15.1.2009
Stoic philosophers sharply distinguish between two sorts of affective
responses: emotional and eupathic ones.
1) A few cautionary remarks: in fact, the affective spectrum
acknowledged by the Stoics is even broader, including further sorts of
phenomena; but for my present purposes, emotional and eupathic responses
are the most relevant.
As always when Stoic philosophy is concerned, we have to be careful not to
be mislead by a more modern use of some key terms. Most obviously, the
word 'stoic' does of course refer to the philosophical position of the
ancient tradition called the Stoics, not (as modern usage suggests), an
attitude of suppressing affective responses, promoting an indifferent
attitude towards what goes on in our lives. My use of the term 'emotion',
in accord with the recent literature on the Stoics, is also more narrow than
one might expect. Emotions, in this narrow sense, are distinguished from
moods and feelings in that they have cognitive and voluntary aspects, and
even within the class of responses so delimited, there are responses (which
I have referred to as 'eupathic') which don't count as emotions in the narrow
sense. Sometimes the wider (modern) usage of the term 'emotion' is
accommodated by calling eupathic responses 'good emotions', implying that
'emotion' is a genus term with the two species of the narrowly emotional and
the euthpathic. (This is fine as long as the difference is made clear, but
I'll mostly avoid this language.)[1]
Strictly speaking, the Stoics ascribe eupathic responses only to ideally
wise people, or sages (who have perfectly developed their rational
capacities, always act virtuously, and are fully consistent in all their
beliefs and actions). I'm leaving aside this dimension and assume that both
emotional and eupathic responses can simply be described by their cognitive
features, without reference to the theoretical construct of the Stoic
sage.
2) Both emotional and eupathic responses are voluntary reactions
to bits of experiential intake in propositional form, i.e. they are reactions
to impressions, or appearances, that things are thus and so. It seems to us
that the situation we are facing is thus and so, and we assent to that
impression, forming a belief that things are indeed thus and so. More
precisely, the character of the impression is not so much descriptive; it
implies rather a suggestion of an appropriate response. More correctly, then,
we are not assenting to an impression that things are thus and so - we are
assenting that things are thus and so and that it is appropriate to
respond in a given way.
So far, this description fits not only affective responses, but actions in
general: actions are assents to particular impressions that include a
suggestion for reacting in a certain way. What is special about affective
responses is that the reaction so suggested has the phenomenological
mark of feeling in a certain way. Affective responses have the
characteristic 'feel' of what we (in modern usage) call 'getting emotional'.
(But note that this applies to both emotional and eupathic responses.)
In addition to the phenomenological difference, these affective responses
contain a particular essential element in the involved beliefs, namely an
evaluative element of a specific type. The beliefs that are formed in
affective responses (again: both emotional and eupathic) are beliefs that
judge something in the situation as good or bad in a strong sense. When we
call something 'good' or 'bad', this assessment may mean nothing more than
that in just this situation, something is welcome or unwelcome, but not much
depends on it; it could mean that we recognize a happy coincidence in that
what happens is beneficial given the circumstances, but might have been
irrelevant or indeed positively spoiling our plans would the circumstances
have been different. In any such cases, we might see something in the
situation as 'good' or 'bad', but this sense of the terms is wider than
what is meant when we say that an affective response includes an evaluation
as 'good' or 'bad'. The latter talk intends a much stricter sense: something
counts as 'good' or 'bad' only if it is positive (or negative) absolutely,
that is, in any context, at whichever point in our plans, and independently
of what else goes on.
Obviously, we wouldn't expect too many things really to be good or bad in
this very strict sense. A general rule that something is really always
to be evaluated so would on reflection look implausible for many kinds of
thing. Having just received a given sum of money, for instance, may look
'good' in the wider sense (because, say, it amounts to half a month's income,
and is a welcome extra just now), but not in the stricter sense, for we can
easily imagine situations where we would be quite indifferent to that
particular sum. Thinking a little more about this, we might expect that
this is true for situations where we receive money much more generally:
whatever sum, it is conceivable that we might be in situations where it
wouldn't be that helpful. Likewise, the thought of losing some amount of
money may look bad in the wider sense for some situations, but again, losing
money seems on reflection not be the sort of thing that justifies calling it
'bad' in the narrow sense.
Affective responses, according to the Stoics, are characterized as
assenting to impressions that something good or bad, in the narrow sense
explained above, is present. (Or, alternatively, something good or bad in
that sense is in store for the agent, i.e. will possibly happen to her.) Let
us call this sort of evaluation uncompromising, and
evaluations as good or bad in the wider sense (not including cases of good
or bad in the narrow sense) restricted evaluations.[2] This
terminology has the nice feature that it seems to capture some of the
compelling nature of emotional responses, in the connotations of the term
'uncompromising'. It is this aspect that interests me, and I'll explore it
further below. First, however, we still need to clarify the difference
between the two kinds of affective response (so far, we have only looked
at what they have in common).
3) While the analysis of affective responses, as outlined so far,
is not particular to the Stoics, the difference they make between the two
sorts (emotional and eupathic) is specifically Stoic. It is not actually
empirically based, but has a normative force: it's a distinction between
those affective responses that we should have and those that we shouldn't
have. The Stoics think that the possible objects of affective responses
can be divided up into those which merit an uncompromising evaluation and
those that don't. All affective responses involve uncompromising evaluations,
as a matter of fact, but with some objects, such as the loss of a sum of
money in the example above, this isn't actually justified - we shouldn't
treat such a loss that way, that is: we shouldn't have an emotion about
losing money. We can't have a eupathic response, too, because these are
responses to situations which involve objects that do merit them.
Loss of money isn't among them. So loss of money is an example for something
we should never respond to; if we do, wrongly, then that is called an
emotion. If we respond affectively to those things which according
to the Stoics are indeed worth of an uncompromising evaluation (which
are our own virtuous or vicious behaviors), then these responses are ok,
from a Stoic point of view, and called eupathic.
4) I have described the difference between emotional and eupathic
responses as a normative difference, as a matter of what should or
shouldn't be. However, this is not all we can say about that
difference.
I have given an example for a typical object of emotions: gain or loss
of a sum of money, and I have noted that the Stoics do not think this is
the sort of object which merits an uncompromising evaluation (an evaluation
of the event as good, or bad, in the narrow, absolute sense). Therefore,
if an affective response is made towards it, we have a case of an emotional,
as opposed to a eupathic response. I have also briefly mentioned that
one's own virtuous or vicious actions are the sort of object which make
a response eupathic, rather than emotional. So what is the general criterion
for telling objects of emotions apart from objects of eupathy? Since it
is a distinction which is taken to have a normative force, the difference
should be one in value. So we have to turn to the Stoic value system next.
At this point, another distinction comes into play, namely the distinction
between external and integral objects of
affective responses. The Greek philosophical tradition treats as externals
all the things we may pursue (such as wealth, a career, or reputation) but
which we cannot ascertain just by ourselves: in order to get what we want,
we rely to a certain extent on luck or help, which may not be forthcoming.
That is, when pursuing an external, even if we do our best, it is not
guaranteed that we actually achieve it. In contrast, there are things that
are fully under our control, namely: how well we act. (Again, with the
qualification that we actually do our best here.) This is completely up to
us, and therefore called integral.[3]
A basic principle of the Stoic value system is that only integral
objects have the sort of value towards which an uncompromising stance is
appropriate. In other words, an uncompromising stance towards an external
object is always mistaken; whereas an uncompromising stance towards an
integral object is justified. (We must be careful here: both an
uncompromising evaluation as good and one as bad are possible towards a
given integral object, but only one of them can be correct in being both
uncompromising and have the right direction, i.e. good vs. bad. So there is
at least a theoretical possibility of an evaluation of an integral object
being wrong, even though it is directed towards an integral object. Thus
uncompromising evaluations are always wrong of external objects, but not
necessarily always right of of integral objects. But it is always right
to be uncompromising about integral objects, as opposed to evaluate them in
a restricted way.)
5) As Terence Irwin points out,[4] the plausibility of the Stoic
theory of value depends on their claim that the uncompromising character
of affective responses (and in particular, emotions, which we know well
from everyday experience) is connected to the particular sort of
evaluation involved, i.e. our treating, in an emotional response, some
object as good or bad absolutely, in the way described above.
Let us first note that restricted evaluation seems to be the more
sophisticated process, whereas uncompromising evaluation looks more like
a direct, simplifying reaction. Given a certain object, an evaluation of
it as good or bad would stand until a restriction is found; in other words,
an evaluation would be good or bad in the narrow sense by default, and only
when supplied by an additional consideration of a possible circumstance
it can be recognized as requiring a restricted evaluation.
This line of thought adds some support to the idea that restricted
evaluations are more appropriate, except in the special cases where they
are impossible, because there is no possible restriction (i.e. in cases
where integral objects are involved). Whenever there is a possibility to
get, via reflection or examination, to a restriction, we should do so.
Ideally, we would end up with a restricted evaluation always when
externals are concerned, for the view is that there is necessarily at
least one restriction to be found. Only in the case of integrals we can
safely assume that there is no restriction to the evaluation (because the
value of an integral doesn't change with context), and therefore we are
justified in an uncompromising evaluation once we recognize that it is in
fact an integral object. Likewise, we should never feel justified in taking
an uncompromising stance once we have recognized the object as an
external.
Still, this doesn't help us with the claim that an uncompromising stance
follows once we think of something as good (in the narrow sense). So
long as we don't make a distinction between integrals and externals, and
consequently don't try to adjust our responses by taking that distinction
into consideration, it seems that we react instinctively with an
uncompromising evaluation to any object that we believe to be good or bad
in the narrow sense. How plausible is this idea?
Even if we accept that our evaluations and motivations follow our believes
about what is good or bad (which is often at least not explicitly so), we
would have to assume that we instinctively apply the difference between a
good or bad thing in the narrow sense and the rest (i.e. indifferent, or
possibly good or bad things in the wider sense). As an indicator that this
assumption actually holds, we might treat evidence about ways in which our
evaluations, and therefore responses, change once we change our beliefs
regarding what's good and bad, and in which sense it is. So from making some
progress in changing our responses there, we might infer that there is
something in this view. (But it seems a long shot.)
__
[1] The terminology is well sorted out by Margaret Graver in her
Stoicism and Emotion, Chicago: University of Chicago Press 2007,
35-60.
[2] Again, the terminology is taken from Graver, op. cit.,
46-47.
The term 'uncompromising evaluation' has been suggested by Terence Irwin,
see "Stoic inhumanity", in: Juha Sihvola and Troels Engberg-Pedersen (eds.),
The emotions in hellenistic philosophy, Dordrecht: Kluwer 1998,
219-242. My explanation so far relies on the discussion in Graver, but I'll
soon turn to Irwin's further analysis of this interesting aspect of Stoic
philosophy.
[3] Graver, op. cit., 47-48.
[4] Irwin, op. cit., 225.
