Jump to ContentJump to Main Navigation
The Retreat of Reason$

Ingmar Persson

Print publication date: 2005

Print ISBN-13: 9780199276905

Published to Oxford Scholarship Online: February 2006

DOI: 10.1093/0199276900.001.0001



The Retreat of Reason

Ingmar Persson (Contributor Webpage)

Oxford University Press

Abstract and Keywords

The chief purpose of this chapter is to provide a classification of types of emotion which will be of use for the discussion of responsibility in part V. The classes of emotions which are of particular relevance for this discussion are ‘agent-oriented’ emotions, such as anger, gratitude, remorse, and feelings of guilt, and ‘comparative’ emotions, like pride, shame, admiration and contempt. These emotions all involve taking someone to be an original or ultimate cause of something in an epistemic sense of a cause which is not in its turn taken to be caused by something else. It will be seen that this link is something these emotions share with the concept of desert.

Keywords:   nger, admiration, contempt, gratitude, guilt, ultimate cause, pride, responsibility, shame

IN this chapter I shall put forward a classification of emotions. (It could be set beside Roberts's more extensive analysis (2003: ch. 3) which unfortunately was published too late for me to be able to benefit much from it.) As I said at the end of the last chapter, my classification is designed to serve the discussion of the rationality of desert- and responsibility-related emotions in Part V. I do not deny that there might be other purposes for which other ways of dividing emotions are more useful.

As pointed out in Chapter 5, emotions can differ to the extent of being given different names in everyday language without their consisting in different kinds of felt patterns. This happens when an emotion is a species of another emotion in the sense that the content of the former is a specification of the content of the latter, as in the case of indignation and anger, etc. It also happens when two emotions are distinguished on grounds of differences in their strength or intensity, that is, differences in respect of the strength or intensity of the felt pattern. In this way, fear differs from terror,horror, and dread,anger from fury and rage,indignation from outrage,surprise from astonishment, puzzlement from amazement,gladness (or joy) from delight,admiration from reverence, and so on. Assuming a contrast between differences in degree and differences in kind, I shall not count these differences as differences of kind in the felt patterns of bodily changes. So, along with cases in which one emotion is a species of another, cases in which emotions differ merely with respect to intensity will not here be counted as different kinds of emotion in the relevant, basic way.

Although emotions are to be classified as being of distinct kinds if and only if they consist in different kinds of felt somatic pattern, it would be unwise to try to elucidate this classification by scrutinizing merely these felt patterns. For it is notoriously tricky to describe bodily sensations (and, for that matter, sensations and sense-impressions in general), and without any proper equipment for measuring the physiological processes of which the sensations are sensations, these processes cannot be precisely recorded, either. Fortunately, it is logically necessary that if a felt pattern is to constitute an emotion, it must by caused by a propositional thought which is typical for that kind of pattern. As a consequence, it is reasonable to expect that a marked difference as regards (p.80) the effect—that is, the felt pattern—reflects a difference in the cause. Conversely, differences in the structure of the contents that emotions take should provide clues to differences in the resulting felt patterns. Since differences in the former respect are easier to describe, I shall approach the latter via the former.

Factive and Epistemic Emotions


(1) A is glad that p (e.g. that he has won on the Pools)


(2) A hopes that p.

(1) is true only if A is convinced or certain that p. Moreover, (1) entails that it is true that p. In fact, it can plausibly be argued that this statement entails that A knows that p (see Gordon, 1987: 40–3). Needless to say, this does not mean that gladness is only possible if there is knowledge; it merely means that the content of gladness is properly expressed in the manner of (1) only if there is knowledge. Somebody who is of the opinion that not-p would have to say something like ‘A is glad because he is convinced that p’. If A's attitude to the truth of p is weaker, the correct report could be ‘A is glad because he thinks it is probable or possible that p’.

A lot of positive emotions behave in these respects like being glad: being happy, pleased, overjoyed, delighted, and relieved. But not the emotion of hope. A's hoping that p does not entail that p is the case and, more significantly, it is incompatible with his being certain that p. It requires A to think that there is some probability that p, that given his body of evidence it is not excluded that p. But this (subjective) probability must not amount to certainty; that not-p must be assigned some probability. Gordon calls emotions like hope epistemic and ones like gladness factive (1987: 25 ff.).

Of course, this distinction cuts across negative emotions as well. Consider the following pair:

(3) A is sorry that p,


(4) A fears that p.

Like (1), (3) entails that A knows that p. In this respect to be sorry is like a lot of other negative emotions: to be sad, unhappy, disappointed, displeased, to regret, and to grieve. All of these are factive emotions, while fearing (worrying, dreading, or being anxious about) p is an epistemic emotion like hope.

Since believing that p is probable to some degree less than 1 goes with believing not-p to be probable to the remaining degree, there is a complementarity between hope and fear. Suppose that A believes there to be a 0.5 probability that he has won on the Pools and (p.81) a 0.5 probability that he has not; then he is exposed to feeling both hope that he has won and fear that he has not. However, he cannot be in the grip of both hope and fear at the very same moment. Which emotion he is feeling at a certain moment is due to which alternative he is episodically thinking about at that time: if he dwells on the possibility of his having won, hope will be called forth, while if the other possibility occupies his mind, he will experience fear. Thus, a person in this sort of situation is subject to a characteristic vacillation between hope and fear. Which alternative one most frequently dwells upon in such situations is determined by one's temperament (whether one is an ‘optimist’ or a ‘pessimist’), one's present mood, or external circumstances.

There is, however, an asymmetry between fear and hope. Consider the distinction between hoping and longing. When life feels miserable, I may be longing for death, but I cannot hope for death (i.e. hope that I shall die), for I am certain of my mortality. I can hope for an early death (that I shall soon die), for this is something that may be merely epistemically possible, but this makes it inappropriate to long for an early death. In short, if it is true that I long for death—that is, long for my dying to occur—then it must be true that I know that this state will occur (I may even know when it will occur). If I had been uncertain in this matter, I would have to hope for death.

As opposed to this, it is obviously possible to fear death, even though one knows one is to die. The fact that a prisoner in a death row knows that he will be executed in a few hours does not prevent him from fearing death. This fear of death feels like the fear that one may die at any moment, which is a fear of something that one regards as merely possible. It is, however, not an epistemic fear, but a factive fear properly expressed by: ‘The fact that the prisoner will soon die makes him afraid.’ (This is probably an understatement; the prisoner is more likely to dread his imminent death.)

This factive fear is the negative counterpart of longing. In other words, fear seems to be the negative counterpart of both hope and longing. Does this mean that ‘fear’ designates two kinds of emotion or that ‘hope’ and ‘longing’ designate the same emotion? My suggestion is that in the more basic sense, in which different kinds of emotion imply different kinds of felt bodily reaction, there is just one positive and one negative emotion, but there is obviously a difference as regards the propositional content or cause.

To hope for something and to long for it both involve wishing it. As we saw in Chapter 4, a wish is a non-intelligent desire for something that the subject is convinced that it cannot cause to materialize. The difference between hope and longing lies in the thought causing the wish and consequently in the precise content of the wish. A hope that p seems to be a wish to ascertain that p is the case caused by the thought that it might be. It thus presupposes that the subject sees it as uncertain whether or not p will be (has been or is) the case. In contrast, a subject longing for it to be the case that p must take itself to be certain that p will occur (and perhaps even certain about when it will occur). Longing is a wish, caused by this conviction that p will be the case, that p be the case sooner, that time so to speak be ‘speeded up’, so that the gap between the present and the realization of this state of affairs is sooner closed.

Note that to say that hoping and longing consist in wishing does not imply that they are desires rather than emotions, for they possess the essential passivity of emotions: they are (p.82) felt, bodily states—though primarily or exclusively of a conative nature—caused by the having of certain propositional thoughts.

Fearing that p also involves having a wish (that p not be the case). It may be objected that the applicability of ‘fear’ extends further down the phylogenetic ladder than the applicability of ‘wishing’ (and of ‘hope’ and ‘longing’). But it should be remembered that a wish is a non-intelligent desire, and when such a desire activates behaviour which does not succeed in achieving its aim of eliminating the danger, the animal is in a situation similar to that of an animal which thinks that it cannot eliminate the danger, even though it is incapable of entertaining this kind of thought.

If factive fear that p were strictly parallel to longing, it would involve a wish that time be ‘slowed down’ so that p materializes later. But surely, when one fears that p what one wishes for is in the first instance that p never materializes; one wishes to avoid it altogether. That is, whether one is subject to factive fear, and believes that p will be realized, or epistemic fear, and believes that it may not be realized, one will tend to have the same wish, that p not be the case. This may be the reason why we speak of ‘fear’ in both cases.

As mentioned in the foregoing chapter, fear includes more than a wish, namely, physiological changes, like a quickened heart-beat, cold sweat, pallor, trembling, etc. I think these changes are present in both forms of fear, though it may be that the respective effects inherit some differences from the difference in respect of the epistemic modality of the causing thought.

The noted vacillation between hoping that p and fearing that not-p then involves swinging between a wish that p be the case caused by the thought that it might be and a wish that not-p not be the case caused by the thought that it might well be. In contrast, there is of course no complementarity between longing and factive fear, since longing does not presuppose epistemic uncertainty.

To anticipate a point of some importance for Chapter 16, longing of the sort that involves impatience is essentially directed at the future, to a state of affairs that is held to materialize in the future. Of course, it makes no sense to wish that time be speeded up so that some past event happens sooner. Instead, one might grieve over the inexorable ‘passage’ of time which removes the event further and further from the present. Similarly, it would be curious to fear factively that something happened in the past. As long as something has not happened, one may have an absurd hope that it will not happen, even though one firmly believes that it will. But when one knows that it has happened, there is only room for the wish that it had not, that is, for regret that it did happen (or for desireless emotions such as sadness and despair).

Positive and Negative Emotions

In expounding the distinction between factive and epistemic emotions, I have called upon another distinction that cuts right across it, namely the distinction between positive and negative emotions. As examples of positive emotions I have enumerated to hope, long, to be glad, pleased, happy, overjoyed, delighted, and relieved. Others will be added to the list (p.83) later on: to be proud or grateful and to admire. Among the negative emotions we find: to be afraid, worried, sad, displeased, unhappy, horrified, terrified, disappointed, ashamed, angry, jealous, envious, to regret, despair, and have contempt. In fact, with the exception of surprise (astonishment), puzzlement (amazement), and wonder (awe), it is arguable that every emotion named in everyday language is classifiable as either positive or negative.1 Consequently, the ground of this distinction merits attention.

Gordon suggests that whether an emotion is positive or negative turns on whether it involves a wish that the content be true or that it be false (1987: 29–32). In the case of positive factive emotions this wish is assumed to coexist with a conviction that it is fulfilled. Thus he writes:

we may say that the positive factive emotions arise from wish-satisfaction—a state in which one simultaneously wishes it to be the case that p and believes that it is the case that p. (1987: 53)

Similarly, negative factive emotions allegedly arise from wish-frustration, wishing something not to be the case when one believes (often correctly) that it is the case.

In view of the earlier criticism of the idea that all emotions involve desires, we should regard Gordon's proposal with suspicion. I do not think it is true, for instance, that despair involves having a wish or desire that things be otherwise. Despair is felt when hope is crushed or, more precisely, when not only a firm conviction is established, but it quashes the wish that things be otherwise. Despair or hopelessness seems to involve feelings of hollowness and powerlessness which occur precisely as the result of a wish being definitely defeated. Certainly, it is possible to retain a wish that p were not the case, in the face of a conviction that it is the case. As remarked, this appears to be what regretting that p is the case consists in. But this regret is clearly different from, say, being subject to despair of its being the case that p.

The criticism can, however, be made more telling in the case of positive factive emotions, for while it makes perfectly good sense to be absolutely certain that p and still wish that p were not the case (this is what happens when regret is experienced), it makes doubtful sense to wish that p is the case when one is absolutely certain that it is the case. Surely, the wish here gives way to a feeling of satisfaction. Of course, in some cases a certainty that p generates a desire to see to it that p remains the case, but this is not always so. Suppose I wish to spot a certain rare bird, say, a hook-billed kite, and I become utterly convinced that I now have succeeded in spotting one. Then it is surely not the case either that I go on wishing to spot this bird or that this wish is transformed into a wish that it remains the case that I have spotted one, for once this has become the case, it will remain so forever.

It might be objected that I would not be subject to a positive emotion when spotting the kite if beforehand I had not had a wish to spot this kind of bird. I think this is false, too (in addition, it is hardly what Gordon—see (1987: e.g. 52)—has in mind). Consider factive emotions in the production of which, I admit, an antecedent wish or desire does figure. One cannot be pleased or satisfied that p unless one has had an (occurrent) desire for p and (p.84) one realizes that p has come to be. This emotion consists in the feelings of relaxation and increased vigour or heightened vitality that can result from realizing that a desire has been fulfilled. The feelings of gladness, joy, and delight appear no different. But in their case it seems not required that there has been a pre-existing, occurrent desire or wish that the objects of gladness, joy, etc. be the case. Hence, the most reasonable conclusion seems to be that these feelings of relaxation, vigour, etc. can have another source than desire fulfilment. (As indicated in the foregoing chapter, I agree, however, that one must be disposed to have wishes regarding, or be concerned about, these objects.)

On the negative side, to be displeased or frustrated requires the presence of a pre-existent desire that to some extent survives the conviction that it has not been fulfilled, for it is the feeling of tension that results from awareness that a desire has been obstructed in its path. (In this respect, they resemble regret.) If the obstacle is overwhelming enough to crush the desire, a state of despair ensues. Thus, this state presupposes a pre-existing occurrent desire. Although sadness, sorrow, and grief as regards p consist in similar feelings of lack of vigour and energy like despair, I think it is not implied that the source of the feeling lies in the extinction of an unfulfilled occurrent wish for not-p. Rather, it is simply a feeling of being powerless, so that one is incapable of wishing this in face of one's current conviction that p is the case.

If, however, it is false that all emotions that can be classified as either positive or negative are so classifiable because they involve occurent wishes or desires, what could be the ground of this distinction? A suggestion that lies close at hand is that an emotion that is positive is so because it involves feelings that have the sensory quality of being pleasurable, while a negative emotion is one that involves feelings having a sensory quality that is intrinsically disliked, like that of being painful. This idea is not popular nowadays, but its merits may be undervalued. It is plausible to hold that moods can be divided into positive and negative ones, just like emotions. But moods lack content; so the division cannot depend on desires having a related content. The current proposal is that it instead has to do with what the felt pattern is intrinsically like.

Unfortunately, this is not unproblematic. Although it certainly feels pleasant to be glad and unpleasant to be sad, it is sometimes unclear that admittedly positive and negative emotions encapsulate such feelings. Above I expressed my inclination to hold the view that hope involves nothing but the wish that something thought to be epistemically possible, but beyond one's power, be found out to be the case. If hoping is nothing but having a wish, then it is hard to see how it could involve much of pleasurable sensations. It seems far more plausible to think that the positive character of hope has to do with there being a chance of something being as one wishes it to be. Consider also the negative counterpart of hope, fear. Although feeling fear is for the most part unpleasant, it seems that feeling slight fear—mere ‘tickles’ of fear—can be pleasant.

This adds up to the following conception of the distinction between positive and negative emotions. An emotion with the content that p is positive just in case it consists in a felt pattern of bodily changes that is intrinsically desired and/or it involves a wish that p be the case, while it is negative just in case it consists in a felt pattern intrinsically disliked and/or it involves a wish that p not be the case. Sometimes it is a defect of a definiens that (p.85) it includes a conjunction or a disjunction. This is so when it is not possible to detect any reason why the particular limbs in question are brought together in the definiens. But in the present case the limbs obviously have a common feature in the reference to desires.

It should be kept in mind, however, that even if an emotion is positive in itself, for the sake of the felt pattern it encompasses, it may possess ‘extrinsic’ features on account of which it is negative. This is true of Schadenfreude. It is in itself positive or desirable because it is a species of joy, but one's being in this state is likely to have bad consequences for others, since it implies that one has malicious standing wishes or desires, to the effect that someone else be harmed. Precisely the opposite is true of pity and compassion which I take to be the negative counterparts of Schadenfreude. Here one is sad or sorry because of the way something has gone for another, just as in the other case one is glad or delighted that this is so. ‘Pity’ and ‘compassion’ designate a negative emotion such that having it is likely to lead to positive action because it is linked with a benevolent concern for the weal and woe of another.

Agent-Oriented Emotions

I have discussed some factive emotions the possible contents of which are virtually unrestricted. Individuals can be glad or sad about something they have themselves done or something they have passively undergone, about something done or undergone by others, about something that will happen in the future or has happened in the past, and so on. The contents of epistemic emotions like hope and (one kind of) fear are similarly unrestricted, though an epistemic constraint is here in operation. Accordingly, I shall call these factive and epistemic emotions—as well as specifications of them, like compassion and Schadenfreudeplain emotions.

Within the class of factive emotions, I shall now distinguish two broad groups of emotions, each of which has characteristic restrictions on their contents. The first of these groups will be called agent-oriented emotions. Anger is the agent-oriented emotion most often discussed. The emotion of anger is necessarily elicited by the thought that that with which one is angry has caused some effect which is negative or which one wishes had not occurred. Anger consists in a set of bodily responses that include a tendency to behave aggressively towards this cause or agent. Thus, imagine that Cain inflicts pain on Abel by hitting him with a stick. This would make Abel angry with Cain only if he sees Cain as having caused him the pain.

It might be objected that if Cain beats Abel with a stick, the (movement of the) stick is also a cause of the pain felt by Abel, so why is Abel not angry with the stick as well as—or instead of—with Abel? A possible reply is that that with which one is angry must be an agent in the qualified sense of being capable of acting intentionally. It has been suggested that the target of anger is “(roughly) another's ill will” (Gordon, 1987: 64) or an agent who has “culpably offended” (Roberts, 2003: 204). In terms of our example, Abel would then be angry with Cain because he sees him—and him alone—as having intentionally caused him the pain.

(p.86) This idea is, however, hard to reconcile with the fact that persons often get angry not only with infants and non-human animals, but also with inanimate objects: for example, with their cars when they will not start. To effect a reconciliation here one would have to contend that anger in these cases is made feasible by one's, at least momentarily, taking these inanimate things to be agents capable of having the intention to cause harm to other beings (see Gordon, 1987: 56). But it appears to me as an altogether too fanciful hypothesis to ascribe, even momentarily, to the person kicking a car in anger an animistic conception of the car as an entity motivated by evil intentions.

The solution that I favour is instead this. Abel is angry with Cain rather than with the stick because he sees Cain as causing the movement of the stick. He is not angry with the stick because he sees it as not being, as I shall put it, a blank cause of the pain, but as a cause he fits into a causal network as an effect. In contrast, he is angry with Cain because there is nothing external to Cain which he sees as causing him to act in the manner he does—that is, he sees Cain as a blank or original cause of the pain he is feeling. In other words, one is angry with something only if one sees it as an in this epistemic sense blank cause of something one wishes to be without, like a pain felt. So the man who kicks a car in anger does so because he is for the moment not (episodically) aware of anything external to the car causing it to behave as it does. Had it straightaway occurred to him that the car's failure to start is the effect of, for example, someone's sabotage, the anger would have been directed at this agent rather than at the car. The idea that individuals be in this way temporarily oblivious to the external causes of inanimate things' behaviour is clearly less extravagant than is the idea that they temporarily subscribe to positive conceptions that are palpably false, like animistic ones.

This account can be buttressed by an evolutionary explanation of why anger—or more precisely, the tendency to behave aggressively, to hit back, etc.—should be geared to what is seen by the subject as a blank cause in the present sense of not seeing its activity as being caused by the activity of something external (rather than in the sense of positively seeing its activity as being uncaused by the activity of all other things). The reaction of taking vengeance, of for example doing something designed to cause pain to somebody who is believed to have inflicted pain on oneself, probably has survival value because it is liable to deter (potential) aggressors from causing harm in the future. Now, it could have this value only if it is directed at proper targets. Proper targets are entities with consciousness and, more particularly, a capacity to feel pain, since the behaviour one automatically tends to engage in when one is in anger is designed to cause pain. But tracing an effect, such as a pain one is feeling, to its causal conditions in the mind of another is a rather sophisticated feat. If the performance of this feat had been necessary for being angry, only a few non-human animals would be capable of it (and would thereby be the beneficiaries of its advantages).

It is a far less advanced feat to view perceivable events as being causally connected. Consequently, we should expect there to be a large group of animals with the ability to trace effects (upon themselves) to causes in the physical activities of other beings, but without the ability to see these activities as the outcomes of the mental states of the other beings (mental states which must not be seen as the results of factors external to (p.87) these beings). All of these animals will see the disappearance of a causal chain into a consciousness as the end of it; they will view the bodily movements immediately resulting from mental states as blank causes, that is, they will neither see them as being causally determined nor as undetermined. Hence, if their anger reactions are triggered by the idea of something as a blank cause, the targets of their reactions will include entities with consciousnesses that guide their behaviour. Certainly, in some cases the target will be improper, as in the case of the man kicking his car. But this is often so for the reason that human beings have invented machines the causes of whose activities are almost as little open to view as the causes of the behaviour of conscious beings. In a natural environment, anger directed at blank causes will in a majority of cases be oriented towards conscious beings. Admittedly, the precision is not so great as it could have been if the triggering thought had been to the effect that the agent was a conscious being, but this is outweighed by the fact that anger with the first kind of causal source could be much more widely spread among organisms.

Of course, an adult human being is normally capable of tracing the behaviour of other conscious beings to the causal antecedents in their minds. Indeed, in our reflective moments we are even aware of the fact that, at least in many cases, these mental states are the upshot of causal factors outside these conscious beings, that now react in a certain way because of their earlier lives. But that this is so is not anything that is perceivable: we do not perceive the mind of another and hence not its reactions to something external to it. That is why, in our more unreflective frames of mind, we may be oblivious to the fact that an activity which is caused by the mental states of another could have a further cause, and may instead view this agent as a blank cause. The possibility of being angry with an agent rests on this unawareness. This is the important point for, as will be seen in Part V, it accounts for the naturalness of viewing the target of anger as deserving punishment, since attributing desert to something involves taking it to be a blank cause.2

The positive counterpart of anger is, “perhaps surprisingly” (Ortony et al., 1988: 151), gratitude, which is felt to what is seen as a blank cause of something which is positive or which one is glad about. As anger comprises a tendency to behave in ways that are designed to cause pain to another, gratitude comprises a tendency to behave in ways that are designed to cause pleasure to another. Note that gratitude seems a privilege of more developed animals (it seems that, say, birds can get angry, but it is more doubtful whether they can be grateful). This parallels our findings in the case of hope and fear. It is also noteworthy that the more intense emotions of terror/dread and rage/fury have no (labelled) counterpart on the positive side. The explanation of this fact is presumably that the reactions of fleeing and fighting back have greater survival value than their positive counterparts.

In view of these considerations, it should not be surprising that there is something peculiar about anger and gratitude being directed at oneself. There is, for instance, no need to deter oneself from inflicting further pain on oneself by inflicting pain on oneself: the pain (p.88) already inflicted on oneself should be a sufficient deterrent. Here regret, or a wish that things had been different, is a more natural reaction. Remorse is regretting that one has responsibly brought about something one wishes that one had not brought about because it contravenes a moral norm.3

Feelings of guilt are appropriate whenever remorse is, but the converse does not seem to hold. These feelings are induced by the belief that one is blameworthy by owing somebody something. The reason why one owes somebody something may be that one has acted wrongly (but, it may seem, not necessarily morally wrongly, or wrongly to others4). This is not always so, however.5 Suppose a friend has done you a favour and you know that you ought to return the favour now because the friend is in a fix. If you akratically postpone returning the favour, the thought of your suffering friend is likely to make you feel guilty, but you are unlikely to feel remorse as long as the opportunity for action is still as good. It is odd to wish that you had acted rather than decide to act when the opportunity is still there. Furthermore, you may feel guilty because of a wrongful action done not by yourself, but by some agent so intimately related to you that you identify with that agent. Thus, Germans born after 1945 may feel guilty about the Nazi atrocities, but could scarcely feel any remorse in this regard.6

Indignation is a species of anger that is directed at someone who is regarded as having acting wrongly not merely from one's personal point of view, but from the point of view of morality (cf. Gordon, 1987: 56–7). Resentment is likewise a species of anger, but it is more long-term: if one is powerless to prevent someone repeatedly acting in a way that makes one angry with him, one will grow to resent him on account of his actions (cf. Ben-Ze'ev, 2000: 396). Therefore, resentment requires greater mental capacities than anger to be felt, and it will probably be felt only towards responsible agents. In this it is like indignation, but unlike the latter, it does not imply a condemnation of the target as having acted immorally (contrast Rawls, 1971: 488). As pointed out by Rawls (1971: 445–6, 483) and Gibbard (1990: 139, 295–300), remorse and feelings of guilt are as a rule experienced by subjects at whom indignation, resentment, or anger could be properly (p.89) oriented. Bitterness is an emotion that their victims could experience, since it is an emotion that one will have if one thinks one has been wrongly treated or, more generally, has received less than one deserves.

Jealousy appears to manifest itself in a mixture of anger, fear, or sorrow.7 It arises in a three-party context (Farrell, 1980: 530): when one is jealous, one typically regards oneself as running the risk of being deprived by some third party of some personal favour—like affection, attention, or esteem—that a second party bestows on oneself. Jealousy occurs because the loss of this favour is seen by one as diminishing one's personal worth. This is the core of this emotion. It is the possibility of the loss of the favour, and the consequent loss of personal worth, that induces fear and the certain fact of their occurrence that induces sorrow. The strength of this fear or sorrow, and thus of jealousy, is a measure of how important it is for one's self-esteem to be favourably received by the other. The anger is normally directed not only at the third party whom one sees as threatening to rob one of the favour of the second party, but also at the second party, since it, too, is regarded as partly responsible for the favour being in the process of shifting its objective. The presence of anger is evidenced by the fact that jealousy often issues in violent acts, crimes of passion. It is this element of anger which makes me place jealousy in this section, despite the fact that it is not a straightforward agent-oriented emotion.

Jealousy is often confused with envy. The latter is, however, frustration because one sees oneself undeservedly worse off in some respect than another. This respect might be a personal favour of some other being but, in contrast to the case of jealousy, not the favour of some particular being to whom one has some sort of valued connection: the favour of any being as good will do (cf. Ben-Ze'ev, 2000: 282, 290). Thus, envy leads to competition over the same particular goods only if there is scarcity. There is also a connection between jealousy and envy for the reason that, if another possesses a desirable asset, there will often seem to one to be a risk that, in virtue of that asset, the other will rob one of the favour and esteem of individuals important to one. Thus, Salieri might be envious of Mozart in so far as he sees him as possessing qualities of a kind that Salieri himself wishes to possess, namely, outstanding musical gifts. But he is jealous of Mozart to the extent that he thinks that, owing to these gifts, Mozart will or might rob him of the reputation and esteem particular individuals would otherwise reserve for him.

If one is frustrated by the thought of another's possession of some good even though it is something one could not have oneself or the possession of which would not make one better off (because one has already got something better), one begrudges that individual. Rawls (1971: 533) characterizes begrudging and being jealous as the “reverse” of envy. He then rightly sees envy as presupposing the envious subject's being worse off (1971: 532). But begrudging does not presuppose the superiority of the subject: I can begrudge people that their health is just as good as mine. Begrudging appears to me to be the desire to deprive others even though one does not think that one is worse off than (p.90) they and that deprivation would make oneself better off (except relatively, by making them worse off).

By taking up emotions like being envious and begrudging I have, however, reached emotions that with greater justification should be put in the next category to be introduced. This overlap between my categories indicates that my categorization is rough-and-ready, but it can still serve its purpose of summing up classes of emotions that are relevant to the discussion of responsibility in Part V.

Comparative Emotions

The second class of factive emotions on the contents of which it is possible to impose substantial restrictions is comparative emotions.8 The most discussed member of this category is pride. I shall follow this practice of spending most energy on pride.

Hume sometimes expresses himself as though pride and self-satisfaction were the very same emotion (1739–40/1978: 320). This is wrong already for the reason that, as I shall soon argue, pride is not necessarily directed at oneself: one can be proud of other things than oneself. So what is true is, at most, that to be proud of oneself is to be satisfied or pleased with oneself. This is also false, but it is instructive to see why.

I can be proud of myself because my ancestors are noblemen or because of the ample gifts—intelligence, beauty, etc.—with which nature has endowed me. But it would take an odd view of the world for me to be pleased or satisfied with myself for these reasons. For saying that I am pleased or satisfied with myself because I possess these properties seems to imply that it is the result of a desire of mine to bring it about that I am equipped with these features. (As indicated above, the emotion of being pleased or feeling satisfied necessarily results from the realization that a desire of one's has been satisfied.) But one would have to be mad to ascribe such a feat to oneself.

Superficially it might seem that the same thought can cause both self-satisfaction and pride, for example the thought that one has won a marathon race. But this thought leads to self-satisfaction only if it is taken as showing something like one's having exercised one's powers well—this being something that can intelligibly be assumed to be due to one's will. The victory can be a source of pride even if it is taken as demonstrating one to possess certain powers or abilities which are such that no sensible being could take itself as being capable of equipping itself with them.

The properties of oneself on account of which one is proud of oneself need not, then, be such that one sees them as resulting from one's own will or efforts, but they must not be such that one sees them as resulting from the activity of causes external to oneself. In this respect, pride resembles agent-oriented emotions: the features in question must be ‘blankly’ ascribed to oneself. One can be proud of oneself because of one's beauty and intelligence only if one's causal investigation stops short at one's possession of them. If (p.91) one sees one's beauty and intelligence as the upshot of hereditary factors that might have bestowed these properties on another, one's pride is undercut.

There is another difference between pride of oneself and being pleased with oneself. I can be pleased with myself in some respect, for example in respect of my ability to type, without being proud of myself in this respect. This is so, if my ability to type is sufficient for my purposes, but in no way superior or remarkable. As Hume puts it, the quality of ourselves that supplies the ground for pride must be “peculiar to ourselves, or at least common to us with a few persons” (1739–40/1978: 291). I would rather put it like this: when one is proud of oneself because of some feature F, one must find oneself superior in respect of F to the members of some reference class of importance to one, and thereby deserving of praise, which presupposes that being F is blankly ascribed. Since the superiority makes one deserving of praise, it goes without saying that F must be a feature thought to have positive value.

Needless to say, it is this element of comparison that provides the rationale for the label ‘comparative emotion’ (contrast Roberts, 2003: 275, who denies that pride is essentially comparative). The comparison issues in an (alleged) superiority making one deserving of praise that gives the felt pattern of pride its distinctive character of comprising a ‘swelling’ of the self and a desire to draw attention to oneself. Presumably, it is also this entailed conception of oneself as superior that makes some moralists condemn pride (of oneself) as a vice.

The opposite of pride is shame. Consequently, the state of being ashamed of oneself is nurtured by the thought that one is deserving of blame because one is inferior in respect of some valuable property, one's instantiation of which is seen as a blank occurrence. Shame consists in feelings of contraction and in a wish to hide away.

Admiration is another comparative emotion which, like pride, is directed at somebody who is held to be superior to oneself in some respect that makes for praiseworthiness. It differs from envy in that one holds the admired object to deserve its superiority. As Roberts points out (2003: 265), we commonly say that we admire such things as sunsets and natural scenery. But, unless we do not mean simply that we enjoy or take pleasure in these things, these attitudes require that we put down natural beauty to some agent that it makes sense to praise.

Initially, it might be tempting to think that pride and admiration differ in that while the former is essentially directed at oneself, the latter is essentially directed at some responsible agent other than oneself. Hume, for instance, claims that the object of pride is always “self, or that succession of related ideas and impressions, of which we have an intimate memory and consciousness” (1739–40/1978: 277). Recently this thesis has been reaffirmed by Gabriele Taylor: “pride is always self-directed” (1985: 35 n.). I cannot here enter into the details of Hume's and Taylor's views, but I think it is fair to say that they cannot be sustained unless the following is true: whenever a subject is proud, that of which the subject is proud can always be (re)described as itself. Thus, if a man is said to be proud of his wife because she is a woman of great intelligence, sensibility, and beauty, this can be rephrased as his being proud of himself because he has a wife who is a woman of great intelligence etc. This is so because to be proud of his wife, he must see her (alleged) superiority as reflecting on himself, showing him to have superior qualities as well.

(p.92) No doubt there are men who fit this description, but hopefully there are also men who are proud of their wives, but who do not wish the attention and admiration their wives receive to be redirected onto themselves—and even such men who are ashamed of themselves and think themselves not worthy of their wives. In some cases, it may be even more bizarre to construe instances of pride as at bottom self-directed. It is a commonplace that, when talking to foreigners, people often express pride of some of their compatriots because of their feats and achievements. It seems wrong to hold that, at bottom, people are here expressing pride of themselves because they belong to the same nation as some eminent person.

A more plausible interpretation is that people can be proud on behalf of others because they have the ability to ‘identify’ with them, the ability to imagine, from the inside, what it is like to be them and, consequently, to be concerned that their desires be fulfilled (e.g. to imagine being in the shoes of a player of the national team and receiving cheers). This is the same sort of attitude as we adopt to future stages of ourselves when we anticipate how things will be like for them (hence, the appropriateness of speaking of an “extended self”, as does, e.g. Ben-Ze'ev, 2000: 516). As we will see in Chapter 23, it is a mark of the imagination involved in (experiential) anticipation that it is not voluntary, but automatically elicited. It seems reasonable to think that the same must be true of the imaginative identification in pride. Given my hypothesis that, like anger, pride rests on a belief—or rather an absence of beliefs—about causal origin, it appears natural to conjecture that when one is proud of another, one must believe there to be some connection with respect to origin of the other to oneself.

Thus, the fact that national bonds sometimes—especially in the presence of foreigners—allow people to be proud of their compatriots (as Gabriele Taylor recognizes, 1985: 30) would then have to do with their belief in a common origin which is more likely than other origins—for example, the present foreigners'—to culminate in greatness. (One need not believe that it has culminated in greatness in oneself, so one need not be proud of oneself.) Analogously, when a man is proud of his beloved wife, his love of her could enable him to identify with her imaginatively, to see things from her perspective and endorse her aims as she herself is presumed to endorse them; but for him to feel proud of her he must believe that the origin of her having the properties that he is proud that she has somehow has to do with himself, though not necessarily to the extent that he has reason to be proud of himself, too.

So, although I agree that there has to be belief in some link to oneself as regards the origin of the individual of which one is proud, I claim that there are instances of pride that are irreducibly pride of something other than oneself and that this is so because persons have the capacity to identify imaginatively with other beings (but not, of course, with inanimate things; so therefore pride of, for example, one's material possessions is always pride of oneself). This conclusion is amplified in Part IV when I argue that some of the psychological relations to which I have appealed in my account of pride—for instance, imagination from the inside and the concern consequent upon it—is not, contrary to what some have assumed, essentially directed to the self. If this is (p.93) so, any alleged essential self-directedness of pride could not be anything more than a linguistic accident to be remedied, for example, by the coining of a new term like ‘other-pride’ (by analogy with ‘self-pity’).

In contrast, it is incontestable that admiration must be directed at another. A person admired is thought of as superior to oneself: it is of the essence of admiration that the person admired is somebody to whom one looks up and can aspire to emulate. However, this does not seem to explain the difference between admiration and pride, for the latter could also be oriented towards somebody judged superior to oneself. But in fact the characterization of admiration given suggests wherein the differentiae is to be found: one can look up to, and aspire to emulate, somebody only if one ‘sets oneself apart’ from this person, that is, only if one does not imaginatively identify with this person in the sense that being proud of somebody requires identification with him or her. Thus, whether one is proud of or admires another depends upon whether or not one's connection with him or her is such as to permit imaginative identification. (When superiority is vast, reverence takes the place of admiration.)

Contempt is the negative counterpart of admiration: it results from regarding somebody, with whom one does not identify, as inferior to oneself in some respect of some blankly ascribed feature, and so deserving to be blamed or badly treated. But there is a noteworthy difference between admiration and contempt: the latter is not necessarily other-directed, for there is such a thing as self-contempt. How could that be? The reason is that one could dislike oneself as one was at some time other than the present to such a degree that one is no longer able to identify with this person. It is then no longer possible to be ashamed on behalf of this person, for shame, being the opposite of pride, demands identification. So one will feel contempt for this being, contempt being an emotion that presupposes the absence of identification. But to admire oneself for what one was in the past, one would have to regard one's past self as superior, and under these circumstances one has no reason to give up identification with one's past self. Consequently, one will end up being proud of one's past self rather than admiring it.

Akin to shame is the emotion of embarrassment; it is so like that it is tempting to view embarrassment as a mild “particular form” of shame (Williams, 1993: 89). Underlying embarrassment is the thought that somebody with whom one identifies is deserving of laughter and ridicule, and this response might seem to be a milder form of what one deserves in shame. But the inferiority in embarrassment stems from something external to one, such as one's being observed in special circumstances rather than from one's then possessing any inherent features as in shame. Thus, one can be embarrassed by being caught in the act of undressing, though the fact that one undresses is nothing to be ashamed of: everybody undresses, so it is nothing that makes one inferior. It is just the fact that one happens to be observed while doing so that makes one inferior. Therefore, although embarrassment and shame have in common the relation of inferiority, they are best regarded as differing not only in intensity. Embarrassment has to do with being inferior due to circumstances one happens to be in.

(p.94) Some Leftovers

I have distinguished agent-oriented and comparative emotions from plain ones with an ulterior purpose in mind: the discussion of responsibility and causal origin in Part V has a special bearing on the rationality of the former two. Nonetheless, this tripartition could have a point even were one to classify emotions for no other purpose than to understand their nature. But I am willing to concede that, from this perspective, my tripartition would leave something to be desired. For instance, although remorse is probably a species of regret, the former comes out as agent-oriented and the latter as plain. Moreover, because of its link to anger, jealousy must be counted as an agent-oriented emotion rather than as comparative. But envy is more of a comparative emotion, yet these emotions are so similar that they are often confused. However, this need not worry me as long as my typology does not miss any fundamental emotion to which reference is relevant in the discussion of responsibility in Part V.

Of course, I do not claim to have surveyed every kind of emotion, for there is an indefinite number of them. Feeling lonely, locked up, confident, on top of the world, and so on may all be different kinds of emotion, caused by beliefs to the effect that one is lonely, in a situation like that of being locked up, etc. Presumably, though, they are merely specifications of such emotions as sadness, fear, hope, joy, etc.

In my review, some para-cognitive attitudes are missing, although they are often cited as prime examples of emotion, namely, love and hate.9 The reason for this omission is that they straddle the distinction between desire and emotion. To love, or like, doing something is to desire to do it, just as to hate, or dislike, doing something is to want to avoid doing it.10 Loving, or liking, somebody, because of certain features of hers, is an emotional state by the passivity criterion of being a state which is identified by its cause, but it is a conative state of loving or liking to engage with her in various activities related to the desire-arousing features.

Loving somebody differs from merely liking her in that it typically includes what in Part IV I shall call concern for (the well-being of) her, that is, desires to the effect that the desires of her be satisfied for their own sakes. Liking can be purely instrumental: if one likes someone because she is good at something, one will desire to engage in this activity with her, and one may desire that her desires be fulfilled only to the extent that this is necessary to make the engagement in this activity profitable. Similarly, dislike of somebody (p.95) need be nothing more than a desire to stay away from her and, if one dislikes her on the ground of some aspect of her behaviour, a desire that she be hindered from indulging in this form of conduct. In contrast, hate also involves malevolence, that is, a desire that life in general for this individual be made difficult.

Loving and hating somebody differ from the agent-oriented emotions of anger and gratitude in that, while one may be angry with or grateful to somebody, because of a single fact noticed about her, love or hate are normally sustained by multiple grounds that are proverbially hard to sort out. It seems typical of hatred of somebody that it grows out of being angry with this person for several things, over time, in circumstances in which one is unable to avenge oneself. There may be a transition from anger, via resentment of various aspects of a person, to hate of the whole person. In opposition to this, love does not primarily grow out of gratitude, though it may partly do so. To love somebody is to be attracted to her, while to hate is not exactly to be repelled by someone or finding her unattractive. The opposite of love is rather both hate and something like repulsion or disgust than simply hate. Love and hate will be further discussed, largely by implication, in Part IV when I examine their constituents (that is, in the case of love, liking, and concern). (p.96)


(1) For this reason, Ortony et al., 1988: 32, deny that surprise is an emotion, but that seems to me counter-intuitive.

(2) In presupposing the notion of a blank cause, anger differs from the similar emotion of irritation or annoyance. A buzzing sound or an itching or tickling sensation may make you irritated or annoyed, but you can only be angry with the insect producing it, not with the sound or sensation.

(3) Cf. Gabriele Taylor (1985: 98). She also stresses that regret, but not remorse, may be felt because of events for which one regards oneself as in no sense responsible, like the passing of the summer.

(4) Cf. Wallace (1994: 238–40). If so, it follows that one may feel guilt without feeling remorse, simply because one does not take one's wrongful action to be morally wrong.

(5) Apart from the fact that Allan Gibbard mistakenly associates feelings of guilt with the idea of acting morally wrongly rather than owing somebody something, he puts the cart before the horse when he defines moral wrongness by reference to feelings of guilt—and resentment (1990: 45). Surely, it should rather be the other way around: the feeling of guilt is the feeling that is occasioned by the thought that one has perpetrated a wrongful act against somebody. Gibbard comes closest to facing this objection when he discusses whether the feeling of guilt involves the thought of one's being at fault (1990: 148–50). He claims that the feeling does not involve this thought because there is such a thing as feeling guilt senselessly. To this it could be retorted that feeling guilt senselessly does rest on a thought that one is at fault, but a thought that one recognizes to be irrational, a thought that one cannot help having, although one realizes that it is at odds with one's evidence (see the Nazi example in the next footnote).

(6) It follows that I cannot agree with Bernard Williams that what he terms “agent-regret” is “psychologically and structurally a manifestation of guilt” (1993: 93). Also, I differ from him in denying that guilt can rationally and properly occur even when one is not responsible, though if agent-regret is properly regret, it can rationally be felt in these circumstances. (Thus, I think it is irrational for young Germans to feel guilty about Nazi crimes, but quite proper for them to deeply regret them.) Williams thinks that branding the guilt felt in these circumstances as irrational “carries no useful message” (1993: 93).

(7) Roberts (2003: 258) mentions these three emotions, while some other writers, like Farrell (1980) and Neu (1980), mention only anger and fear. As Roberts stresses, bringing out the connection between jealousy and these emotions is not enough to pinpoint it.

(8) The present distinction between agent-oriented and comparative emotions has been arrived at independently of, and does not coincide with, Jon Elster's distinction between emotions of interaction and emotions of comparison (1999: 141).

(9) For instance, in the tripartition of emotions that Ortony et al. (1988) present, they constitute the third category, emotions that focus on objects, alongside emotions that focus on events—which roughly correspond to my plain emotions—and emotions that focus on agents—which roughly correspond to my agent-oriented and comparative emotions.

(10) Contrast Gaus who asserts that liking and disliking are emotions (1990: 65) and who even goes as far as to claim that “the overwhelming majority of emotions, if not all, can be described—not fully, but partly—as a type of liking or disliking of something” (1990: 69). The latter claim—with which Ben-Ze'ev chimes in (2000: 94)—must be false if, as argued in the foregoing chapter, it is false that all emotions involve desiring or wishing. Contrast this claim to Dent's view that “love … underpins all our other emotional responses” (1984: 82)—even hate (p. 84)! As his discussion of hate shows, this claim does not mean that love is an ingredient of all other emotional responses, but rather that they arise from it. This is in line with my concession in the foregoing chapter that something like concern dispositionally understood can feature in the explanation of an emotion.