Hallvard J.

Fossheim
Aristotle on Virtuous Questioning of
Morality

One Thought Too Many?

It is often thought that asking oneself “Why be moral?” is a sign of ethical im-
maturity. Asking oneself whether one should act justly, or honestly, or bravely
—or perhaps, asking oneself whether one should perform an act from the appro-
priate ethical motivation—appears in itself to betray some sort of shortcoming.
The more you are able to think seriously outside the ethical box, as it were,
the more it might seem that this is not a good thing, but rather a bad thing.
One might think of Bernard Williams’s famous notion of “one thought too
many” in this context.¹ His comments, as I understand them, are primarily
about the questionable psyche of the person who needs a detour through
moral theory (in his example: the idea of impartiality) in order to decide what
is the right thing to do. But something like that picture is relevant to the
“Why be moral?” question too. For asking this question in full earnestness
seems to require that the agent be sufficiently alienated from, or uncommitted
to, an ethical stance for him to be able to step outside it and question its validity.
Asking oneself “Why be moral?” can be characterised as a shortcoming per-
haps not least from a virtue ethical standpoint. For according to a common un-
derstanding of what virtue is, it means that your moral motivations have not
grown into a fully-fledged ethical character. You are, as it were, still open for sug-
gestions coming from outside the sphere of the morally responsible, which is to
say that you are not good. If something is part of your character, you cannot dis-
tance yourself from it; it has become part of who you are, and not something you
can take off like a shirt or a watch. So being able to ask the question implies that
morality is not settled in you as your character.

The Claims
What I would like to suggest in what follows, is that this is an undue simplifica-
tion of the matter. On the contrary, I would like to try to argue for two claims of

 Cf. the final paragraphs of Williams ().

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
66 Hallvard J. Fossheim

which the first is difficult to unite with the “one thought too many” story just out-
lined. And I am going to do this with Aristotle as umpire and argumentative
source.
One of the claims will be that when it comes to the individual’s process of
betterment, the question “Why be moral?” and its family of related questions
are highly useful and form an absolutely necessary prerequisite in the process
of moral improvement.
The main conclusion that I will try to make a case for is that the good person
too needs to be able to ask something very like the “Why be moral?” question.
That is, not only in our development towards ethical goodness, but in the state of
ethical goodness itself and as an integral part of it, the person will see and feel a
pull from other, non-moral options and alternatives. This is something which
puts the moral stance (if we can call it that) into relief, and amounts to a capa-
bility to ask something like the “Why be moral?” question.

The Question
Before proceeding, I should say something more about how I will approach the
“Why be moral?” question. I take the question not as a request for arguments, in
the abstract, for why any individual, or any rational individual, should follow
morality. Such arguments go in and out of fashion, and can build on, say, ego-
tistical considerations, normative ideas of consistency, or prudence. Rather, I
hope it is already becoming clear that the angle I want to test has to do with
what it means for agents like ourselves—and those less good than us, and
those better than us—to be asking basic questions about how to act and why.
The question “Why be moral?” is ambiguous in several ways. Let me point
out a couple of the ambiguities before advancing any further. One source of am-
biguity concerns whether, when I ask myself why I should be moral, I am asking
whether I should from an external point of view act morally in a given situation I
am facing, or whether I should be moral in the sense of being motivated by the
right sort of reasons. The implications of the two can differ greatly, as one might
perform the same set of external motions with very different motives.
A related ambiguity concerns whether I am asking myself the question with
a view to a singular setting, or with a view to broader vistas—ultimately, my life
as a whole. This ambiguity is related to the previous one in that we tend to think
of acts as singular, while being one way or another characterizes the person and
the person’s life more broadly. But the two are not identical, since it is possible to
ask about one singular situation and find that one worries about one’s soul, or to
ask about one’s personal qualities and find that one only cares about this be-

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 67

cause of how it might affect one’s performance in the present singular situation.
For present purposes, I would like to hold on to both sides of both ambiguities.²

Goodness, Character, and Akrasia

Let us start by briefly considering akrasia and her militant cousin enkrateia, as
these seem, at least at first blush, to constitute Aristotle’s main instances of
the sort of psychological openness characteristic of the “Why be moral?” ques-
tion. The acratic seems to be characterized by a two-sidedness where one of the
thoughts or impulses overrules the notion that there is indeed a positive reply
available. It bears noting that if it is anything like what the literature normally
suggests, akrasia must be extremely widely diffused among us. The acratic on
this conception cannot be a rare bird, but will be exemplified by most of us at
least from time to time.
The acratic has standarly been used to establish a picture of Aristotle’s good
person as someone who no longer sees, or is able to consider, what the acratic
sees and is affected by. However, we don’t need to think of the motivational
set of the good person as having silenced all other considerations than the ethi-
cally ideal ones. I suspect that part of what has made such readings possible has
been a wrong-headed and overly broad conception of the acratic and encratic
types. Aristotle’s analysis of the acratic and the encratic types does not force
us to admit a class into which most of us fall most of the time.
You may or you may not agree that the good person must be open to sugges-
tions in a way not allowed by what we might call the “total silencing” interpre-
tation. (I call it the “total silencing” interpretation, because it is not entirely clear
to me where John McDowell’s reading lands on the question of how the virtuous
agent experiences the lure of alternative courses of action, or whether she does
so at all.³ What sort of experiential difference from the encratic’s experience does
his talk of “not count for him as any reason” (91) or “count for nothing” (92)
imply? If silencing is substantially different from outweighing or overriding,
then it is difficult not to see this interpretation as yielding a view of goodness

 By these comments, I at the same time want to stress that I do not mean to limit the question
“Why be moral?” to instances where the individual understands herself as asking, hyper-exis-
tentially, once and for all and across the board whether she would like to belong to morality
or not. Most instances of people posing themselves the “Why be moral?” question clearly are
not like this, and the ones that are, perhaps are not in the end as important as the person at
the time feels them to be.
 Cf., e. g., McDowell (), pp.  – .

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
68 Hallvard J. Fossheim

which is not only idealized to the point of being non-existent, but—as as I’ll try
to illustrate in what follows—ethically impoverished as well.

Being Good and Seeing Bad

Here is one argument for the necessity of the good person in a certain sense hav-
ing badness in him. In order to act well, one must be able to grasp the charac-
ters, motives, and schemes of others. No act is an island. What the act signifies,
and not least, whether it succeeds, will depend on the agent having grasped
what motivates the other individuals involved. Furthermore, success in action
depends on the agent’s ability to give some sort of prediction of how each of
the others involved will see and react to what one does. Now this in turn depends
on one being able to somehow align with each of them. Understanding another’s
point of view in a practical sense is not a matter of treating him or her as a black
box, and through some algorithmic, statistical, or accidental approach calculat-
ing the actions that will follow. On the contrary, it is a matter of seeing the whole
from that person’s point of view. What are the goods at stake? What is the other
person up to? Which alternative courses of action seem most desirable to him,
and why? Et cetera. But you cannot do this without, as it were, having a taste
of those goods, desires, and perspectives as the other person experiences them.
Whether we think of this quality or ability in terms of emotional intelligence
or as being savvy, these other perspectives must be available as part of the hori-
zon for the agent, if he or she is to be successful in action. How can we get to a
closer determination of what sort of stance the good person must have towards
this plethora of shortcomings?
Now “having a taste of” is not the same as “having a taste for” in a full sense
of actually being ready to go in for whatever is the object in a given case. So the
goods in question are not the individual’s own preferred goods, simply because
the individual is able to read the other person’s perspective. But on the other
hand: the way the good person relates to the other person’s perspective must
normally be more than only knowing that the latter person goes for that sort
of thing. It must in fact involve a real perspective on ways of construing options
that imply the felt presence of the “Why be moral?” question, in some cases even
the denial that there is a positive account which could yield a reply to it.
Looking to Aristotle, he is clear that many less-than-perfect desires and acts
are not to be liked, sympathized with, or even tolerated. Often, part of being good
will be reacting instinctively to something—a personal characteristic, a motive,
an action—as disgusting or as a provocation. So understanding is not the
same as condoning. Nor is it, in such cases, being overwhelmed by the point

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 69

of view of the other. But the other’s experience must still be somehow available
and recognizable to the agent.
Do we have Aristotle on board here? On the one hand, Aristotle seems very
clear that being good includes swift and merciless denunciation of certain ac-
tions. There is no reason to think that there is a crucial difference between
how the good person is supposed to react instinctively to others’ actions and
how he is to react to their motives. An action, in Aristotelian parlance, normally
includes the motive or motives behind it. And just as the person shows himself as
being all the better for not having to think the matter through, but straightaway
initiates the proper response, we should think (all else being equal) that the good
person will be immediate in his emotional-cognitive responses to others and to
their actions.
Is this picture compatible with the notion that the good person has direct
access to bad motives and desires?

Emotions and States

A basic question can be articulated as: “What is it like to be a virtuous person?”
On what I understand to be the common interpretation, the good person is some-
one who feels and reacts only in the ways that are identical to, or expressive of,
virtue as a way of seeing the practical field. But we have seen that there are rea-
sons to think otherwise. Not least is the consideration that in order to act well,
you need to be able to grasp—or “see”—the point of view of all kinds of people.
This consideration also reminds us that in a way, what I take to be the re-
ceived interpretation can be said to represent a sort of ethical poverty when it
comes to the capacities of the good person. Now what is unavailable to the
good person on this interpretation, is every point of view other than the good
person’s own, perfect, take on things. We might call this “the narrow view con-
ception of virtue”. Of course, it would all be worth it in the end, if this is what it
takes to be good. But at the same time, this interpretation of the good person
holds that goodness includes a massive impairment. If being in the state
which allows you to act well also entails that you lack all perspectives other
than that of goodness, can we then not properly speak of virtue as a sort of cog-
nitive shortcoming?
I want to suggest that goodness-as-cognitive-impairment, in the sense just
described, does not necessarily represent Aristotle’s view. But I will admit that
the received interpretation does have textual evidence working for it. In a man-
ner very different from that of Plato, Aristotle seems to take as part of the core of
his moral psychology that growing up and being brought up means taking on a

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
70 Hallvard J. Fossheim

character, a more or less consistent set of states definable in terms of virtues and
vices. And such a character is presented as a solid and lasting accomplishment.
Once you have a character, changing it in any substantive way is usually present-
ed as impossible, or close to impossible.
In order to see whether openness to the “Why be moral?” question and its
concretizations in given situations might be construed as part of Aristotle’s ac-
count of a virtuous person, it is necessary to consider more closely how Aristotle
classifies states of character. Aristotle says,

[s]ince there are three conditions arising in the soul—emotions, potentialities and states—
virtue must be one of these. […] If, then, the virtues are neither emotions nor potentialities,
the remaining possibility is that they are states.⁴

So virtues, and by implication vices, are states. A state or hexis is something du-
rable and determining for action. It is not like a coat of paint, which on a whim
can be scraped off or painted over in some other colour. A state is more like a
second nature.
Furthermore, a hexis represents a serious narrowing in relation to a capacity
or dunamis. Says Aristotle, at a later point in the Nicomachean Ethics, “while one
and the same capacity or science seems to have contrary activities, a state that is
a contrary has no contrary activities”.⁵ The underlying logic seems to be that if
nature provides us with a capacity to become good, then a hexis is a realization
of that capacity. That is to say, one and the same capacity functions as the basis
for different and opposing states. Capacities belong on a level of determination
which can “hit both ways”, as it were, in relation to virtue and vice. The same
capacity thus stands as an explanation, or at least a partial explanation, of dia-
metrically opposed ethical states. To the extent that a capacity for anger, for in-
stance, is a basis for the virtue of mildness, the same capacity also stands in the
same relation to the vice of hot-temperedness. Both the virtue and the vice, and
all the other more or less vicious states building on a capacity for anger, can be
traced back to that capacity as a natural condition for their existence. This is why
capacities are said to relate to actions in the way the possessor of medical sci-
ence relates to acts of healing: the doctor is the one eminently capable of both

 EN II, v, b – , a – ; Irwin’s translation, modified by substituting, for the
sake of consistency, “emotion” for “feeling” and “potentiality” for “capacity” throughout.
 EN V, i, a – .

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 71

killing and saving her patients. And correspondingly, a capacity has the structur-
al role of underlying the corresponding virtue and vice alike.⁶
This would seem to constitute a strong argument in favour of the received
interpretation. An ethical hexis determines the person. One hexis characterises
person A and her relevant actions, while another and opposed hexis character-
ises person B and her relevant actions. So if your state is one of mildness,
then mildness is all you feel in the relevant cases. Hot-temperedness is alien
and nothing to you.
But does this really follow? I think there is reason to tread carefully here. The
first claim seems unproblematically to follow: a hexis narrows a person’s range
of possible actions, and this is why it can function as explanans to the explan-
andum of her action.
But the second claim does not follow. The premises do not necessarily entail
that a person with a given hexis does not have any access to other emotive reac-
tions. Consider the following passages.

in so far as we have emotions, we are said to be moved; but in so far as we have virtues and
vices, we are said to be in some condition rather than moved.⁷

First, then, neither virtues nor vices are emotions. For we are called excellent or base in so
far as we have virtues or vices, not in so far as we have emotions [kata ta pathê]. We are
neither praised nor blamed in so far as we have emotions; for we do not praise the
angry or the frightened person, and do not blame the person who is simply angry, but
only the person who is angry in a particular way [pôs]. But we are praised or blamed in
so far as we have virtues or vices.⁸

There is more than one reason why it is important for Aristotle to distinguish
sharply between hexeis (virtues and vices) and emotions. One of them is that
without this distinction, there is no room for a theory of acquired ethical virtue
of the sort he is out to defend. So the distinction is at least principled, in the
sense that emotions and states are different even if it turns out that they are
one and the same moral psychological entity in any given instance.
But this does not rule out the possibility that one might have a wide variety
of emotional reactions to a given situation, as long as one of them is the ruling

 Not all potentialities relate in this way to two diametrically opposed states. The motion of the
elements, e. g., is unidirectional, in that the elements (such as a piece of earth) do not harbour a
potentiality for developing more determinate states. The capacities for human goodness, howev-
er, do hold such possibilities.
 EN II, v, a – .
 EN II, v, b – a.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
72 Hallvard J. Fossheim

one and the one shaped into part of one’s character: that is to say, as long as one
of them is the one that foreseeably determines action.
To what extent does Aristotle think of the emotion as remaining somehow
intact “underneath” the state? He cannot think of emotion as something
which as a matter of fact exists in abstraction from what we might term the
“how-ness” of virtue or vice. For any case of emotion must always present itself
in some way or other. Aristotle says both that emotion as such belongs on a gen-
erally animal level, which is transcended by virtue, and that virtue is nothing but
the alteration, rather than the overcoming, of emotion. So the question really is
how much can be said about human nature, in a narrower sense, as something
which is still a presence in the habituated individual. The following passage
seems to support the idea of emotion as a constant aspect of, or basis for, our
states and actions, inherent in us from nature. A virtue of character

is concerned with emotions and actions, and these admit of excess, deficiency and an in-
termediate condition. We can be afraid, e. g., or be confident, or have appetites, or get
angry, or feel pity, in general have pleasure and pain, both too much and too little, and
in both ways not well; but [having these emotions] at the right times, about the right things,
towards the right people, for the right end, and in the right way, is the intermediate and
best condition, and this is proper to virtue.⁹

A virtuous state consists at least in part in being well off in relation to emotion.
And a virtuous state marks, Aristotle seems to say, a middle point in a determi-
nate range of states, the extremes of which are full-blown vices. Therefore, it
makes good sense that at least some virtue-vice continuities get much of their
unity from relating to some emotion or other.
Naturally, it is not necessarily only emotion which defines an ethical virtue.
Aristotle also seems to place some emphasis on the notion of a sphere of action
in this respect. But emotions do yield many of the virtues. Bravery relates in this
way to fear.¹⁰ Temperance relates to the basic appetites. Generosity, magnifi-
cence, magnanimity, and “the virtue concerned with small honours” can of
course deal with goods which are ostensibly objects of appetite. But ultimately,
all these virtues are also ways of relating to the emotions, such as anger, a form

 EN II, vi, b – .

 And to confidence, as noted by, e. g., Urmson (), pp.  – . After Pears () and
in particular Hursthouse (), I do of course not mean to set forth the idea that each virtue is
delimited by experiencing the right amount, quantitatively, of one emotion. What I wish to do is
bring out two central aspects of the relationship between emotions and virtues/vices.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 73

of the self-assertion of thumos. ¹¹ In these central cases, then, we see that emotion
is imperative in providing focus for the virtue-vice continuity.¹²
But if a virtue-vice continuity can be said to be based on emotion, that tells
us something about the cognitive strength of such an emotion. The relation be-
tween emotion and state seems to be partly a logical one: that is, in a more or
less completed or habituated individual, it is possible to “peel off” in analysis
the state determination and be left with a notion of emotion which amounts
to much more than only a reconstruction of some original natural state. To
name but one example again, bravery is concerned with the emotion of fear,
which is thus already conceived in terms of a desire to avoid danger. We must
therefore think of Aristotelian emotions as always already more or less determi-
nate. Expressed in terms of desire, then, fear will be a desire to avoid danger,
irrespective of which state one has, that is, of how one is disposed towards
one’s fear. The emotion has sufficient cognitive strength to be determined
apart from whether the human being in question is “well or badly off” with re-
spect to it. On a more general level, then, the virtuous and the non-virtuous per-
son may share the relevant emotion as a way of experiencing or grasping the sit-
uation.
There is support for such a reading outside the Ethics too. For this, presum-
ably, is the level on which the Rhetoric’s definitions of emotions operate, allow-
ing for its characterisation of fear as “a sort of pain or agitation derived from the
imagination of a future destructive or painful evil”.¹³ This is a definition which
determines the general cognitive direction of an emotion, that is, its general
sort of object, without including an account of just how fine-grained or advanced
the cognitive structure might be. But the point remains that, even in abstraction
from the state, Aristotle’s characterisation of fear allows it an important role in
cognising the relevant situations. Sticking with the present example, a danger is
something which threatens one’s own being; this appears to imply that as an
emotion, fear has to include some grasp of oneself as opposed to others

 Slightly confusing in this context is how Aristotle seems to conflate honours with what seem
to be merely material goods, but the clue for grasping the connection is probably that both
money and honour classify as external goods.
 More problematical is, e. g., general justice, which relates to all the emotions via its relating
to all the other virtues considered as a whole, or even special justice, which is the part of general
justice dealing specifically with honour or wealth or safety, and accordingly relates to emotions
like anger, appetite, and fear. Not to mention a virtue like wit, which seems to relate more ex-
clusively to a realm of action (dialogue or discussion) than to any particular recognisable emo-
tion.
 Rhetoric II, v, a f; translation George A. Kennedy.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
74 Hallvard J. Fossheim

(whose danger is an object of pity rather than of fear); of the pain of injury or
obliteration as opposed to the pain or discomfort of, for instance, mild hunger;
and of a future state of affairs, that is, of something which is not yet present. But
it does this without saying how the individual relates to that fear—well (the vir-
tuous state) or badly (some more or less vicious state). This quietude is, I think,
central to allowing the notion of emotion the importantly non-ethical, or rather
pre-ethical, status it holds. For, given the way it is defined, an emotion like fear
might apply equally to a logical analysis of the virtuous character of a mature
individual and to the doings of a child (or perhaps even an animal).
So the state of affairs is something like this: in mature human beings, every
external display of emotion is at the same time a display of some state or, in the
case of natural virtue, state-like manner in which it is set, as a cognitive “tuning”
of that emotion. Emotion seems to be treated at least partly as the aspect of a
virtue or vice which is catered for by nature. A state proper, by contrast, is not
the result only of nature as an internal principle of development, but has been
formed by habituation. So any mature display of anger, for instance, is simulta-
neously a display of an emotion and of a state.
Thus, in certain contexts emotion can even be treated as if it existed in sep-
aration from virtue and vice. This is evident from Aristotle’s advice to the public
speaker. Aristotle’s insight, in the Rhetoric, is that since the public speaker can-
not fully take into account the individually developed character of each person
in the audience, the capacities for emotional response, which are universally
shared among all or most, form the level at which a reflective analysis of public
address must be carried out. One person is more rash than another, and one
more egoistic than another, so the address cannot be made to serve as a perfect
vehicle for ensuring any fine-tuned aggressive response in each and every one
among them. What the speaker must do, is attempt to create a desire for revenge
in each by considering the more generally shared level of emotion. In this sense,
the more general emotive level can still be said to be directly available in the ma-
ture, state-determined individual. And it may be Aristotle’s view that the emotive
perspectives of other characters will be available to the good or decent member
of the audience, although she will not act on what has in this way been made
emotively available to her. This means that a good person may be able to expe-
rience something like the impulses and perspectives not only of those who feel
the existential force of the “Why be moral?” question, but even of those who
have settled on a denial that the question has a positive answer. While the
good person will not herself be unsettled by its force, its experiential recognition
will enable her both to act and to speak more wisely.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 75

Three Possible Objections

I want to very briefly respond to three possible objections concerning my inter-
pretation of the specifics of Aristotle’s theoretical framework. They concern, re-
spectively, (1) the settled quality of character, (2) the difference between praxis
and technê when it comes to their characteristics, and finally (3) an objection
of a more purely normative sort.
As for the first possible objection, this would be the suggestion that, al-
though we are “somehow co-responsible”¹⁴ for our character, this very phrasing
reminds us that character is not something we can change like a shirt. While this
is surely true, it does bear noticing, however, that Aristotle in this passage is
speaking only of people who are coming up short. He does not in fact say any-
thing about any possible limitations of the good person. The responsibility in
question concerns a less than good person who has done something that is no
good, and this also means that the concern, in the context, is to point out
that such a person could have made something else of himself. And in fact,
even here the setting implies that the person in question is defending himself
in a way that requires his ability to see that what he did was wrong. If you
are blaming something on your own weakness or shortcoming, then you are cer-
tainly capable of seeing it as a shortcoming. So this passage will not save the
“narrow view conception of virtue”.
The second objection makes things more interesting. The claim here is that
Aristotle himself implies a strict POV (“point of view”) limitation for the good
character too, in his definitional contrasting of character from technical knowl-
edge. While technical knowledge and know-how enables you to swing both ways
—say, using the same medical knowledge to save or to kill—ethically embedded
insight is defined by opening up a space of action and reflection in only one di-
rection. While the doctor can save and kill, the good person can perform only
good deeds. So while the bad person is trapped in the direction of evil, the
good person is equally trapped in the direction of good, unable to perform un-
just, or gluttonous, or cowardly actions.
But this is a misrepresentation of the case. More accurately, the criticism de-
pends on a confusion of levels. We can agree that a certain character is partly
defined in terms of a certain set, or range, of actions—actions that are just rather
than unjust, for instance. But this is not what is at issue. What is at issue, is
whether the stability that a character seems to provide is provided through
that character’s inability to think and feel otherwise—that is, to think and feel

 Aristotle’s wording is sunaitios pôs (EN III, , b).

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
76 Hallvard J. Fossheim

what would, had the experience been action-effective, lead to acts other than
those of the good person. And assuredness about this latter contention is
what the argument so far has hopefully served to undermine.
A final, more normative kind of criticism can also be raised. The claim will
be that we do not want a state of affairs where goodness is that fragile. A world
without solidly good people, without characters that are unambiguously virtuous
and will remain so at least in the near future, is a world of distrust and anxiety. If
there is no solidity in goodness, why bother disciplining oneself or providing oth-
ers with a good upbringing?
Here, it is of course possible to respond simply that our not liking something
is not necessarily a knock-down argument against its being true. But we can also
say more. For again the objection misses the mark, much in the way the last ob-
jection did. The vision of a world off its hinges does not follow from the sugges-
tion that if character entails a significant measure of stability, it does not do so
through simple constriction. I want to end by trying to make this difference in
levels more clear by taking a developmental perspective on character.

Insights from a Developmental Perspective

I hope to have established that Aristotle does not necessarily claim that the good
person only has access to emotions perfectly attuned to performing good actions.
And correspondingly, nothing in what he says denies that the less-than-good
person might have emotive access both to material approaching the good per-
son’s effective state, and to the perspectives of other, alternative, less-than-
good characters. On the contrary: in the case of less than good people, how
could we grasp their betterment as taking place without an appreciation of
other perspectives than those they act upon?
This also means that Aristotle does not necessarily hold that it is only the
akratic and the enkratic who have experiential access to two or more emotive
perspectives. What distinguishes the akratic and the enkratic is not that they
have this emotive access, but that they are seriously in two minds about how
to act.
We get support for this reading of what is crucial to virtue—and to vice—by
considering something which Aristotle himself sees as a central issue in his dis-
cussion of virtues, namely, the question of how we develop them. It is striking,
once you think about it, that Aristotle does not present any specifically mental
exercise as paramount to developing virtue.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 77

A state [of character] arises from [the repetition of] similar activities. Hence we must display
the right activities, since differences in these imply corresponding differences in the
states.¹⁵

Virtue of character [i. e. of êthos] results from habit [ethos]; hence its name “ethical”, slight-
ly varied from “ethos”.¹⁶

Aristotle’s unambiguous response to the question of what is required to become

virtuous does not include any exercises designed to purify one’s soul or to
cleanse one’s psychological set-up. What he presents as the all-important
thing is ensuring that the acts one performs are uniformly expressive of virtue
(partly defined in terms of wishing the act for its own sake). Not only is there
nothing here which suggests that there is no room in the good person for alter-
nate perspectives, as long as they do not seriously interrupt the flow of virtuous
acts; the training method would seem to be positively ideal for not rooting out all
plurality of perspective, as long as the predictable, harmonious solidity of good
agency is ensured.
So while such solidity naturally requires feeling the right pleasures and
pains, this does nothing to exclude a richer emotional life than the narrow
view conception of virtue would allow. There is plenty of room in the logical
space between Goody Two-Shoes and the enkratic for good agents who can
still see and feel the attractions of other routes than that defined by virtuous ac-
tion.

Emotional Phantasia
Do we have any more positive evidence that Aristotle might harbour a view like
the one I have tentatively been attributing to him? This brings us to the topic of
phantasia, a concept that has been understood by readers of Aristotle in widely
different ways. It would go far beyond the present inquiry to try to construct any-
thing like a unified Aristotelian theory of phantasia. But the few points directly
relevant to our question might be brought out without attempting anything so
ambitious. For as it turns out, what he has to say on the matter in De Anima
(III.3, 10 – 11) and in De Motu Animalium (esp. 6 – 8) strongly suggests a psycho-
logical model in line with the sort of openness I have been advocating. Consider
the following passages.

 EN II, i, b – .

 EN II, i, a f.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
78 Hallvard J. Fossheim

This, then, is the way that animals are impelled to move and act: the proximate reason for
movement is desire, and this comes to be either through sense-perception or through phan-
tasia and thought.¹⁷

For the affections suitably prepare the organic parts, desire the affections, and phantasia
the desire; and phantasia comes about either through thought or through sense-percep-
tion.¹⁸

The sketch is rudimentary, but clear. What should strike us first of all is that the
action-inducing desire is not present at the outset. Rather, the more unwieldy
and manifold power of phantasia generates a desire (orexis), which then in
turn generates an emotional reaction (pathos) proper. This means that phantasia
must also be seen as part of the early searching phase (zêtêsis) of action. In this
phase, several possibilities might suggest themselves to the agent. We must
imagine that this is the rule rather than the exception: most cases of practical
searching will lead to more than one option, although they will not all be active-
ly endorsed by the agent in the sense of generating what we might call the “rul-
ing passion”—the action-efficient reaction.
A search may be instigated by the agent’s own broad practical agenda, or by
a physical state, or by something the setting unexpectedly presents him with. To
concretize, the search can be initiated through a general notion to do some good,
or as a reaction to an organic lack of water, or by a lion suddenly jumping up in
front of him.
So the story does not start with a determinate, character-defining desire gen-
erating an action. This belongs only to the final stages of the story. In the begin-
ning is the much more open-ended phenomenon of the imagination doing its job
of displaying various hypothetical situations or results.¹⁹ That it makes sense to
think that the prospects come in the plural is also suggested by the basic exam-
ple of Aristotle’s, coming by the help of phantasia to see that “This is drink”, and
then drinking (MA 7, 701a32– 33). Surely, in many situations there will be more
than one option. And in those cases, the narrowing down takes place after phan-
tasia has done the job of conjuring up the various possibilities.
The progression phantasia → orexis → pathos does not rule out that there is
some emotive activity in the agent during the initial phase before orexis and then
pathos are generated. Rather, Aristotle seems to suggest that there may be appe-

 De Motu Animalium , a – ; Nussbaum’s translation.

 De Motu Animalium , a – .
 That what is imagined are resulting states which also include the agent, is suggested by the
specification at EN III (, a – ) that the lion does not enjoy the look or sound of the
lamb, but “the prospect of a meal” (hoti boran hexei).

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
 Aristotle on Virtuous Questioning of Morality 79

tites, thumetic impulses, and more intellectual desires present here too. Again,
what is effected by the agent fastening upon a given phantasia is not that
only then does anything like an emotional reaction start—but, that only then
has the action-effecting desire and emotion come to rule the day.

Conclusion
If my suggestions make for a viable interpretation, then the question “Why be
moral?”, surprisingly, does have a place in the life of Aristotle’s good agent. Al-
though she will not be led into temptation by it, she will be able to ask this and
similar questions with a seriousness and understanding available only to the
person who can see and somehow feel their pull. And this ability, on the part
of the good agent, makes a real difference both for the richness of her horizon
and for her practical ability to realize the perspectives of other agents less
good than herself.
Presumably, there will be a limit to what sort of questions the good agent can
ask herself without it being indicative of a lack in goodness; similarly, there will
be a limit to how insistently she can pose them. But these limits seem to be well
outside the borders set by ethically ideal agency as this is traditionally con-
ceived. Far from being a sign of moral immaturity, then, a questioning and
self-questioning from well outside the safe centre defined by virtuous motivation
can be seen as a prerequisite for full ethical agency. Giving a twist to Williams’s
formulation, we might say that in this context, something like “one thought too
few” is in fact what poses the greater threat.²⁰

Bibliography
Hursthouse, Rosalind (1995): “A False Doctrine of the Mean”. Proceedings of the Aristotelian
Society 81/1, pp. 57 – 72.
McDowell, John (1998): “Are Moral Requirements Hypothetical Imperatives?” In: McDowell:
Mind, Value, and Reality. Cambridge, Massachusetts: Harvard University Press,
pp. 77 – 94.
Pears, David (1980): “Courage as a Mean”. In: A.O. Rorty (Ed.): Essays on Aristotle’s Ethics.
Berkeley: University of California Press, pp. 171 – 187.

 I am grateful to the participants of the “Why Be Moral?” conference held at UiT, The Arctic
University of Norway for their helpful suggestions, and to the editors for their valuable advice in
the final revision of this piece.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM
80 Hallvard J. Fossheim

Urmson, J.O. (1980): In: A.O. Rorty (ed.): Essays on Aristotle’s Ethics. Berkeley: University of
California Press, pp. 157 – 170.
Williams, Bernard (1981): “Persons, Character, and Morality”. In: Williams: Moral Luck.
Cambridge: Cambridge University Press, pp. 1 – 19.

Brought to you by | Stockholms Universitet

Authenticated
Download Date | 9/24/15 2:33 AM