Emotion, Evaluation, Desire, Behavior and Goals: a Eudaimonistic View

Abstract

In this essay I examine the conceptual relation between emotions and their corresponding evaluations, desires, behavior and goals. Such conceptual relation is of the utmost importance in order to account for the unity or oneness of emotion, for if the different aspects of emotion are linked conceptually, then to have one such aspect would imply having all the others. After I discuss how emotions are related to their corresponding evaluations, desires and behavior, I show how each aspect of emotion is related to the others. Next, I argue that indeed there is such conceptual or quasi-conceptual relation between emotions, evaluations and conative or desiderative states (i.e. desires or wishes). However, while intentional behavior in terms of actions is conceptually related to some emotions, other times the emotion-specific desires are either non-forward-looking, and thus do not give rise to intentional actions, or they may be overridden, and the resultant behavioral manifestations are not conceptually related to emotion. Next, I argue that emotions are eudaimonistic, in the sense that they are essentially concerned with important goals, concerns and attachments that comprise our conception of the good life and well-being. Finally, I examine the conceptual relation between emotions and their respective goals, and argue that while emotions logically presuppose such goals, the reverse is not the case.

1 Introduction

In this essay I argue that emotions are eudaimonistic in the sense that they are essentially concerned with important goals, concerns and attachments that comprise our conception of the good life and well-being. First, I examine the conceptual relation between emotions¹ and their corresponding evaluations, desires, and behavior. Such conceptual relation is of the utmost importance in order to account for the unity or oneness of emotion, for if the different aspects of emotion are linked conceptually, then to have one such aspect would imply having all the others. After I discuss how emotions are related to their corresponding evaluations, desires and behavior, I show how each aspect of emotion is related to the others. Next, I argue that indeed there is such conceptual or quasi-conceptual relation between emotions, evaluations and conative or desiderative states (i.e. desires or wishes). However, while intentional behavior in terms of actions is conceptually related to some emotions, other times the emotion-specific desires are either non-forward-looking, and thus do not give rise to intentional actions, or they may be overridden, and the resultant behavioral manifestations are not conceptually related to emotion. Finally, I examine the relation between emotions and goals, and argue that while emotions presuppose certain eudaimonistic goals, the reverse is not the case. Having a goal does not conceptually require an emotion. Due to space limitations however, I will bypass a discussion of the affective aspects of emotion such as the bodily feelings and the relevant pleasures and pains, which I also think are essential to the emotions.²

2 Emotion and Evaluation (or Appraisal)

“Appraisal” is the preferred term in psychology, while philosophers use mostly the term “evaluation”.

Although scholars of emotion disagree about the exact definition of emotion, most believe that an emotion requires some type of evaluation or appraisal.^{4 , 5} While in the last two decades appraisal theory suffered some drawbacks largely because of the findings of the neuroscientist J. LeDoux (1996) and his view that one can have an emotion without frontal cortical involvement, several critical reviews and new experiments have cast doubt on LeDoux’s findings. (See Clore and Ortony 2008; and Salmela 2014). Although there is no consensus regarding the nature of such evaluation(s) or appraisal(s), I think that it is safe to assume that according to most scholars, an evaluation identifies the emotion as the kind of emotion it is, and provides its intentionality. For instance, when one is afraid, one evaluates the situation as being dangerous, when one is angry, one evaluates the situation as being insulting/unjust, when one is happy, one evaluates the situation as being wonderful, etc.

Elsewhere I have argued that such an evaluation may be either cognitive, such as an evaluative belief, judgement, or thought, or non-cognitive, such as a mental picture, which lacks a conceptual content.⁶ For instance, we can imagine a case where a person, who generally likes spiders to be afraid of a particular spider because she believes, judges or thinks that it is poisonous or dangerous. On the other hand, a person may know that the little spider in front of him is not dangerous, and yet be afraid. Certainly, most people would not accuse such person as being irrational! In such cases, a non-conceptual, less-epistemically-committed state, such as a mental picture, can provide the requisite evaluative intentional state of emotion.⁷ Although mental pictures are closer to imagination than to belief and propositional thought, they are also representational (i.e. intentional), and some can portray an actual object aptly. However, there is no truth/falsity dimension associated with such aptness. Whether a person sees a mental picture of X to be an apt portrait of Y is largely a personal matter. Therefore, mental pictures cannot be reduced to beliefs or similar cognitive, propositional states. There are no objective criteria by which one can judge or persuade another that a certain picture is an apt portrait of someone or something. One cannot take someone else’s word for it, as one might in order to believe a proposition. Similarly, one cannot be said to have made a mistake, if one does not see a picture as an apt portrait of someone or something. Yet, one can be said to be mistaken, if one forms a false belief about someone or something. On the other hand, such mental pictures are not instances of free imagination either, as there must be a belief or similar state that functions as an external constraint, which enables a person to picture or see X as Y. However, the constraints that generate the picturing do not need to be objectively true.

So, in the arachnophobic’s case, although the subject does not believe that the spider is dangerous, he pictures or sees it as something genuinely dangerous, a black widow, perhaps. Seeing this picture of a genuinely dangerous spider as an apt portrait of the present harmless spider explains and identifies his fear. In addition, our arachnophobic most probably has beliefs in the “region” of dangerousness of spiders, such as “spiders are ugly and disgusting creatures”, “some spiders are dangerous”, “to be bitten by them is unpleasant”, “this spider resembles a black widow”, etc., which function as external constraints and conditions for picturing or seeing the present spider as a dangerous “black widow”. Had the arachnophobic none of these beliefs, he would not be able to see the picture of a dangerous spider as an apt portrait of the present spider.

So far we have seen cases of emotion where the evaluation is either cognitive (e.g. evaluative belief) or non-cognitive and non-conceptual (e.g. mental picture).⁸ But how will this view account for cases of emotion where the evaluation is unconscious, rapid and automatic?^{9 , 10} To this I would like to argue that even cognitive, propositional accounts of emotion do not require an explicit, conscious propositional analysis of the situation before the emotion arises. Indeed, Clore and Ortony note in their article “Appraisal theories: How cognition shapes affect into emotion”, “it is not clear that anyone is holding such a view”:

Seeing one’s groceries spill out surely would produce distress directly and without thought, but a propositional analysis does not suggest otherwise. Whatever cognitive activity is involved is implicit. A number of models... have addressed such issues, arguing that any requisite cognitive activity can be exceedingly rapid and without conscious thought (p. 631).

Thus, even those who endorse a cognitive account of appraisal, like Clore and Ortony, accept that cognitive appraisals can be rapid and automatic without necessarily involving conscious thought.

Our discussion of evaluation so far has shown the following: An emotion logically presupposes an evaluation (or appraisal). In adult, human emotions an evaluation is either cognitive, such as an evaluative belief, judgement, or thought, or non-cognitive and non-conceptual, such as a mental picture. And, even though an emotional evaluation can be conscious, more often than not it is non-conscious, rapid, and automatic.

Yet, we have not considered in our present discussion the possibility of the sufficiency relation between evaluation and emotion. Since there can be evaluations which are typical of emotion without having an emotion, there is no entailment relation from the evaluation to the emotion. But can we say that such cases are simply aberrant? I think that we cannot. Mr. Spock of Star Trek is pictured to be incapable of emotions and yet he is evidently capable of making evaluations. But even if one does not agree with this, it suffices for my argument that there are plenty of cases where an evaluation is not attended by an emotion.¹¹

3 Emotion and Desire

Most scholars of emotion seem to take it for granted that emotions are somehow related to desires, and thus they do not discuss them in great detail. Some notable exceptions are Kriegel (2016); Döring (2003); Goldie (2000a and 2000b); Green (1986 and 1992), Marks (1984), and Kenny (1963) among others.

I would like to argue that it is difficult to imagine someone to be afraid and not ever having the desire to avoid the danger, be in love and not ever having the desire to be with the beloved, or be angry and not ever having the desire to retaliate in some way. If a person says that he is afraid but he does not want to avoid the danger, we would have difficulty understanding what exactly he means when he says that he is afraid, and we expect some explanation. Does he not desire to avoid the danger because he is with his beloved and he wants to show that he is a brave man? Or does he believe that the “danger” is not a real danger, and thus his desire to overcome his unjustified fear, has overridden his desire to avoid the danger? Unless an explanation is given, it will be difficult to make sense of the person’s fear. The same is true with anger. If a person does not have any desire to retaliate, but still insists that she is angry, we are not going to make sense of her anger, unless, again, she gives us some explanation. Does she want to overcome her “negative” emotions and thinks that a way to do this is by eliminating her desires? Or, does she think that desiring to hit back against the offender is wasting too much time on someone who is not really worth it? Likewise with romantic love. If a person has never desired being with her beloved, then we would fail to understand her emotion, and again, we would demand some type of explanation. Did the beloved demand that she should not desire him and because she loves him does not want to go against his wish? Does she think that by desiring him, she is actually harming the beloved, and she loves him too much to ever harm him? So, it is safe to assume that at least the emotions of fear, anger and love require some type of desire that is conceptually presupposed by the emotion.

But are there any desires required for grief, which involves a negative evaluation over the loss of something or someone? Since the object of grief involves something that already happened in the past, it is difficult to assess the desire involved in such emotion, since desires are mostly forward-looking. Of course, one could say that a person still has a desire, namely the desire to change the past. However, it seems that in this case the term “wish” would better describe the person’s conative state, since wishes, unlike desires, are not necessarily forward-looking.¹² For one may wish one were living in fifth Century BCE Athens, but one does not have the desire to do so, because one knows that such a desire is unrealizable. In this light then, we can say that in the case of grief there is still a conative state involved, i.e. a wish that the loss had never happened. If one is grieving over the death of a beloved, one must have the wish that the beloved had never died. Indeed, without such a wish, one’s grief would not make much sense, and again, an explanation would be required. So, even in a backward-looking emotion such as grief, a conative state such as wish is required in order to make sense of the emotion. I think this discussion suffices for now to show that in the absence of defeating conditions¹³ there is a conceptual connection between emotions and conative states, in the sense that an emotion conceptually presupposes some type of conative state (i.e. desire or wish).¹⁴

4 Emotion and Behavior

But how does behavior feature in this conceptual scheme? Most scholars of emotion do not discuss the relation between behavior and emotion in a great detail. Some noteworthy exceptions are Mulligan and Scherer (2012); Döring (2003); Goldie (2000a, b); Green (1986 and 1992); DeSousa (1987); Frijda (1986); Marks (1984), and Kenny (1963).

I would like to argue that in the absence of defeating conditions we lose our grip on the occurrence of a given emotion, if the expressed behavior is not in accord with such an emotion. If a person says that she is afraid of dark alleys and yet we see her passing through dark alleys all the time while she could avoid it, we have hard time believing that such person is truly afraid of dark alleys. Unless an explanation is given such as “she walks through dark alleys because her therapist told her that this is the only way to overcome her fear”, or “she became fatalistic about her life”, we will be at a loss in understanding what such person means when she says that she is afraid of dark alleys. For in such a case the conceptual connection between fear and behavior would have been lost. The same goes for other emotions. If I never show any sign of joy when I receive the news that my paper was accepted for publication, one would be at a loss to ascribe to me “joy that my paper was accepted.” Or alternatively, if I don’t show any signs of anger when I receive an insult, then people would be right to question whether I am angry. Hence, I would like to assert the following: In the absence of defeating conditions¹⁵ emotions conceptually require behavior.

Yet, in our discussion we have not indicated whether behavior should be understood as intentional actions, or simply as non-intentional bodily expressions and movements. Although I am addressing this in more detail in later sections, for now I would like to argue that if we understand behavior reductively, i.e. as simply non-intentional movements, expressions and sounds, then the conceptual connection between behavior and emotion will be lost. For if behavior is understood reductively, then to be afraid of the snake would be analyzed in terms of tense muscles, accelerated heartbeat, one’s limbs moving away from the snake, one’s mouth making loud sounds, and so on. However, such behavior is not conceptually related to fear, for it is a contingent fact about our species that we express fear in such a way. On the other hand, if we understand behavior as an intentional action, then we can say that one is fleeing from the snake because one evaluated it as being dangerous and has the intention/desire to avoid the danger. So, I would like to argue that the behavior required for the emotion is in the form of intentional actions. If we interpret behavior as intentional actions, then intentions/desires, and evaluations become part of the behavior.

However, one may object, that even behavior qua action is not necessary for emotion. One may be afraid of the snake that is behind glass, but in order to avoid being thought of as being irrational by others, she tries to conceal her feelings and not show any signs of fear. Or similarly, a person may be angry at his friend for not having called him, but because he is preoccupied with another emotion, say grief over a loved one’s misfortune, he may not exhibit the usual signs of anger. In these cases a person feels the emotion, but he or she does not express it behaviorally in the expected ways. Indeed, cases like these led some scholars to believe that behavior is not essential to emotions.¹⁶ For instance, Mulligan and Scherer do not consider behavior as a necessary condition for a working definition of emotion (Mulligan and Scherer, p. 352). However, I would like to maintain that these are aberrant cases whereby the existence of defeating conditions does not allow the connection between emotions and behavior. Later we shall see what is the nature of such defeating conditions, and why some emotions do not exhibit the expected behavior.

5 The Unity of Emotion

So far we have found that emotions require evaluations, conative states, and behavior qua actions (in the absence of defeating conditions). However, a question arises about how all these different aspects of emotion are related together to form the emotion. Elsewhere I have contended that in order to account for the unity or oneness of emotion the various aspects of emotion cannot be merely empirically or “externally” related, and argued against the prevalent causal view,¹⁷ according to which an emotion is a conglomeration of different parts mediated by efficient causation (See Adamos, 2002 and 2007) . Some proponents of this view are Schachter and Singer (1962); Lyons (1980), and Goldstein (2002). So, bypassing a detailed discussion let me at least roughly sketch the problems facing such view.

5.1 The Problem of Accidental Causal Connections

Suppose I have just learned that my paper has been accepted for publication and I evaluate this positively (E) in part because I can now take time out from research and engage in my favorite indulgence – dancing. The causal result is (B) – my jumping up and down, and enjoying this exercise, a smile breaks out on my face along with various bodily and behavioral expressions related to joy (e.g. jumping up and down). E caused B, but this is not an instance of joy that my paper was accepted (though there may be joy in dancing). That is, the good news about the acceptance of my paper is a causal factor in my dancing, and subsequently in my joy. Yet, I am taking joy in dancing, and not that my paper has been accepted. The general problem is that insofar as causal chains from Es to Bs can be expected to have many links, the causal theorist faces the difficulty of differentiating fortuitous and non-fortuitous chains between Es and Bs.

One may object that this is an aberrant way in which an emotion may arise. However, my point is that the causal view fails to distinguish an emotion from a mere conglomeration of different parts that happen to be causally related. Thus, it is unable to give an account of unity or oneness of emotion even in cases where the causal connection is not fortuitous.¹⁸

5.2 The Problem of the First Person Authority Over One’s Emotion

We seem to have a special first person authority over our emotions. Although such authority may not occur over all our emotions, it exists in most standard cases. Surely such authority is not infallible, but it does have a privileged status. Imagine that I see suddenly a snake in my room and I am afraid. Given that the causal view is supposed to analyze the meaning of fear in terms of the causal relations involved, it must admit that since I have a first person authority over my fear of the snake, I must also have a first person authority over the causal relations between the different aspects of my fear. But surely I do not have any special first person authority over such causal chain. Although one can object that we are routinely wrong in identifying our emotions, my point is that it can still be shown that I can know without investigation that I am afraid of the snake without having to determine the causal relations involved in my fear.

The reason that the causal view fails to explain the unity of emotion is that it sees the relation between the different aspects of emotion as empirical and external. What all this points to is that such a relation should instead be internal, in the sense of “conceptual presupposition”. A conceptual relation between the different aspects of emotion would provide an explanation of the unity or oneness of emotion, for in such case one aspect will imply all other aspects. In what follows I will examine whether there is such a conceptual link between the different aspects of emotion.

6 Desire and Evaluation

Let us consider a case of anger. Most scholars agree that anger presupposes the evaluation that someone has wronged us. How is this desire related to such an evaluation? Of course, a mere desire Jill has to harm John need not presuppose any evaluation about John having wronged Jill; one could just take pleasure in the suffering of others. However, once we describe Jill’s desire as a desire for revenge or a desire to get back at him, this presupposes that Jill thinks or imagines, etc. that John has mistreated her in some way or other. It is contradictory or simply incoherent to say: Jill desires to get back at John, but she does not at all evaluate John’s action negatively. So, I would like to claim the following: A desire to x conceptually presupposes a positive evaluation of x.

Yet, while there is a presuppositional, conceptual relation from the desire to the evaluation, the reverse is not the case. That is, it does not appear that evaluations are sufficient for having a corresponding desire. One might evaluate something positively without desiring it; or, alternatively, one might evaluate something negatively and desire it. For instance, I could evaluate running five miles a day positively because I think that it is a superb exercise for my health, but do not have the desire to do it. Or, I might evaluate negatively eating a chocolate pudding and yet fail to have the corresponding desire of avoiding it.

Let us first consider why I may not desire to run five miles a day despite evaluating such an exercise positively. It seems that one factor is that the term “desire” can be understood in a narrow and in a wide sense. That is, in the narrow sense, “desire” means something like “craving” or “itching” to do something. So, if we understand desire in the narrow sense, we will be able to explain my not wanting to run five miles a day, by saying that I do not have the desire in the narrow sense to do it. Under this interpretation, even if I decide to run five miles a day for my health, I could still say "I hate it; I do not want to do it"! In this narrow sense of desire, we cannot expect any kind of sufficiency relation from evaluation to desire.

However, we can also understand “desire” in a wide sense. According to this sense, “desiring x” means something of the following: a state, which in conjunction with evaluative intentional states, leads to a decision to do x, and results in disappointment or discontent if x is not performed. If I do evaluate running five miles positively, I might be expected to be somewhat discontented or disappointed with myself if I do not do it. In this wide sense of desire (which from now on I will indicate by “w-desire”), one w-desires the means to attain ends, which are themselves desired (in a more narrow sense).

But there is another complication regarding desires: sometimes we have a variety of competing desires and even if we w-desire x, all things considered, we may desire or w-desire to do something incompatible with x instead. For instance, it could be that I despise looking sweaty and tired, and this desire (i.e. to avoid looking sweaty and tired), may override my w-desire to run five miles a day. In that case my w-desire to run five miles a day with the evaluation "it is good to run five miles a day", will not result in my deciding to run, and I may feel no disappointment with myself for not running.

In order to overcome the difficulty it may be useful to borrow W. D. Ross’ notion of prima facie duties and apply it to desires (Ross 1930, pp. 17–46). So, I would like to propose the term “prima facie desire”, which may be defined as follows:

One has a prima facie desire to do x, if doing x would be what one w-desires (or desires), all things considered, and if it were not for competing desires or w-desires.

Seen in this light then, we can surmise that if one evaluates doing x positively, one has at least a prima facie desire to do x. But now our earlier problematic cases will not be counterexamples to this suggestion. For, if I positively evaluate running five miles a day, I would at least have the prima facie desire to run five miles a day. Under this reading then, my prima facie desire to run five miles a day is overridden by my competing desire or w-desire not to look sweaty or tired.

Now, if one is internalist like Hume and believes that evaluations entail desires, one would agree that there is a conceptual truth that evaluations entail prima facie desires. On the other hand, those who deny the internalist position might be meaning to deny that evaluations entail desires in the narrow sense, but not deny that evaluations entail prima facie desires. Be this as it may, can we make any sense of someone evaluating doing x positively, and yet failing to have any prima facie desire to do x? But, what would my positive evaluation of running five miles a day amount to, if there is no prima facie desire to run five miles? This would mean that even if I do not have any competing desires or w-desires (such as not looking sweaty, etc), I would still not w-desire to run five miles all things considered. But, then, how are we to explain this situation?

In what follows I will use the term “quasi-sufficiency” to refer to a condition that would be sufficient for the occurrence of a state or an event, if there were no overriding conditions. Therefore, in order to take into account the defeating conditions of the overriding desires, we can conclude with the following: Desiring or w-desiring to do x logically presupposes a positive evaluation of x, and such evaluation is conceptually quasi-sufficient for the w-desire to do x.

7 Non-Forward-Looking Evaluations and Desires

Thus far, I have only considered “forward-looking evaluations”. However, there are also non-forward-looking evaluations where one evaluates something that is already done. For instance, I may positively evaluate my having just won the lottery. How can such evaluation be quasi-sufficient for any desire (in both senses), since desires are typically forward-looking? Well, one might say that when buying the ticket I had the w-desire to win the lottery. So, in this case I can explain my positive evaluation by referring to a prior w-desire for what is now positively evaluated.

But it seems that some times when I evaluate something that is already done, I do not have any prior desires. For example, if I unexpectedly receive a legacy and the thought of receiving a legacy never crossed my mind, it would be strange to claim that I desired to receive a legacy. How can we explain such cases? Well, we can say that in this case I was disposed to desire a legacy: if the thought had crossed my mind, I would have desired it.

Our previous conclusion relating to forward-looking evaluations can now be generalized: Actually or dispositionally w-desiring (or desiring) x, logically presupposes a positive evaluation of x, and such evaluation is conceptually quasi-sufficient for w-desiring x (or being disposed to w-desire x in case of non-forward-looking evaluations).

8 Forward-Looking Desires and Behavior

What we have found to be true of the relation between evaluations and w-desires will be mirrored in the relation between w-desires and behavior, as long as we are dealing with forward-looking w-desires, and “behavior” is understood as intentional actions. For instance, my action of running away from the dangerous snake presupposes my intention/w-desire to get away from the snake. Hence, I would like to assert the following: An action x logically presupposes the intention to do x and, in this sense, the w-desire to do x is logically presupposed. Of course, one may go to the dentist to have a cavity filled without desiring (in the narrow sense) to go to the dentist. But one does w-desire to go to the dentist.

In our previous discussion of the relation between emotion and behavior, I argued that emotion necessitates behavior (qua actions) in the absence of defeating conditions. Now I would like to suggest that our recent discussion of prima facie desires and evaluations enables us to clarify the nature of the defeating conditions that prevent a w-desire from culminating into an action. It seems that they fall into two categories:

• (a) The w-desire may have been only a prima facie desire, which was overridden by a competing desire or w-desire. In this light, my current desire for a cigarette may override my w-desire to stop smoking.

And,

• (b) Some cases may be classified as “behavior incapacitation”. I might for instance, attempt to fix the car, and yet I do not succeed. Although there is the behavior of trying to fix the car, there is some kind of incapacitation that prevents me from successfully completing the intended action. Or, I might w-desire to talk to my parents that I have not spoken to for years, but be unable to do so. In this case one has no means of satisfying one’s w-desire.

Therefore, we can conclude that in the absence of overriding desires or w-desires, and behavioral incapacitation, it seems to be a conceptual truth that the w-desire to do x would lead to the decision to do x. Hence, we can assert the following: Forward-looking w-desires are conceptually quasi-sufficient for intentional behavior qua actions.

9 Non-Forward-Looking Desires and Behavior

Yet, while forward-looking desires are conceptually quasi-sufficient for actions and are logically presupposed by them, dispositional desires based on non-forward-looking evaluations of something already done do not relate to actions or behavior in the same way. For, if I positively evaluate my unexpected legacy, even if I were disposed to w-desire it (and if the thought had crossed my mind), still such a w-desire would not lead to action, since no actual w-desire is present. Now, for negatively evaluated situations, one can appeal to the conative state of wishing it had not been so. For example, in the case of the negative evaluation of a hurricane destroying my house, one can appeal to the wish that such a thing had not happened. But wishes contrast to desires in the sense that wishes, unlike desires, which are forward-looking, are mostly about things already done, in which case one is not in a position to do anything to make one’s wish come true. Of course, non-forward-looking evaluations may be the basis for future desires. However, since the evaluations of things done engender no desires, no intentional action is to be expected. Thus, we can conclude with the following: Non-forward-looking conative states such as wishes do not engender behavior qua actions, and thus there is no conceptual relation between wishes and behavior/actions.

10 Emotions, Evaluations, Desires, and Behavior

Having seen the conceptual relations that exist between evaluations, w-desires, and behavior qua actions, how do occurrent emotions fit to this conceptual scheme? Such relations can be diagrammed as following:

In the above diagram we see that emotions are conceptually quasi-sufficient for w-desires and actions as long as the presupposed evaluations are forward-looking. Seen in this light, fear would presuppose the negative evaluation that the situation is dangerous, and this evaluation would be quasi-sufficient for the prima facie w-desire to avoid the danger and the action of escaping the danger.

However, in some cases the evaluation presupposed by the emotion is non-forward-looking: the joy in receiving an unexpected legacy, the grief over the death of a loved one, the anger over the insult received. In this sense, emotions could be seen as emotional reactions to what has already happened. But then the question arises of how such emotions relate to desires and actions. There is no uniform answer.

Let us first examine anger. The prima facie desire in anger is the desire (in the narrow sense) to retaliate. But usually this desire is overridden by the competing w-desire to abide by the norms, which advise against acting on our desire. However, the desire to retaliate does not disappear; it still remains and one often settles for yelling at the wrongdoer, or saying nasty things to her, etc., since these, too, are forms of retaliation. So, we can now assert the following: Anger presupposes the prima facie desire to retaliate, which is quasi-sufficient for the corresponding action of retaliation and it is logically presupposed by it.

However, it seems that there is more involved in the action out of anger, which fulfills the desire to retaliate. Because the prima facie desire to retaliate is often frustrated (but not canceled), even yelling at the wrongdoer could be overridden by other w-desires, and we usually resort to ineffective substitutes, such as banging on the table, or clenching our fist, etc., which are also considered behavioral expressions of anger.¹⁹ Now, insofar as banging on the table is an intentional action, there is also a desire (in the form of an intention) to bang on the table. Why then not to say that this desire is also required for anger? I think that this is to miscast the matter. For, it is not that anger entails the desire to bang on the table, or to clench one’s fist, or to smash the vase, etc. There is only one central desire, the desire to retaliate, and one’s behavior relates to this desire either as its fulfillment, or, by assuming ineffective substitutes for what was desired. Now, there is the possibility that even ineffective substitutes such as banging on the table can be overridden by other w-desires, and we may only be left with a flushed face or tensed body.²⁰ However, none of these behavioral expressions are conceptually related to anger, as it is a contingent fact about our species that we express anger in these ways. So, we can now assert the following: When the prima facie desire of anger (i.e. the desire to retaliate) is overridden, the resulting behavioral manifestations and indications neither logically presuppose, nor are they made conceptually quasi-sufficient by the prima facie desire.

When we turn to reactive emotions like joy and grief, the relation to behavior becomes even more problematic, since such emotions do not seem to entail any prima facie desires. Typical behavior attached to grief is crying, and to joy “jumping up and down”. Now, jumping up and down and crying do not seem to be conceptually linked to their respective emotions. Unlike the case of anger, whose behavioral manifestations could be seen to be derivative in one form or another from the central desire to retaliate, with joy and grief we can only point to typical actions or physical reactions (jumping up and down or crying) and their variations, such as a smile or a sad look. But, although it seems true that it would be difficult to ascribe joy or grief in the absence of any inclination to such actions or behavior, we cannot find a conceptual connection between the emotion and the behavior, since nothing in the intentional object of the emotion (having gotten the legacy, or the death of a loved one) makes contact with such behavior. It seems to be a contingent fact about us, humans, that we express joy and grief in these ways. Hence, I would like to suggest the following: Since the backward-looking emotions of joy and grief do not engender forward-looking w-desires, the resulting behavioral manifestations and indications are not conceptually related to these emotions.

11 Emotions and Goals: A Eudaimonistic view

We have not discussed yet the role of emotions in one’s overall goals, projects and attachments. I would like to now assert that emotions are eudaimonistic.²¹ That is, they feature in and are concerned with what a person takes to be his or her idea of the good life and general well-being. My own eudaimonistic account follows very closely Nussbaum’s view of eudaimonism, although there are many important differences between our overall account of emotions, which I will not discuss here. Nussbaum states that emotions “view the world from the point of view of my own scheme of goals and projects, the things to which I attach value in a conception of what it is for me to live well (Nussbaum 2001, p. 49). Emotions enable us to focus on or draw our attention to these goals and attachments, and make us see the world from a personal point of view. Nussbaum makes the point forcefully:

Emotions contain an ineliminable reference to me, to the fact that it is my scheme of goals and projects. They see the world from my point of view. The fact that it is my mother [who died] is not simply a fact like any other fact about the world … It is not just the fact that Betty Craven has died. It’s the fact that Betty Craven is my mother. In short, the evaluations associated with emotions are evaluations from my perspective, not from some impartial perspective; they contain an ineliminable reference to the self [her italics] (p. 52).

Nussbaum makes a powerful argument here for the self-referential nature of emotion, and in particular, the very personal nature of grief for the loss of someone who was very important in her conception of eudaimonistic life.

However, the idea of the good life is different for each person, even if it contains several similar goals. For instance, a part of the good life is to ensure that a person has taken care of his or her biological needs for survival, which is a general goal that applies to all humans.²² Yet, my particular goal of having this paper published, living in a particular neighborhood, or having a particular friend may be peculiar to me and my circumstances.

By alerting us to our goals and attachments, emotions also reveal what Nussbaum calls our “neediness”. We are needy creatures that require other things and people in order to survive and live the good life. Nussbaum puts this eloquently: “…[W]e need others to survive and flourish, but do not at all control their movements” (p. 13). As much as I would like to be self-sufficient, I still have biological needs such as food and shelter, freedom from harm, as well as other psychological and personal needs such as freedom of movement, people to love and be loved by, financial security, a job, philosophy, etc.²³ If I am lucky enough to have such things, then I am prone to feel fear when they are threatened, anger when they are treated unjustly, joy when they are realized, and grief when they are lost.²⁴ Such emotions would never occur, had these attachments not transpired. Thus, because other things and people are essential parts in my conception of eudaimonistic life, I (consciously or unconsciously) want to make sure they are protected, maintained, and, when it comes to other people and creatures, as much as it is within my power, help them achieve their own overall eudaimonistic goals.^{25 , 26} Indeed, because such things and people are very important to my life, a crucial eudaimonistic goal of my life is to protect them and ensure that they continue to be in my life, all things considered.²⁷

Most people realize that their eudaimonistic goals should be compatible with each other and the overall goal of achieving the good life. Thus, we try as much as possible to achieve such compatibility. For instance, if my goal is to continue my relationship with Julia, I may choose not to make a big fuss about her not calling me yesterday. Also, if my goal is not to be thought of as a coward in front of my students, I may choose not to scream in fear when I see the cockroach. These goals, in turn, should be compatible with the overall goal of achieving (or trying to achieve as much as possible) the general goal of my well-being.

Yet, at times our goals are not compatible at all. One may know that having an affair with a married man is not compatible with one’s overall conception of the good life, and yet do it. Or, a person may recognize that a certain individual uses him as a means to her ends and still continue with the relationship, even though he knows that it does not accord with his overall eudaimonistic goals. In such cases a tension arises between the person’s actions and his eudaimonistic goals, which is perhaps indicative of the messiness of human lives. Given that we, humans, are not perfectly rational beings, at times we will do things that do not accord with our own idea of practical rationality, and thus, our actions do not always conform to our goals or aspirations.

So far I have maintained that emotions are concerned with important goals that make up our conception of the eudaimonistic life. Now I would like to discuss a bit more the issue of importance in the list of the eudaimonistic goals. Usually, we view some goals as being more important than others. For instance, the “harm” of a mosquito-bite might not engender an emotion at all because I don’t consider such harm as a very important threat to my goal of protecting myself from harm. On the other hand, if I believe that the mosquito carries the West Nile virus, then I may feel fear because I perceive that this endangers my life, and protecting it from harm is an important goal of mine. Likewise, the loss of my beautiful pen might upset me because it prevents me from using it. However, it will not engender grief because having this pen in my life does not feature as salient among my eudaimonistic goals. On the other hand, the loss of someone dear to me, will engender grief, because it alerts me to the dissolution of the vital goal of having such person in my life.²⁸ This difference in importance of the various goals in one’s conception of the good life also accounts for the different intensity of emotions. When the object of the emotion is related to an important goal, the emotion is very intense. On the other hand, if the object is not related to an important goal, then the emotion seems to be less intense.²⁹

The above discussion also shows that the evaluations that feature in the conceptual relations of the emotions are not simply neutral evaluations. Instead, they are either positive or negative because they relate to either a frustration or a satisfaction of a goal.³⁰ So when I am afraid, my evaluation that I am in a dangerous situation is negative because it is the frustration of the goal of “living without a threat to my life”. Alternatively, the evaluation in joy that something good has happened is positive because it is a realization of the goal of “being a subject of good things which advance my overall eudaimonistic goals”.

I believe that our discussion of the eudaimonistic nature of emotions also shows the reason why emotions do not by sheer chance alert us to their objects, and why they are conceptually related to some important “eudaimonistic” goal of ours. Several psychologists recognized this.³¹ For instance, Lazarus says that emotions are appraisals that enable a person to become aware of and focus on something important in his or her environment that is related to his/her personal goals (Lazarus 1991, p. 30). Emotions focus our attention on what is important, as we tend not to feel emotions about unimportant events. Moreover emotional appraisals show the significance of the event to one’s personal well-being. Indeed, for Lazarus an emotional reaction “tells us that an important value or goal has been engaged and is being harmed, placed at risk, or advanced (Lazarus 1991, pp. 6–7). For Lazarus the goals that a person (or an animal) has may be due to biology, society, personal, etc. They may be conscious or unconscious.

Like Lazarus, Oatley accepts that emotions are appraisals in relation to a person’s goals, but he adds that the appraisal is of some perceived change in the status of these goals. According to Oatley an emotion happens when there is a change and interruption to a “psychological tendency” or “a smoothly flowing action” (Oatley 1992, p. 46). Such change is very tightly connected to the probability of progress toward a particular goal (p. 46).

Also, Roseman understands emotions as “a coherent, integrated system of general-purpose coping strategies, guided by appraisal, for responding to situations of crisis and opportunity” (Roseman 2013, p. 141). Like Oatley, Roseman thinks that positive and negative emotions are concerned with change in the goals or motives of a person (p. 146).³² So, again we see that like Lazarus, and Oatley, Roseman believes that an emotion occurs in relation to the goals a person has.

I believe the above discussion suffices to support the claim that emotions are related to the eudaimonistic goals a person has. But how such relation features in the overall conceptual analysis of the emotions?

I would now like to argue that emotions are essentially related to their respective goals. For instance, my fear of the car that is advancing rapidly towards me alerts me to the threat of the important goal of self-preservation. If protecting my life were not a central goal of mine, I would not feel fear. Also, anger is not trivially connected to the goal of not being subject of injustices. My anger at my colleague for the insulting remark alerts me to the risk of the goal of not being subject of wrongs. If such goal were absent, then anger would not occur. Likewise, my joy that my paper was accepted for publication is closely related to my overall goal of being subject of good things, which advance my eudaimonistic goals. If such goals were absent, the emotions would not occur. Indeed, emotions function as alert mechanisms, which, as Lazarus has argued, either focus our attention to a cessation of an important goal (grief), or warn us that a goal is at risk (fear and anger), or simply alert us that a goal has been advanced (joy). So, we are now equipped to claim the following: Given the eudaimonistic nature of emotions, an emotion logically presupposes an overall aim or goal.

But are eudaimonistic goals sufficient for emotion? I think that they are not. My goal of not being a subject to dangers in certain cases might generate fear, but in others it may simply engender a set of evaluations, which together with their relevant desires will guide me away from the danger. The same is the case with other emotions. The goal of not being a subject of injustice might simply generate a certain evaluation that an offense has occurred, but it might not push me to seek revenge, but forgive the person instead.

Can we claim that such cases are simply aberrant and say instead that a eudaimonistic goal is quasi-sufficient for the emotion? I think that we cannot, as there are plenty of cases where such goals do not engender an emotion. We can imagine Mr. Spock of Star Trek having the goal of not being subject of injustices but not feeling anger. Also, we can imagine a circus trainer having the goal to protect herself from dangers, and yet not be afraid of the dangerous lion.

But how the overall goals, which are relevant to the emotion and the particular goals elicited by the desire and the intentional action are related? For instance, in fear the overall goal is protecting the self from danger or harm. However, the activated goal elicited by the desire is escaping the danger. Of course, one would expect that the two types of goals to be functionally coherent. Yet, what does the goal of not being a subject of a threat has to do with the goal of escaping danger? It seems that the activated goal of escaping the danger is conceptually related to the overall goal in fear of not being an object of a threat or harm, because once this activated goal of escaping is satisfied, the overall goal of not being object of threat will also be satisfied. In a way the activated goal of escaping is an auxiliary goal to the overall goal of fear of protecting survival. Thus, in fear the overall eudaimonistic goal of not being subject of harm conceptually presupposes the goal of escaping danger. However, the particular acts of escaping such danger are not so conceptually presupposed, as it is a peculiar fact of our physiology that we, humans, express escaping danger in such ways. Thus, one may escape the danger by screaming loudly, or hiding, or running away, or playing dead, etc.

Likewise with anger. The overall goal of anger is not being a subject to some wrong, and the relevant desire and activated goal is retaliation. How such goal is related to the overall goal of anger “not being subject of some wrong”? I would like to claim that if the activated goal of retaliation is satisfied, then at least in one’s mind, the perpetrator of the injustice who, through the retaliation, will be “punished” for the wrong action, is less likely to commit such an injustice again towards one. The act of retaliation is supposed to show the perpetrator (as well as the community) that there will be repercussions should he/she (or anyone else for that matter) repeat the wrong. This in turn will satisfy (even if briefly at times) the overall goal in anger, i.e. ‘not being object of injustices”. So, in anger the overall goal of not being object of injustices conceptually presupposes the activated goal of retaliation. However, as our previous discussion of anger has shown, the actual behavior of retaliation might not occur because of other competing desires or w-desires, in which case we are left with a flushed face or a clenched fist neither of which are conceptually presupposed by the anger. And, although the activated goal of retaliation is logically presupposed by the overall goal of not being subject of injustices, the particular acts of retaliation and their respective goals such as hitting back, screaming at the perpetrator, going to the police, talking to the perpetrator’s boss, etc., are not so presupposed.

As we have seen, the backward-looking emotions of joy and grief do not engender forward-looking, action-oriented desires, and thus, they do not involve particular action-related, activated goals. The physical reactions of crying in grief and jumping up and down in joy, do not have an essential connection to the frustration of the goal of having certain people or things in my life, or to the realization of the goal of being a subject to good happenings that advance my other overall goals, respectively. It simply happens that we, humans, express grief and joy in such ways. Therefore the destruction of the eudaimonistic goal of having certain people or things in my life is not conceptually connected to crying or similar behavior in grief. Similarly, the overall eudaimonistic goal in joy of being subject of good happenings is not conceptually related to the usual activated goals of jumping up and down and smiling, since it is a contingent fact about us humans that we express joy in such ways. In fact, the eudaimonistic, overall goals related to grief and joy in a way explain why such emotions do not employ forward looking desires, or particular intentional actions. Since grief alerts us to a destruction of a goal, there is not much we can do to change this. Unfortunately, the destruction is total and irreversible. As for joy, the overall goal of being subject to good happenings that advance my other eudaimonistic goals has already been achieved, so to speak. No action or activated goal will enhance this particular joy and its overall goal. Of course, this joy may become more intense if I find out that my paper has also won the best paper award. Also, my present joy may lead me to pursue other similar endeavors, try to publish another paper, say. However, it will not be joy that this paper was accepted.

The above discussion of the eudaimonistic goals of emotion, also explains the motivational nature of emotions. The reason that certain emotions move us to act is because in the end we would like to protect, maintain, or alter important goals that fit in our conception of the good life. Given that goals are motives and reasons for action and that emotions conceptually involve certain eudaimonistic goals, they, through their respective desires and evaluations motivate us to act in certain ways that will ensure that our conception of the good life is satisfied.³³

12 Conclusion

Given our analysis of evaluations, desires, behavior, and eudaimonistic goals, we are now equipped to say that with respect to forward-looking emotions, as well as some backward-looking emotions, an account of unity of emotion has emerged, which would avoid the problems of fortuitous connection and first person authority that afflicted the causal account. Since having one aspect of emotion implies (or quasi-implies) all others, there is no room for an accidental connection of the “antecedent” and “consequent” parts of the causal relation. Also, since the conceptual unity of emotion does not allow for one to have a first-person-authority over one aspect of emotion and not have similar authority over all other aspects, it does not face the first person authority problem that afflicts the causal view.

However, since the backward-looking emotions of joy and grief do not engender forward-looking desires and active goals, we have not found a conceptual connection with behavior, and thus, no account of unity has emerged for these emotions. For as we have seen, the tears of grief, as well as the smile of joy cannot find a place in the web of conceptual relations; yet, they are surely central to grief and joy. How such behavioral expressions can be incorporated under the umbrella of conceptual unity will be the subject of another essay.³⁴