re
absente, sicut imaginatio; et ideo semper sensus apprehendit rem ut est, nisi sit impedimentum in organo,
vel in medio; sed imaginatio ut plurimum apprehendit rem ut non est, quia apprehendit eam ut praesentem,
cum sit absens; et ideo dicit philosophus in IV Metaph., quod sensus non est dicens falsitatis, sed phantasia
(Aquinas 1970).
17
() quod virtutis cognoscitivae est una conversio in speciem rei et in rem ipsam; () non dicitur
conferre (Aquinas 1929).
18
If sciences were to be about objects in our mind, they would all turn out to be psychology. See S. Th. (I,
q. 85, a. 2) (Aquinas 1952).
15

Philosophia (2013) 41:589602

597

act of conversion toward two formally identical objects, the intelligible species and
the quiddity of the external object.
As coherent as this reading might be, it rests on two fallacious understandings: the cognitive implications of apprehensio and the semi-reifying of the
species, namely their understanding as the immediate object of cognition. If
we take a closer look at what Aquinas states, in various19 places the intelligible species is treated, in simple cases of cognitive acts, as the quo of
cognition and only in the case of introspection as the quod of cognition. In
addition, the Latin apprehensio usually has no cognitive connotations, so it
would be rather unusual for Aquinas to use it in Q.D.V. (q.11, a.1, co.) as
Pasnau suggests. A better translation and understanding of apprehensio would
be, in my opinion, to grasp, a verb which has causal connotations. But this is
just a personal opinion; the real flaws in Pasnaus argumentation rest on
ignoring the double existence of the form, and on a deficient understanding
of the formal identity. Aquinas speaks about only one act of conversion
toward the species and the object because the species and the object have
the same formal content which exists in two different ways: a material and
natural way in the extra-mental object, and an immaterial and intentional way
in the species.20 So far weve seen that Pasnaus line of argumentation fails to
align with Aquinass doctrine of the species. The only irrefutable piece of
evidence is the primum visum passage (P1) which, unfortunately, is never
reiterated by Aquinas.
The second line of Pasnaus case rests on the observation that later on, in his
work, Aquinas does not reject the act-object doctrine, but an idealist version of
representationalism. Pasnau is aware of the second article from the 85th question of the first part of Summa Theologiae, where Aquinas says that the
intelligible species is that by which we cognize, and not the object of cognition, but he doesnt find it a strong argument against what he calls representational realism knowledge is concerned with the external world, but both
the species and the object are understood; the species are directly understood
and we never directly apprehend the external objects, though cognition is never
distorted. This type of representationalism is opposed to the representational
idealism which states that the only things we ever know are our inner representations and we never understand the real object. It is this latter form of
representationalism which Pasnau believes is rejected in Summa Theologiae (I,
q.85), and his belief is right. Here Aquinas rejects Protagoras and Platos
views according to which we only cognize species. The consequences of such
a theory, relativism and subjectivism, are most bizarre. But let us take a look at
Aquinass words:
I answer that, some have asserted that our intellectual faculties know
only the impression made on them; as, for example, that sense is

19
20

See S.c.G. (II, 75, 5) (Aquinas 1961); S.Th. (I, q. 5, a. 2, sc; I, q. 85, a. 2) (Aquinas 1952).
For a similar reading of the passage see Perler (2002), 85.

598

Philosophia (2013) 41:589602

cognizant only of the impression made on its own organ. According to

this theory, the intellect understands only its own impression, namely, the
intelligible species which it has received, so that this species is what is
understood. This is, however, manifestly false for two reasons. First,
because the things we understand are the objects of science; therefore if
what we understand is merely the intelligible species in the soul, it would
follow that every science would not be concerned with objects outside the
soul, but only with the intelligible species within the soul; thus, according
to the teaching of the Platonists all science is about ideas, which they held
to be actually understood. Secondly, it is untrue, because it would lead to
the opinion of the ancients who maintained that whatever seems, is true
[Aristotle, Metaph. iii. 5, and that consequently contradictories are true
simultaneously. For if the faculty knows its own impression only, it can
judge of that only. Now a thing seems according to the impression made
on the cognitive faculty. Consequently the cognitive faculty will always
judge of its own impression as such; and so every judgment will be true:
for instance, if taste perceived only its own impression, when anyone with
a healthy taste perceives that honey is sweet, he would judge truly; and if
anyone with a corrupt taste perceives that honey is bitter, this would be
equally true; for each would judge according to the impression on his
taste. Thus every opinion would be equally true; in fact, every sort of
apprehension. Therefore it must be said that the intelligible species is
related to the intellect as that by which it understands: which is proved
thus. There is a twofold action (Metaph. ix, Did. viii, 8), one which
remains in the agent; for instance, to see and to understand; and another
which passes into an external object; for instance, to heat and to cut; and
each of these actions proceeds in virtue of some form. And as the form
from which proceeds an act tending to something external is the likeness
of the object of the action, as heat in the heater is a likeness of the thing
heated; so the form from which proceeds an action remaining in the agent
is the likeness of the object. Hence that by which the sight sees is the
likeness of the visible thing; and the likeness of the thing understood, that
is, the intelligible species, is the form by which the intellect understands.
But since the intellect reflects upon itself, by such reflection it understands both its own act of intelligence, and the species by which it
understands. Thus the intelligible species is that which is understood
secondarily; but that which is primarily understood is the object, of which
the species is the likeness. This also appears from the opinion of the
ancient philosophers, who said that like is known by like. For they said
that the soul knows the earth outside itself, by the earth within itself; and
so of the rest. If, therefore, we take the species of the earth instead of the
earth, according to Aristotle (De Anima iii, 8), who says that a stone is
not in the soul, but only the likeness of the stone; it follows that the soul

Philosophia (2013) 41:589602

599

knows external things by means of its intelligible species.21 S.Th. (I,

q.85, a.2, co.)
After presenting the outcomes of Protagoras and Platos positions, Aquinas goes
on by stating the adequate understanding of species as those by which (id quo) the
sight sees and the intellect understands. The intelligible species becomes the objects
of cognition only when the intellect reflects upon itself; and even then, the species are
what is understood only secondarily, because the extra-mental object remains the
primary object of cognition. Indeed, Aquinas does not argue here against representationalist realism (or the act-object doctrine), but he doesnt defend it either. The text
posits clearly that the intelligible species play a causal and not a cognitive role. If
Aquinas actually advocated a view like that advanced by Pasnau, why would he not
defend it here? Why would he not present it as the valid option with regard to the
intelligible species? Why did he choose to put forth a direct account of the intelligible
species? To questions like these Pasnau points towards the following passage from
D.S.C. (a.9, ad 6):
For both of these activities, the intelligible species is precognized, by which
the possible intellect is actuated; because the possible intellect does not operate
Respondeo dicendum quod quidam posuerunt quod vires cognoscitivae quae sunt in nobis, nihil
cognoscunt nisi proprias passiones; puta quod sensus non sentit nisi passionem sui organi. Et secundum
hoc, intellectus nihil intelligit nisi suam passionem, idest speciem intelligibilem in se receptam. Et
secundum hoc, species huiusmodi est ipsum quod intelligitur. Sed haec opinio manifeste apparet falsa ex
duobus. Primo quidem, quia eadem sunt quae intelligimus, et de quibus sunt scientiae. Si igitur ea quae
intelligimus essent solum species quae sunt in anima, sequeretur quod scientiae omnes non essent de rebus
quae sunt extra animam, sed solum de speciebus intelligibilibus quae sunt in anima; sicut secundum
Platonicos omnes scientiae sunt de ideis, quas ponebant esse intellecta in actu. Secundo, quia sequeretur
error antiquorum dicentium quod omne quod videtur est verum; et sic quod contradictoriae essent simul
verae. Si enim potentia non cognoscit nisi propriam passionem, de ea solum iudicat. Sic autem videtur
aliquid, secundum quod potentia cognoscitiva afficitur. Semper ergo iudicium potentiae cognoscitivae erit
de eo quod iudicat, scilicet de propria passione, secundum quod est; et ita omne iudicium erit verum. Puta si
gustus non sentit nisi propriam passionem, cum aliquis habens sanum gustum iudicat mel esse dulce, vere
iudicabit; et similiter si ille qui habet gustum infectum, iudicet mel esse amarum, vere iudicabit, uterque
enim iudicat secundum quod gustus eius afficitur. Et sic sequitur quod omnis opinio aequaliter erit vera, et
universaliter omnis acceptio. Et ideo dicendum est quod species intelligibilis se habet ad intellectum ut quo
intelligit intellectus. Quod sic patet. Cum enim sit duplex actio, sicut dicitur IX Metaphys., una quae manet
in agente, ut videre et intelligere, altera quae transit in rem exteriorem, ut calefacere et secare; utraque fit
secundum aliquam formam. Et sicut forma secundum quam provenit actio tendens in rem exteriorem, est
similitudo obiecti actionis, ut calor calefacientis est similitudo calefacti; similiter forma secundum quam
provenit actio manens in agente, est similitudo obiecti. Unde similitudo rei visibilis est secundum quam
visus videt; et similitudo rei intellectae, quae est species intelligibilis, est forma secundum quam intellectus
intelligit. Sed quia intellectus supra seipsum reflectitur, secundum eandem reflexionem intelligit et suum
intelligere, et speciem qua intelligit. Et sic species intellectiva secundario est id quod intelligitur. Sed id
quod intelligitur primo, est res cuius species intelligibilis est similitudo. Et hoc etiam patet ex antiquorum
opinione, qui ponebant simile simili cognosci. Ponebant enim quod anima per terram quae in ipsa erat,
cognosceret terram quae extra ipsam erat; et sic de aliis. Si ergo accipiamus speciem terrae loco terrae,
secundum doctrinam Aristotelis, qui dicit quod lapis non est in anima, sed species lapidis; sequetur quod
anima per species intelligibiles cognoscat res quae sunt extra animam (Aquinas 1952).
21

600

Philosophia (2013) 41:589602

except by being in act through the visible species. Hence the visible species are
not considered that which we see, but that by which we see.22 (P2)
I must admit that I was really puzzled while reading this passage, and, at the same
time, I was willing to agree with what Pasnau (1997, 212) says: How can we conclude
from such a premise that species arent themselves apprehended? But the problem is it
is not clear what Aquinas means here by precognized (praeintelligitur), so maybe it
would be more careful not to assume that precognition is a form of cognition. On the
other hand, neither is one allowed to conclude from such a premise that species are not in
any sense cognized. The most we are entitled to conclude is that species play a certain
causal role: that the presence of species is a prerequisite for certain later operations.
If precognition is correlated with what P1 states and with the fact that the intelligible
species are not the only ones we cognize,23 one might think we are justified in saying
that Aquinas has a representationalist position at the level of the intelligible species.
Despite Pasnaus very elegant and quite luring demonstration, I think it rests on
insufficient textual proofs and the rest of the argumentation is mainly based on circumstantial evidence. Allow me to explain the reasons behind my disagreement. As already
noted P2 stands alone and without any real reverberations, if it is not correlated with P1.
But let us take a closer look at what P2 really says. Everything seems to revolve around
praeintelligo whether we understand it as a kind of cognition, or we read it as a type of
preliminary data, emphasising the anteriority. If we assume the first reading, that
praeintelligo means the precognition of the intelligible species, then we have to shed
some light on the subject of this action, the active or the possible intellect. If the
intelligible species is precognized by the possible intellect, we would find ourselves
with a case of cognition within cognition, because, for Aquinas, the process of cognition
involves a change in the subject; if the possible intellect was to precognize the intelligible species, this would in turn trigger a change in the possible intellect resulting in
formation of another intermediary intentional form, and, in the end, we would find
ourselves in the paradoxical case of having not one, but two intelligible species which
stand in a formal identity relation to the external object. This addition would be useless,
therefore I think Aquinas is implying that the agent intellect, and not the possible
intellect precognizes the intelligible species. Then, why wouldnt the same consequence,
cognition within cognition, apply at the level of agent intellect as well? First and
foremost, because things are a bit different at this level: due to the fact that the agent
intellect acts before the actual cognzing of the quiddity taking place, we can name this
level a pre-cognitive24 one. Within this pre-cognitive framework the praeintelligitur of
Utrique autem harum operationum praeintelligitur species intelligibilis, qua fit intellectus possibilis in
actu; quia intellectus possibilis non operatur nisi secundum quod est in actu, sicut nec visus videt nisi per
hoc quod est factus in actu per speciem visibilem. Unde species visibilis non se habet ut quod videtur, sed ut
quo videtur (Aquinas 2000).
23
Si igitur ea quae intelligimus essent solum species quae sunt in anima, sequeretur quod scientiae omnes
non essent de rebus quae sunt extra animam, sed solum de speciebus intelligibilibus quae sunt in anima.
S.Th. (I, q. 85, a.2, co.) (Aquinas 1952).
24
By saying that the intelligible species operate at a pre-cognitive level, I am in fact stating that they are a
part of the process of thinking, which stands for the moving of the mind while it still deliberates. More
clearly, it acts as a sort of preparatio for the actual act of understanding, by providing the matter for the
possible intellect: acting at a pre-cognitive level means being a part of a process which unfolds before the
actual act of understanding. See also Pasnau (1997), 13940.
22

Philosophia (2013) 41:589602

601

the intelligible species can be understood not so much as a kind of cognition, but as a
kind of preparation (praeparatio) for the actual act of cognition. Three reasons stand
behind this option: (1) in compound adjectives the prefix prae is used to form words
which indicate that something takes place before a particular action, event or date; (2)
the root intelligo is used in this particular case because the whole action is located at the
level of intellectual cognition; (3) the only functions the agent intellect has are the
abstraction of the intelligible species from the phantasms and their illumination in
other words, the agent intellects only role is to actualize the intelligible character of the
sensorial data, so as to make it fit for the possible intellect. The preparation for the actual
process of cognition would not mean, in this case, a fore-cognition, but an actualization
of the intelligible character of the intelligible species needed for their proper receival by
the possible intellect. Still, why cannot praeintelligo be understood as a form of
cognition? The reason is very simple and it rests on textual evidence: Aquinas describes
the agent intellect as having only the two aforementioned operations. Against the
cognitive reading of praeintelligo I can appeal to the formal identity and the double
existence of a form. If my reading of Q.D.V. (q. 1, a. 11, co.) is correct, the same reading
should apply here too. Therefore, my conclusion is, once more, that praeintelligo and
apprehensio should be understood as having only a causal meaning. If praeintelligo is
understood as a preparation for cognition, P1 stands alone, casting Pasnaus ingenious
interpretation into the shadow.
After indicating the passages that support the act-object doctrine from both the
earlier and the later writings of Aquinas, Pasnaus demonstration reaches its final
step: it underlines the complementarity of the official view position with the actobject doctrine. According to the former, the species are not, ordinarily, the object of
cognition themselves, but they are the medium through which the extra-mental
objects are apprehended; according to the latter, the fact that the species are not
cognized translates into the claim that the species are not the objects of our judgements or beliefs. This translation is justified by appealing to passages from Quod.
(VII, q. 1, a. 1, co.) and S.c.G. (I.53), appendix II8 from the 1961 Marietti edition,
where the extra-mental objects are described as objects which we are first drawn to
and which we first attend to. We are first drawn to the extra-mental objects and not to
the species because we form concepts and judgements not about the species, but
about the objects. (Pasnau 1997, 216) The main body of the problem is not that we
produce concepts and judgements about the objects instead of the species, this is a
fairly uncontroversial statement easily accepted by any direct realist. The real problem is the cognitive understanding of expressions like turning toward, apprehending,
precognizing, in cases where they are nothing but alternative expressions of the
taking on of the species, as Scott McDonald suggested to Pasnau (1997, 208).
Let us now return to the leading question of this paper: what role do the intelligible
species play in the process of human cognition? Have I managed to gather enough
data for settling the dispute between the direct realist and the representationalist
interpretations of the intelligible species? As I have pointed out, there are more
arguments in favour of treating the intelligible species as causal entities, and not so
many arguments for treating them as cognitive entities. On the one hand, formal
identity, double existence, double aspect, id quo and similitudo blend in coherently
and are supported by numerous passages ranging from Aquinass early to his later
works. On the other hand, Panaccios attack on the formal identity lacks the resources

602

Philosophia (2013) 41:589602

to persuade a careful reader to go in the direction of his version of representationalism. And this happens for two reasons: first, he does not take into account the double
existence and the double aspect of a form, and second, he uses a very narrow
understanding of identity, namely numerical identity.
In the case of Pasnaus reading things are not equally obvious, and this happens
because he embraces the formal identity. By accepting the fact that the content of the
intelligible species is identical with the essence of the extra-mental object, Pasnau
advances an interpretation that is halfway between direct realism and representationalism. However, his entire demonstration rests on fragile grounds. Here I have in
mind the formal structure of his argumentation: A is affirmed in Aquinass early
works, A is not explicitly denied in Aquinass later works, A is compatible with the
official view of species, and therefore A can be maintained. But it is not enough to
say that A is not explicitly negated, in order to be fully justified in maintaining A. If
we rely only on the fact that A is not negated, there is the possibility for A to be
maintained, but there is also the possibility for A not to be maintained. If we zoom in,
things still do not seem to stand on a solid ground, because the cognitive understanding of verbs like apprehendo and praeintelligo contradicts the id quo and the double
existence status of the intelligible species. However, until someone finds passages in
Aquinass later works where the act-object doctrine is explicitly defended, the
validity of this view is still opened to debate. Thus, the most robust answer to the
question regarding the role of the intelligible species in the process of cognition
remains in the hands of direct realists, and, consequently, the intelligible species play
a causal role in the process of human cognition.

References
Aquinas, T. (1979). Compendium theologiae ad fratrem Reginaldum. In R. A. Verardo (Ed.), Opuscula
theologica (Vol. 1). Torino: Marietti.
Aquinas, T. (1929). In P. Mandonnet (Ed.), Scriptum super libros Sententiarum magistri Petri Lombardi
episcopi Parisiensis. Paris: P. Lethielleux.
Aquinas, T. (1952). In P. Caramello (Ed.), Summa theologiae. Torino: Marietti.
Aquinas, T. (1961). In C. Pera (Ed.), Summa contra Gentiles. Torino: Marietti.
Aquinas, T. (1965). In P. M. Pession (Ed.), Quaestiones disputatae de Potentia. Torino: Marietti.
Aquinas, Th. (19701976). Quaestiones disputatae de Veritate, in Opera Omnia, ed. Dondaine, H. F. XXII.
Roma & Paris: Commissio Leonina & Cerf.
Aquinas, Th. (1996). Quaestiones quodlibetales, in Opera Omnia, ed. Gauthier, R. A. XXV. Roma &
Paris: Commisio Leonina & Les ditions du Cerf.
Aquinas, Th. (2000). Quaestio disputata De spiritualibus creaturis, in Opera Omnia, ed. Cos, J. XXIV-2.
Roma & Paris: Commissio Leonina - Les ditions du Cerf.
Klima, G. (1996). Nulla virtus cognoscitiva circa proprium obiectum decipitur, Critical comments on
Robert Pasnau, The Identity of Knower and Known, http://faculty.fordham.edu/klima/APA.htm.
Accessed 19 July 2012.
Panaccio, C. (2000). Aquinas on Intellectual Representation, in Cahiers depistmologie. Groupe de
recherche en pistmologie compare, no. 2002.
Panaccio, C. (2010). Mental representation. In R. Pasnau & C. van Dyke (Eds.), The Cambridge History of
Medieval Philosophy (Vol. 1). Cambridge: Cambridge University Press.
Pasnau, R. (1997). Theories of Cognition in the Latter Middle Ages. Cambridge: Cambridge University Press.
Perler, D. (2000). Essentialism and direct realism: some late medieval perspectives. Topoi, 19(2), 111122.
Perler, D. (2002). Theorien der Intentionalitt im Mittelalter. Frankfurt am Main: Vittorio Klosterman.
Spruit, L. (1994). Species intelligibilis: From Perception to Knowledge (Vol. 1). Leiden: Brill.