c Peter King, forthcoming in The Oxford Handbook to Aquinas

AQUINAS ON THE EMOTIONS

QUINASS theory of the emotions (passiones animae) is cognitivist, somatic, and taxonomical: cognitivist because he holds
that cognition is essential to emotion; somatic because he holds
that their physiological manifestations are partially constitutive of emotions;
taxonomical because he holds that emotions fall into distinct natural kinds
which are hierarchically ordered. Aquinas supports these claims with a wealth
of argument, presented in his three extended treatments of emotion: sent.
d. q. , d. q. and d. q. , dating from ca. ; uer. qq. ,
dating from ca. , and his masterly sum. theol. a qq. (the
treatise on the emotions), dating from .1 His presentation and defense
proved so eective that later medival thinkers took Aquinass theory as their
starting-point, whether they agreed with it or not, and it remained a contender
in aective psychology up to the end of scholasticism.
Aquinas identies eleven essentially distinct types of emotion, which he
sorts into two kinds, and which for the most part occur in conjugate pairs of
contraries: the six concupiscible emotions of love and hate, desire and aversion, delight and distress; the ve irascible emotions of hope and despair,
condence and fear, and anger (which has no contrary).2 The concupiscible
emotions are directed at objects insofar as they appear to be good or evil,
whereas the irascible emotions are directed at objects insofar as they present
something good or evil that might be hard to achieve or to avoid. But these
are not perspicuous or evident claims; we need to begin with more fundamental issues about emotion in general to appreciate Aquinass theory.
Ill begin with Aquinass account of the nature of emotion as a psychological phenomenon (), as a feature of what he calls sensitive appetite,
after which Ill turn to how emotion is related to other psychological faculties,

1
2

All translations are my own. Latin texts are cited from their respective editions,
with the punctuation as given (not always respected in the translations).
I adopt the dating of these works given in Torrell [].
Respectively amor and odium, desiderium and fuga, gaudium and tristitia; spes and
desperatio, audacia and timor, and ira. Aquinas adopts this list and much of the
structure that supports it from Jean de la Rochelle (Summa de anima), by way of his
mentor Albert the Great: see Knuutilla [] and King [].
1

THE NATURE OF EMOTION

namely cognition and volition (). After situating the emotions in Aquinass
psychological system, Ill then look at his analysis of their internal structure,
which provides a taxonomy of emotional experience (). Ill conclude with
some brief reections on Aquinass theory vis-a-vis contemporary theories of
`
emotion.
. THE NATURE OF EMOTION
Emotion, according to Aquinas, is an objectual non-volitional aective
psychological state. Or, in medival terms, emotion is an actualization of
the sensitive appetite, which is a semi-autonomous faculty of the soul. To see
what these denitions mean and why they are equivalent, more than a little
unpacking is needed.
Psychology, in the aristotelian tradition, is a subordinate branch of natural philosophy. It studies the activity of living beings qua living, and on
this score identies three kinds of clustered activities that living beings exemplify, stemming from three distinct principles, that is, from three types
of soul: (a) nutrition, growth, and reproduction, typical of plants and trees,
whose principle is the vegetative soul; (b) self-movement and perception of
the world, typical of animals, whose principle is the sensitive soul; (c ) thought
and reasoning, typical of human beings, whose principle is the intellective
soul. These kinds of soul are arranged in a hierarchy such that the latter include the former: anything capable of (b) is capable of (a), anything capable
of (c ) is capable of (b) and (a). Aquinas famously held that these clusters of
principles were not really distinct when combined in the same subject his
controversial stand on the unicity, as opposed to the plurality, of substantial
form but this metaphysical disagreement can be set aside in psychology,
since all parties to the dispute agree that human beings, for instance, have
intellective and sensitive capacities, whether they stem from a single unique
principle or a pair of related principles; horses and cats do not have the same
cognitive powers as humans, however these powers be related when found
together. One question to ask, then, is whether the emotions are features of
the sensitive soul (and so common to all animals) or of the intellective soul
(and so particular to human beings).
Yet there is another question that is equally pressing. The cluster of powers associated with the sensitive and the intellective souls are not limited to
cognition. Animals and human beings do not merely acquire and process
information about the world; they engage the world directly, being drawn to
some things and driven away by others. Therefore, side-by-side with cognitive powers to acquire and assimilate information, there are appetitive powers
to move the subject. Another question to ask, then, is whether the emotions
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE NATURE OF EMOTION

are cognitive (and so dealing with information about things) or appetitive (and
so dealing with the things themselves).
Aquinas starts his answer to both questions by considering the nature of
emotion (uer. q. a. and sum. theol. a q. a. ). For an emotion is a
passio animae, literally something that the soul undergoes or experiences
a capacity for being in a given psychological state rather than something
the soul does (the way it reasons, for instance). In medival philosophical
jargon, an emotion is a potency whose principle of actualization is external to
its subject; in contemporary terms, an emotion is a reaction.
First, if an emotion is a reaction, it is therefore passive as regards whatever
brings it about, that is, whatever prompts the reaction. Yet the passivity of an
emotion in itself does not entail that the subject is thereby passive with respect
to that emotion. Sight is likewise a passive potency we can see only what is
there to be seen but we can exercise a measure of control over what we see
nonetheless: we may close our eyes, avert our gaze, turn our head, and so on.
So too we have some degree of control over the emotions. (More will be said
about this point in .) Moreover, an emotion is a reaction that may well have
causal ecacy: fear of the wolf moves the sheep to ee, a perceived insult
causes the proud man to lash out in anger, the hope of winning motivates
the runner to put on a nal burst of speed at the end of the race. Reactions
can cause or motivate subjects to act; their doing so depends on how they
are related to other elements in the subjects psychology. That is, being in a
given state can be the cause of further events, regardless of how the subject
comes to be in that state. Hence the intrinsic passivity of emotion is of no real
importance.
Second, while emotions may be reactions, they are more fundamentally
types of motion.3 The subject of an emotion is moved by it drawn towards
the object, as in the case of desire; or driven from it, as in the case of hatred.
This is more than mere metaphor; in the aristotelian tradition, movement
includes more than just change of place. For in living animals the soul plays
two roles: on the one hand, it is the substantial form of the body which vivies the body and unies the composite; on the other hand, it exercises its
operations through the body which it informs, causing it to change (move)
from one state to another. Emotions, Aquinas maintains, are psychophysical
phenomena: the apprehension of a insult leads to (a) the desire for revenge,
which is the formal aspect of anger, and (b) the boiling of blood around the
heart or as we should say the increase in heart rate, blood circulation, the

See King [], Miner [] Ch. ., and the discussion in .

c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE NATURE OF EMOTION

ow of adrenaline, and so on which is the material aspect of anger.4 The

desire (a) is the formal cause of the physiologial changes described in (b), set
in motion by the soul.5 By the same token the physiological changes in (b)
are what make the associated psychological state an emotion: a desire is only
a desire if its motivational force is felt in the proper way, and mutatis mutandis
for the other emotions.6
These considerations about the nature of emotion suggest a single reply
to the two questions posed above, namely that the emotions properly belong
to the sensitive appetite. Aquinas puts this reply concisely in uer. q. a. :
Since the body is altered due to an operation of the soul in the case of
emotion, it must be the sort of power that is joined to a bodily organ
and whose function it is to alter the body. Hence emotion is not in the
intellective part [of the soul], which does not pertain to the actualization of
some bodily organ. Nor again is it in sensitive apprehension, since from
sense-apprehension there follows movement in the body only through
the mediation of the appetitive power, which is the immediate moving
cause.
Emotions are sensitive rather than intellective because they essentially involve physiological changes, unlike the operations of the intellective faculties
of intellect and will (sum. theol. a q. a. );7 they are appetitive rather
than cognitive because they move the body, that is, because they engender changes in the subjects bodily states even the psychological movement
involved in emotion is typical of the appetitive rather than the cognitive powers, since the appetitive powers engage the subject directly with things in the
world (sum. theol. a q. a. ).8 More generally, Aquinas argues that emotions motivate action, as they unquestionably do, only if representations of
4

See sum. theol. a q. a. : In the case of the emotions, the formal aspect is
the motion of the appetitive potency and the material element is a physiological
change, where the one is proportionate to the other.
Aquinas takes this direction of inuence from the soul to the body as a mark of
emotion, which he calls by the name of animal emotion (passio animalis) in uer.
q. a. .
A corollary of Aquinass insistence that physiological changes are essential to emotion is that nonphysical beings, such as angels and God, strictly speaking do not
have emotions. See Miner [] Ch. . and the references therein, as well as
King [].
Aquinas oers two further arguments for the view that emotions do not belong
to the will: the will is a free active power whereas emotions are not, and the will
tends to a universal object whereas the emotions tend to a particular object (uer.
q. a. ).
This is Aquinass preferred proof when he has to give a brief account of why the
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

their objects occur in a context in which they move the agent (as in the appetite) rather than one in which such representations are merely assessed for
the information they convey (as in cognition). Thus the emotions must belong
to the appetitive part of the soul.
Hence there are four distinct types of psychological activity: (i) sensitive
cognition a. k. a. perception, the domain of the external and internal senses;
(ii) sensitive appetite, the domain of the emotions; (iii) intellective cognition,
the domain of thought and reasoning; (iv) intellective appetite, the domain
of free will. The rst pair are common to all animals and their exercise is
bound up with the body; the latter pair is specic to human beings and their
exercise is carried out independently of the body. In particular, as features of
the sensitive soul, the emotions are common to animals and to human beings.
Aquinass account of emotion must therefore be general enough to apply to
non-human animals as well as to human beings.
A nal point before turning to examine how the emotions are related to
the other types of psychological activity listed above (the concern of ). For
Aquinas, emotion has both a formal (psychological) and a material (somatic)
component, as we have seen; hence an emotion is primarily a state of the
entire unied soul-body composite.9 It is the bereft mother who grieves for
her lost son, not merely or primarily her soul that grieves, or for that matter
her body. Grief is not a mere mental phenomenon, though it of course does
involve mental states. As we would put Aquinass point today, emotions are
states of persons.
. COGNITION , VOLITION , AND EMOTION
The preceding section glossed over the details of how Aquinas takes emotion to be related to other psychological faculties, in particular to cognition
(both sensitive and intellective), and volition. With regard to cognition, there
are two topics to be considered: whether cognition is an essential part of an
emotion, and the extent to which cognition can inuence or control emotion.
The issue whether cognition might be essential to emotion arises from
reecting on Aquinass conclusion that an emotion is a power of the sensitive
appetite, and hence categorically distinct from any cognitive act a consideration that might seem to warrant the inference that Aquinas is a feelingtheorist about the emotions, holding the position that emotions are essentially
(pure) feelings which are known entirely through their phenomenological and
emotions are appetitive: see for example diu. nom. . or in eth. . .
9

See Cates [] Ch. .

c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

qualitative properties.10
Now this inference might be warranted if Aquinas were to hold that the
intentional aspect of emotion could be separated (at least by divine power)
from the actualization of the sensitive appetite, which would thereby lack
any cognitive element. This is not his position, however. Aquinas holds that
the sensitive appetite inherits its intentional character from cognition, which
must therefore gure in the account of emotion. Consider, for example, the
case of fear the emotion consequent upon the perception of some thing
as a (sensible) imminent evil that is hard to avoid, which is associated with
heightened respiration and heart-rate, the sudden ow of adrenaline, and the
like, where fright is the cause of ight. This is not specically human; a sheep
experiences fear when confronted with a wolf, in the well-worn example. Analyzing the stages of the process sketched here will make Aquinass position
clear.
First, the sheep has a cognitive act, more specically an act of sensitive
cognition: it sees the approaching wolf. This act, properly speaking, is an
actualization of the sheeps passive power of vision. As such, like the actualization of any passive power, it takes place due to an external principle, in this
instance the wolf. The wolf, as the external principle of the sheeps cognitive
act, thereby becomes its object; the sheep has a seeing-of-a-wolf, after all. The
point holds generally: the external principles that reduce passive potencies to
acts are their objects.
Sensitive cognition is the paradigm case, but Aquinas does note that the
same analysis can be applied when the object is derived not from occurrent
sensing but rather from memory via the imagination. The cognitive act alone,
however, whether derived from sense or from memory, is not sucient for
an emotion; a camera linked to image-analysis software could just as well
register the (sudden) presence of a wolf without feeling a thing. In addition to
the simple cognitive act of seeing a wolf, an act of another type takes place,
one that links the cognitive to the appetitive faculties. Aquinas describes it as
follows (sum. theol. a q. a. ):11
An animal must seek out or avoid some things not merely because they
are suitable or unsuitable to the senses, but according to certain other
uses and advantages or disadvantages. For example, a sheep seeing an
approaching wolf runs away not due to its unsightly colour or shape, but
as if it were a natural enemy. Likewise, a bird collects straw not because
10
11

See Floyd [], Gorevan [], and Ryan [].

See also sent. d. q. a. , where much the same account is given. Miner
[] Ch. . discusses the estimative and memorative powers in Aquinas with
regard to the emotions.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

it is pleasing to sense but because it is useful for building a nest. Thus it

is necessary for an animal to perceive intentiones of this sort, which the
exterior senses do not perceive. There must be some distinct principle. . .
the proper sense and the common sense are appointed for the reception
of sensible forms, but the estimative power (uis aestimatiua) is appointed
for apprehending intentiones, which are not received through sensing.
The object of the animals cognitive act, be it the wolf for the sheep or the
straw for the bird, is also perceived under some intentio or other: the sheep
perceives the wolf as a natural enemy, the bird perceives the straw as useful
in the construction of a nest.12 These are not perceptible properties. The
wolfs dangerousness is not perceived the way its unsightly colour is. Yet
the emphasis Aquinas puts on intentiones not being perceptible properties
to the extent of postulating a distinct psychological faculty to perceive them
at all is misleading. After all, no dispositional property is strictly speaking
perceptible: rationality, ferociousness, fragility, and the like are not apparent
to the exterior senses alone. (Neither is the wolfhood of the wolf, or in general the substance of anything.) No wonder that friendliness or hatred are not
perceptible (the examples Aquinas mentions in sent. d. q. a. ). Yet
far more important for Aquinass purposes is his recognition that intentiones
are what we should call today evaluative response-dependent concepts. The
sheep regards the wolf as a threat, and the bird regards straw as useful, because of the kind of thing each is. The sheep does not regard straw as useful,
since it does not build nests of straw; the bird does not regard the wolf as a
threat (or as much of one), since it can easily y away. At bottom this is a
built-in capacity: animals perceive these sorts of intentiones only through a
kind of natural instinct (ibid.),13 though it may be modied through conditioning and habituation, as when the family watchdog learns that strangers are
the only proper targets of its anger, or, in the case of human beings, through
the exercise of higher faculties (as we shall see shortly). For now let us take
the point that such evaluative response-dependent concepts occur after acts
of simple cognition.
Aquinas oers a few examples of intentiones: the sheep perceives the wolf
as inimical, the bird perceives the straw as useful, or vaguely advantages and
disadvantages. Now it might be thought that these are a motley collection as
12

13

It is unclear whether this cognitive act is distinct from the initial perception of the
wolf perhaps the sheeps concept of a wolf has dangerousness built into it but
it is enough for our purposes that it is logically distinct.
The natural instinct may be nothing more complicated than the way the animal is
wetwired: whenever a wolf-form occurs in sensitive cognition, the adrenal glands
begin pumping, heartrate increases, and so on.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

varied as the nature of evaluation itself (evaluation being the root meaning
of aestimatio). But there is a general way to describe intentiones as such, that is,
a way to characterize anything that is to count as an intentio at all, much the
way that for anything to count as visible it must have colour.14 It is sensible good
or evil, where sensible refers to the subjects perception of its object as good
or as evil.15 Hence an intentio must present some thing to its subject as a good
or as an evil in some fashion. When the sheep perceives the approaching wolf,
it may evaluate the wolf as a danger that is, as an imminent evil which
is hard to avoid.16 The initial simple cognition of the wolf is augmented with
this evaluative response-dependent cognition, each cognition arguably caused
by the wolf, or more exactly by the wolf given the natures of the wolf and the
sheep.
So much for the cognitive side of things. At this point there is a hand-o
to the sensitive appetite: the lower appetitive power does not naturally tend
to anything until after that thing has been presented to it under the aspect
of its proper object (uer. q. a. ad ), since in the case of animals the
sensitive appetite is apt to be moved by the estimative power, as when a sheep
esteems a wolf as inimical and is then afraid (sum. theol. a q. a. ). The
sensitive appetite, as a passive power, is reduced from potency to act when it
inherits objectual content from the evaluative response-dependent concept
(which is the actualization of the estimative power). That is to say, the sheep
has an act of the sensitive appetite directed at the wolf, which is presented to
the sensitive appetite as a hard-to-avoid imminent evil. This proper object
therefore has a double causal role. On the one hand, it reduces the sensitive
appetite from potency to act, and is thereby an ecient cause of the resulting
act. On the other hand, it makes the resulting act be the kind of act it is, and
is thereby its formal cause. For the resulting act of the sensitive appetite is
the emotion of fear when it is caused by the formal object the wolf as a hardto-avoid imminent evil, with the appropriate associated somatic responses. If
an object were presented as a hard-to-attain imminent good, say, the way the
sheep might appear to the wolf, the act of the wolfs sensitive appetite would
be hope. Emotions are therefore objectual, since the sensitive appetite is the
14

15

16

This is the medival notion of the proper object or the primary object the
terminology was uid of a cognitive potency, in this case of the estimative power.
The terminology is not perspicuous, since the intentio, which must be a form of
sensible good or evil, is, as Aquinas has remarked, not a perceptible feature of the
object. The key point to keep in mind is that sensible here refers to the particular
that is grasped in its presentation as a good or an evil.
As we shall see in more detail in , the formal object of the irascible emotions is
sensible good or evil that is dicult.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

passive recipient of the causal and formal agency of the external principle,
the wolf in the case of the fearful sheep.
Aquinas is therefore a cognitivist about emotion, since cognitive acts are
not only causal preconditions of emotion, but contribute their formal causes
as well. The emotion is not the feeling alone: it literally would not be the
emotion it is without the formal object it has, and there would be no emotion
at all in the absence of a formal object. This is not to say that there cannot be
objectless states of the sort that are so important to contemporary philosophy, such as angst, dread, or boredom, but that they are not to be understood
as emotions: they are rather akin to moods, somatic states that inuence psychological states.17 Despite being a cognitivist, however, Aquinas is also an
externalist. For the pair of cognitive acts do not in themselves have motivational force: they act as ecient and nal causes of the acts of the sensitive
appetite, which do motivate the agent, but even the evalative judgment implicit in the response-dependent conception of something as a good or an evil
does not cause responsive action directly.
The point is important, because the causal link between cognition and
appetite is more complicated in the case of human beings. Even in the case of
animals such as sheep, the natural tendency or instinct to respond emotionally in a certain way can be tempered and perhaps even changed by conditioning and habituation. The extent to which this is possible depends on the
type of animal, and perhaps even on the particular animal. Kittens and puppies raised together often remain quite friendly with one another as cats and
dogs, though they respond in more traditional ways to unknown dogs and
cats; sheep, on the other hand, might never learn to be tolerant of wolves, no
matter how tame. The strength of the causal linkage in each animal, and its
susceptibility to conditioning and to habituation, make all the dierence, and
these are matters for empirical investigation.
Human beings have higher cognitive faculties, and, in consequence, a
more complex and sophisticated emotional life. For one thing, instead of an
estimative power humans have a general cogitative power, which is sometimes called particular reason on the grounds that it combines individual or
17

Moods were often thought to be completely explained in purely physiological

terms, having to do with the relative balance among the four bodily humours.
Acts of the vegetative soul in the higher animals, such as sexual arousal or hunger,
were also thought to be objectless and hence not emotions a view in keeping with
their contemporary classication as drives or urges. Aquinas points out that male
impotence eectively demonstrates the dierence between mere bodily arousal
and sexual desire, a form of the emotion of desire (concupiscentia) and an altogether
dierent thing.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

particular intentiones (sum. theol. a q. a. and q. a. ). Aquinas says little about human cogitative power, and nothing about what it is to combine
intentiones. Nor do the details really matter. What does matter to Aquinas
is that thinking and reasoning aect the evaluative response-dependent concepts that trigger the sensitive appetite, and that human cogitative power is
involved in the process.
We can now turn to the second of the two questions posed at the start of
this section, namely the extent to which cognition can inuence and control
emotions.18 We have already seen that some non-human animals are susceptible to conditioning. And, like other animals, humans have some instinctual
emotional responses (fear of falling) and some habituated responses (pleasure
at the sight of a loved one). But human beings, unlike animals, have an extensive and rich set of conceptual resources that can be deployed even at the
level of mere conditioning: the botanists instant delight at recognizing an unknown species; the anger that follows upon a perceived slight in a complex
but thoroughly assimilated code of honour; and so on.19 Nor is it merely a
matter of human beings having a larger conceptual apparatus; human beings
are much less tied to their present circumstances, being better able to imagine
things in other places and at other times, and in addition are able to conceptualize the world in a universal, rather than merely a particular, way. They
can hope for a happier afterlife, become angry at the memory of a rebuke,
love wisdom, hate spiders, be saddened at the loss of the sculptures of Phidias
all beyond the capacity of animals.
Aquinas mentions two ways in which human emotion is cognitively penetrable, that is, capable of being consciously aected by changes in belief or
thought after the quasi-instinctual initial response of the sensitive appetite.20
For although an emotional response is not completely in our power since it
precedes the judgment of reason, it is in our power to some extent (uer. q.
a. ). First, we can imaginatively present one and the same thing in dierent
lights, via the imagination, and thereby trigger dierent emotional responses
(uer. q. a. ). The divorced spouse can think of the former partner with
18
19

20

See Murphy [] and Uenheimer-Lippens [].

Aquinas notes that the irascible emotions, in virtue of having a more complex
formal object, are thereby closer to reason and more susceptible to its inuence.
He takes this point from Aristotle, eth. nic. . (a b : see for instance sent.
d. q. a. and uer. q. a. .
Of course, humans can condition themselves over time through a variety of techniques: training, education, brainwashing, self-hypnosis, repetitive practice, and
so on. But here we are concerned with Aquinass focus on the ways in which
someone can attempt to aect his or her emotional responses directly.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

COGNITION , VOLITION , AND EMOTION

love or hatred, depending on which past situations and events are recalled
or imagined. Likewise, the imagination can provoke emotional responses by
the force of what it imagines. Aquinas oers the example of a believer who
reects on punishment in the afterlife; imagining the re burning and the
worm gnawing and the like, there follows the emotion of fear in his sensitive
appetite (uer. q. a. ad ).21 Since deliberate imagination of this sort is
in the agents conscious control, it is clear that some emotions are indirectly
subject to the control of reason though it is a bit like controlling digestion
by being able to pick and choose what one eats.
Second, Aquinas notes that the intellect can inuence emotion: anyone can experience for himself that by applying some universal considerations, anger or fear or the like can then be mitigated or even stirred up
(sum. theol. a q. a. ). Reminding oneself of general truths can aect the
understanding of a particular situation. Grief over the death of a friend can
be mitigated by thinking of the general truth that we all die; condence can
increased by the thought that only the brave deserve the fair; and so on. Here
Aquinas is somewhat hamstrung by his view that the intellect is the realm of
the universal whereas sense is the realm of the particular; if we allow him
to relax his strict insistence on this dividing principle, then there are all sorts
of ways in which intellectual cognition can (attempt to) inuence ones emotions: thinking about the stringent air-safety regulations in place in order to
curb ones fear of ying, for example, or thinking about how even lesser lights
have been awarded the Nobel Prize in order to boost ones hopes. The factor
in common in all these cases is that the emotional responses seem to follow
(when they follow at all) merely upon having the thoughts.
Such techniques do not always lead to success. For the emotions do not
always submit to the dictates of reason or imagination; they are unruly and
may resist their commands (sum. theol. a q. a. ).22 Yet strictly speaking,
it is not the role of reason to command at all. That is the province not of
cognition, but of the intellective appetite, i. e. the will.
Aquinas argues that emotion is the province of the sensitive rather than
the intellective appetite on the grounds that the latter, like the intellect itself,
has a purely nonphysical operation (uer. q. a. and sum. theol. a q.
a. ). Yet just as in the case of the cognition, the presence of higher intellective
faculties allows human beings to inuence their emotions in ways that are not
open to animals, though not to dictate them; we cannot simply choose not to
21
22

See Galeazzi [].

See also sum. theol. a q. a. ad . The allusion is to Aristotle, pol. . (b ),
in which the rational part of the soul is said to control the irrational part with a
political and royal rule rather than a tyrannical rule.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

have an emotional response, though we can have some eect on what that
response might be. Aquinas describes three ways in which this can happen.
First, the sensitive appetite is subordinate to the will, and this subordination aects the kinds of emotions that accompany volition (uer. q. a. ):23
In the case of powers that are connected and ordered to one another, it
happens that an intense movement in one of them (and especially in the
higher one) overows into the other. Accordingly, when the movement
of the will is directed to something through choice, even the [emotions]
follow this movement of the will.
If someone makes a choice that wholeheartedly commits him to a course of
action, say, he may thereupon feel delight and hope in its pursuit. Such an
overow from the will is an emotion whose inception is in the very choice
of the will. In general, the process of habituating emotional responses to
choice is part of acquiring practical wisdom, which includes satisfaction with
the choices one has made.24
Second, the will is not only the principle of choice; it also governs consent. We may consent to our emotions, or withhold consent from them,
thereby strengthening or weakening the emotional response. (The latter is
usually described as resisting the emotion.) In the ordinary course of events
an emotion follows upon sensitive cognition, as described above, and so precedes the judgment of reason in such a way that there is no issue of consent,
whether explicit or implicit (uer. q. a. ad ). Once the emotional response
occurs, the will may then endorse or reject it.
Third, the will can directly aect what Aquinas calls the motive or executive power, so as to prevent or facilitate the emotion from being causally
ecacious (uer. q. a. ). When a human being perceives the sudden approach of a wolf, he experiences fear as a natural reaction, but unlike the
sheep he may exercise his free will and choose to stand his ground in the face
of his fear, blocking his natural impulse to ee. He would have no success in
the face of his more immediate somatic responses, however: he may tremble, grow pale, break into a sweat; his teeth may chatter, his knees knock, his
heart race (the physiological expressions of fear which Aquinas catalogues in
sum. theol. a q. ).
. THE TAXONOMY OF EMOTION
Aquinas adopts the traditional distinction of emotions into two fundamental types, namely concupiscible and irascible, and sets out to establish
23
24

See also sum. theol. a q. a. ad .

Aquinas takes this point from Aristotle, eth. nic. . (b ).
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

the distinction on a rm philosophical basis.25 He sketches three arguments

that the concupiscible emotions and the irascible emotions genuinely dier
and are not reducible to a single principle:
The Interference Argument. The two types of emotions must be dierent in
kind because they are able to interfere with one another: stirring up anger
can lessen desire, and conversely stirring up desire can lessen anger.
The Submission Argument. Sometimes the soul submits to distress, against
the inclination of desire, so that it may ght against things opposed to it.
The Champion Argument. The irascible emotions arise from the concupiscible emotions, to help them succeed, whereupon they terminate in them.
Anger, for example, may be born from distress and, in taking revenge,
thereby relieve the distress and end in delight.
A few comments about each are in order.
The Interference Argument, which is ultimately derived from Platos Republic, turns on the fact that the distinct kinds of emotions can be directed at
one and the same real thing while nevertheless diering in their formal (intensional) objects. One and the same person can be simultaneously alluring and
annoying, for example; these contrary characterizations of the same person
conict with one another. Hence the emotions must be dierent in kind, not
merely in degree. For while it is true that we speak of the relative strength of
the concupiscible and the irascible emotions emotions, as when we say that
anger nally won out over desire, this is not to be understood literally. It
is quite dierent, Aquinas maintains, from the case in which two emotions of
the same sort compete with one another when I am presented with a choice
between chocolate ice cream and vanilla ice cream, for instance, where I have
only to consult the relative strength of each desire to make the choice. But
when the emotions are of dierent kinds, their strength does not accurately
predict the agents behaviour. Even a tempting bone will not deect a dog,
once angered, from attacking someone.
The Submission Argument takes a dierent tack. Sometimes the sensible
good that is hard to attain is such that the diculty is an intrinsic feature of
the good in question: we want to win the race, become uent in Icelandic,
learn quantum physics. But sometimes the diculty is in the surrounding
circumstances, not inherent in the object itself. It is no challenge to make
tea, unless one is stranded on a desert island with tea but no kettle. In cases
such as the latter, Aquinas declares, the end can be unproblematically desired, independent of the diculties associated with it; drinking tea is not an
25

The following discussion is derived from sent. d. q. a. ; uer. q. a. ; and

sum. theol. a q. a. .
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

intrinsically dicult or challenging enterprise, even if it is so in certain circumstances. In such cases, though, where the diculties are not intrinsic to
the good sought, one must submit to them to reach the desired goal. The
implications of the Submission Argument should be clear. Aquinas holds that
we cannot understand all behaviour in terms of simple push/pull desires, in
particular instances of submission to present pain, which involve not merely
weighing the relative strength of the desires but at least rudimentary meansends calculation.
The Champion Argument turns on the fact that the concupiscible emotions are comprehensible in their own terms, whereas the irascible emotions make sense only against the background of the concupiscible emotions.
Aquinas presents this as partly a logical claim and partly a causal claim. Overcoming the diculties means attaining the sensible good, which prompts the
emotional response of delight; so much is simple logic. But other connections
among the emotions, such as distress, anger, and revenge, are causal rather
than logical in nature. Aquinass point is that in such causal connections,
the irascible emotions come to the aid of the concupiscible emotions, the
former being champions of the latter. They can do so in virtue of being
a dierent kind of emotion, for otherwise they would be a constitutive part
of the initial (concupiscible) emotional response to the object, not something
further.
Having established by these three arguments that there are two basic
kinds of emotions, Aquinas then turns to isolating and specifying the distinct varieties of each. Before we turn to the details of his account, however,
it is worth remarking on his method. Although Aquinas will be oering a
taxonomic theory of emotion, identifying distinct kinds of emotions and their
interrelations, his procedure is quite unlike the standard sorts of taxonomies
found in ordinary genus/species accounts, or even the less rigid taxonomies
found in botany and biology. The fundamental divide between the concupiscible and the irascible emotions is not due to a dientiating feature, but to a
distinction of their formal objects: sensible good/evil taken absolutely and sensible
good/evil that is dicult. (If anything, the latter looks like a candidate for a
species of the former.) Once we accept this distinction, however, we might
think that we could identify the subordinate kinds of each type by proper differentiae presumably dierentiating the formal objects of each, the irascible
emotions distinguished by dierent kinds of diculty, for instance. But this is
not how Aquinas proceeds. Instead, he uses some technical apparatus from
Aristotles natural philosophy to generate a set of principles so as to arrange
the six concupiscible and the ve irascible emotions into rational groups of

c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

coordinate pairs,26 on the grounds that emotions are, literally, motions of

the sensitive appetite.
In brief, his account is as follows. Physics teaches us that there are two
types of motion: (a) movement either towards or away from a given terminus, in which case it is the direction of the movements that are contrarily
opposed; (b) movement between two contrary opposed termini. Concupiscible emotions are directed at their object solely as (sensible) good or evil. If the
terminus in question is the good, then, Aquinas maintains that the sensitive
appetite cannot fail to move towards it; if evil, it cannot fail to move away
from it. Hence (a) does not apply. The concupiscible emotions are therefore properly understood as types of movement between contrary termini,
that is, of type (b), between the contrary opposites of good and evil: love,
desire, and delight towards the good; hatred, aversion, and distress towards
evil. The grouping of concupiscible emotions into three conjugate pairs has
to do with the avor of the movement each exemplies: love and hate are
simple tendencies of the sensitive appetite with respect to their objects; they
are emotional attitudes toward the object, pure and simple an aective
stance to it. The second conjugate pair, desire and aversion, involve moving
towards or away from their objects, being drawn to or driven from it. The
third conjugate pair, delight and distress, has to do with the aective stance
one is in when attaining the object.
Irascible emotions, on the other hand, are directed at (sensible) good and
evil not in themselves but qua dicult, and so are capable of both directions
of movement implicit in (a) as well as movement between contrary termini as
in (b). They are properly categorized according to (a), however. Aquinas insists that the irascible emotions do not characterize approach and withdrawal
in terms of good or evil but in terms of the surmountability or insurmountability of the diculties associated with the (good or evil) object. Hope is the
irascible emotion that sees its object as a surmountable (attainable) dicult
good, so that the dicult good approaches the agents possession; despair,
with which hope is paired, sees its object as an insurmountable (unattainable)
dicult good, so that the dicult good withdraws from the agents possession. Likewise, condence is the irascible emotion that regards its object as
a surmountable (avoidable) dicult evil, and fear, with which condence is
paired, regards its object as an insurmountable (unavoidable) dicult evil. Fi26

There seems to have been some development in Aquinass conception of how the
emotions are structured and organized. The account given here mostly follows the
mature analysis given in sum. theol. a q. a. ). This is Aquinass major advance over Jean de la Rochelle, who had organized emotions into their conjugate
pairs, but who did not oer any principles underlying their organization.
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

nally, there is the irascible emotion par excellance, anger, which is unique in
not having a contrary; here the sensitive appetite has achieved the evil that
was dicult to avoid and rests in it, taking in the full measure of its diculty.
Each conjugate pair represents a avour of motion, as with the corresponding concupiscible emotion: the simple tendency with respect to the object,
surmountable obstacle to something good (hope) or insurmountable escape
from something evil (despair); movement with respect to the possession of
the object, either towards an attainable good (condence) or away from an
unavoidable evil (fear); and nally anger, which dwells on a present evil in its
possession that is hard to overcome, and so is a form of rest.
As noted, irascible emotions also involve movement of type (b), namely
between contrary opposites, although in a more complicated way than the
concupiscible emotions. Moreover, the irascible emotions are grouped somewhat dierently in respect of (b). Hope and fear are paired together in that
they each regard their (dicult) objects as likely to be possessed by the agent,
hope directed at something good and fear directed at something evil. The
same reasoning presumably applies to condence and despair, each regarding its (dicult) object as likely to not be possessed by the agent, condence
at the prospect of being without something evil and despair at the prospect
of being without something good. Anger, as before, obeys a slightly dierent
logic, since it has for its object a present evil that is already possessed and
dicult to overcome; there is no contrary irascible emotion of overcoming
directed at the possession of a present good; indeed the only proper response
to the possession of a present good is delight, a concupscible emotion.27
Aquinas thus takes the overall taxonomic structure of the eleven kinds of
emotion to be as follows:
Concupiscible Passions
Love Hate
Desire Aversion
Joy Sorrow

[simple tendency]
[movement]
[repose]

Irascible Passions
Hope Despair
Condence Fear
Anger

Given the complexity of the underlying division (not all of which is represented here), it is not clear how to extend the taxonomy to bring into its
scope further species of each kind of emotion. In the event, Aquinas does not
do this, preferring instead to discuss each emotion one-by-one and to describe
the subtypes and varieties of each, as required.
27

This last point illustrates a logical and causal truth for Aquinas, namely that the
irascible emotions begin from and nally terminate in the concupiscible emotions
(sum. theol. a q. a. ).
c Peter King, forthcoming in The Oxford Handbook to Aquinas

THE TAXONOMY OF EMOTION

Two examples will give the avor of Aquinass discussion of the subspecies of emotions, each the paradigm of its type: love for the concupiscible
emotions, anger for the irascible emotions.
In sum. theol. a q. a. , Aquinas puts forward his basic division of
love into friendly (amor amicitiae) and covetous (amor concupiscentiae). This
division follows upon his observation that love, strictly speaking, has two
targets: the item, seen as a good thing, and the subject who recieves the
item. Covetous love is love that gives precedence to the good thing, which
is typically wanted for oneself; friendly love is love that gives precedence to
the recipient, typically someone other than oneself, for whom the good thing
is wanted. The two kinds of love are not entirely on a par. The object of
friendly love is loved simply and per se, whereas the object of covetous love
is not loved for itself but for something else. Friendly love is also known as
benevolence (in q. a. ) because of this concern for the other for its own
sake, which gives it priority over covetous love. Friendly love is even found
among animals, the best instance being the friendly love a mother has for her
ospring. The two kinds of love, then, seem quite close to being species of
the genus love in the traditional sense.
In sum. theol. a q. a. , Aquinas turns to the types of anger (given
traditional expression by John of Damascus): wrath, ill-will, and rancour. He
proposes that each increases anger: wrath, which might better be termed irascibility (as Aquinas remarks in ad ), denotes the facility of the movement to
anger; ill-will the rehearsal of what caused the anger; and rancor the vindictiveness or unquenchable impulse for revenge. But none of these is in any
clear sense a division of the formal object of anger, namely the desire to
overcome a dicult present evil that poses an obstacle. Aquinas seems to
recognize the justice of this point, since in his answer to the rst objection,
which charged that Damascenes enumeration was an accidental division, he
declares that those things that help to complete anger in some fashion are
not altogether accidental to it; as a result, nothing prevents them from providing it with a specic dierentia (ad ). Perhaps not, but that is hardly
the same as saying that they do provide specic dierentiae, which Aquinas
carefully does not say. Nor is there any discussion of what we might think
of as better candidates for species of anger: annoyance, irritation, rage, and
the like. Aquinas seems interested only in nding some way to accommodate
tradition, rather than exploring the question in its own right. Later commentators such as Cajetan took note of Aquinass understated conclusion; in his
commentary ad loc., Cajetan concluded that wrath, ill-will, and rancor are not
genuine species of anger for Aquinas.
The upshot is that Aquinas leaves us with a careful taxonomy of the
c Peter King, forthcoming in The Oxford Handbook to Aquinas

CONCLUSION

emotions at the generic level, along with a variety of observations about

some of their species, with no attempt to systematize them. His work on
this score remains incomplete. Yet there is good reason to think that this is
not what he meant to do. In the remainder of his treatise on the emotions
(sum. theol. a qq. ), Aquinas takes up each of the eleven kinds of emotions in detail. His analysis of each conforms to a rough pattern: he discusses
the causes of the emotion, its nature and formal object, and its eects. Sometimes issues specic to a given emotion are taken up, e. g. whether it is possible
to hate oneself, given the nature of hatred (sum. theol. a q. a. ); sometimes causal connections are explored, e. g. whether sympathy from friends
can help to alleviate distress (sum. theol. a q. a. ). Identifying distinct
species of emotion does not seem to be one of his main concerns. In many
ways, that is just as well, since taxonomic theories have lost the lustre they
once possessed, but Aquinass particular discussions of individual emotions
are still psychologically acute and philosophically rewarding.28
CONCLUSION

The last section might well lead one to wonder about Aquinass theoretical aims, if he is neither giving us quite an empirical theory founded on
observation nor a taxonomic genera-species classicatory scheme. There is
much to say about the sense in which Aquinas oers us a theory of emotion,
but perhaps the best way to approach the issue is to see how his theory is
related to contemporary theories of emotion.
We have already seen in that Aquinas is a cognitivist (of sorts) about
emotion, a nding that puts him with the majority of philosophical treatments
of emotion in the last fty years. But the discussion in gives us an even
closer comparison. Given the role played in the psychological economy by
sensitive cognition on the one side and by sensitive appetite on the other
side, it is clear that Aquinass account of the emotions is in many ways like
contemporary perception theories of emotion, e. g. in Roberts []. Such
perception-theories diverge from standard cognitivist accounts in taking the
evaluative element crucial to emotion to be not a judgment, with all the cognitive apparatus judgments draw in their train, but rather a perception, or
something very like a perception. From the safety of the sidewalk I might
perceive the onrushing trac as a threat even though I know that I am quite
28

There have recently been several studies of Aquinass views about particular emotions: Manzanedo [], Manzanedo [], Manzanedo [], Manzanedo
[], Drost [], Loughlin [], Green []; the need for such careful
studies of particular emotions is the central theme of Miner [].
c Peter King, forthcoming in The Oxford Handbook to Aquinas

CONCLUSION

safe where I am; if the evaluative judgment that I am in danger were a requisite part of the emotion of fear, then it seems hard to explain my fear in
the face of my knowledge of my safety. Perceptions, however, need not be
reasoned, though perhaps permeable to reason; they can be had by nonintellectual subjects, such as babies and dogs, who we want to say experience
genuine if primitive emotions; and they can deploy concepts without requiring their articulation the sheep may regard the wolf as dangerous without
having the concept of danger at all.
Aquinass account of the emotions is in many ways close to such perception theories. Indeed, Aquinas exploits the structural parallel between perception and emotion frequently, and, as with perceptions, he holds that the most
fundamental way to understand emotions is to see them as modes of engagement with the world. Again, Aquinas holds that the cognitive penetrability
of emotion derives from the susceptibility of perception to being aected by
changes in beliefs and thoughts. The rich array of psychological faculties that
Aquinas sketches, with the complex interplay among cognitive and aective
components, oers a congenial background for the contemporary cognitive
scientist accustomed to working with mental modules and their transference
of information via representations. If Aquinas draws a sharper line between
perceptual and intellectual cognition than most theorists are comfortable with
today, that is a small drawback for being centuries ahead of his time. The taxonomic structure he proposes might then be taken as a rst attempt to isolate
the natural kinds of emotion as an aective phenomenon. In short, we could
preserve his insights and his general approach to the emotions, and perhaps
even the general taxonomy, but leave behind some of its more medival
features, such as the appeal to the aristotelian theory of motion, or the necessarily immaterial character of intellectual functions. But a remarkable amount
of Aquinass analysis bears worthwhile comparison to contemporary theories,
as well as being a stunning intellectual accomplishment in its own right.

Peter King University of Toronto

c Peter King, forthcoming in The Oxford Handbook to Aquinas

BIBLIOGRAPHY

BIBLIOGRAPHY
Cates []. Diana Cates, Aquinas on the Emotions: A Religious-Ethical Inquiry.
Georgetown University Press .
Drost []. Mark Drost, Intentionality in Aquinass Theory of Emotions
in International Philosophical Quarterly (): .
Drost []. Mark Drost, In the Realm of the Senses: St. Thomas Aquinas
on Sensory Love, Desire, and Delight in Thomist (): .
Floyd []. Shawn Floyd, Aquinas on Emotion: A Response to Some
Recent Interpretations in History of Philosophy Quarterly ():
.
Galeazzi []. Umberto Galeazzi, Le passioni secondo Tommaso dAquino: De veritate q. in Aquinas (): .
Gorevan []. Patrick Gorevan, Aquinas and Emotional Theory Today:
Mind-Body, Cognitivism, and Connaturality in Acta Philosophica ():
.
Green []. Keith Green, Aquinas on Attachment, Envy, and Hatred in
the Summa theologiae in Journal of Religious Ethics (): .
King []. Peter King, Aquinas on the Passions in Scott MacDonald and
Eleonore Stump (eds.), Aquinass Moral Theory (Cornell University Press
): .
King []. Peter King, Emotions in Medieval Thought in Peter Goldie
(ed.), The Oxford Handbook of Philosophy of Emotion. Oxford University
Press : .

King []. Peter King, Dispassionate Passions in Martin Pickave and

Lisa Shapiro (eds.), Reason and Emotion in Medieval and Early Modern Philosophy. Oxford University Press .
Knuutila []. Simo Knuutila, Emotions in Ancient and Medieval Philosophy.
Oxford University Press .
Loughlin []. Stephen Loughlin, Tristitia et dolor: Does Aquinas Have a
Robust Understanding of Depression? in Nova et Vetera ():
.

Manzanedo []. Marcos Manzanedo, El deseo y la aversion segun San

to Tomas in Studium (): .

Manzanedo []. Marcos Manzanedo, Efectos y propriedades de la de
lectacion in Studium (): .
Manzanedo []. Marcos Manzanedo, El dolor y sus causas in Studium
c Peter King, forthcoming in The Oxford Handbook to Aquinas

BIBLIOGRAPHY

(): .

Manzanedo []. Marcos Manzanedo, La audacia segun Santo Tomas

in Studium (): .
Marmo []. Costantino Marmo, Hoc etiam etsi potest tollerari. . . Egidio
Romano e Tommaso dAquino sulle passioni dellanima in Documenti e
studi sulla tradizione losoca medievale (): .
`
Meyer []. Michel Meyer, Le probleme des passions chez saint Thomas
dAquin in Revue Internationale de Philosophie (): .
Miner []. Robert Miner, Thomas Aquinas on the Passions. Cambridge University Press .
Murphy []. Claudia Murphy, Aquinas on Our Responsibility for Our
Emotions in Medieval Philosophy and Theology (): .
Ricken []. Friedo Ricken, Aristotelische Interpretationen zum Traktat
De passionibus animae (Summa theologiae I II ) des Thomas von
Aquin in Martin Thurner (ed.), Die Einheit der Person: Beitrage zur An
thropologie des Mittelalters. Stuttgart: Kohlhammer : .
Roberts []. . Robert C. Roberts, Emotions: An Essay in Aid of Moral Psychology. Cambridge University Press .
Ryan []. Thomas Ryan, Revisiting Aective Knowledge and Connaturality in Aquinas in Theological Studies (): .
Torrell []. Jean-Pierre Torrell, LInitiation a Saint Thomas dAquin: Sa per`
sonne et son oeuvre. Presses universitaires de Fribourg (e ed. rev. et
augm.).
Uenheimer-Lippens []. Elisabeth Uenheimer-Lippens, Rationalized
Passions and Passionate Rationality: Thomas Aquinas on the Relation
between Reason and the Passions in Review of Metaphysics ():
.
White []. Kevin White, The Passions of the Soul in Stephen Pope
(ed.), Essays on Aquinass Ethics. Georgetown University Press :
.

c Peter King, forthcoming in The Oxford Handbook to Aquinas