Introduction

Chapter One: Emotional Control

1. Introduction

In this chapter, I compare different theories of the emotions with respect to how much volition they allow in the formation of emotional habits or in the experience of emotional events. Attributions of responsibility imply some degree of freedom on the part of the person in question to have responded differently. It is because one is assumed to have been in control of their behaviour or responses that we hold that person accountable. My account of emotional responsibility must therefore begin with the establishment of how much freedom one has in the development of their emotional habits and in particular cases of emotional response.

Below I examine connections between the degree of volition afforded emotion by several contemporary theorists and their version of how emotions are caused. Emotions in the biological theories of Paul Ekman (2003), Paul Griffiths (2003a, 2003b), or Justin d’Arms and Daniel Jacobson (2003) could not really be considered voluntary given that emotions on these accounts are basically reactions to environmental stimuli and operate independently of will and belief. According to Cheshire Calhoun’s cognitive theory (2003), emotions are ways of seeing determined by cognitive sets, patterns of thinking and attending, which are typically passively developed. Ronald de Sousa (1987) grants emotions objective cognitive status through their analogy to perception. Like perception, emotions are also generally automatic, although he allows certain limited powers of inducing one’s own emotional state. These influential contemporary theories, on examination, seem to leave little room for emotional volition. However, de Sousa’s theory includes an indirect method of emotional control he refers to as “bootstrapping” (1987, 11), which allows one to modify one’s emotions through a kind of feigning which becomes authentic. This makes his view compatible with mine and is why I will return to de Sousa when I look at ways in which emotions may be morally assessed in Chapter Three.

In Part 3 below, I look at strategic theories of the emotions, in which emotions either are strategies adopted toward certain ends (Jean-Paul Sartre 1960, Robert Solomon 1976, 2003a, 2003b), or are evaluated in terms of their instrumental value (Patricia Greenspan 1988). I focus on Robert Solomon’s theory of emotions as judgements because I accept many of his conclusions despite rejecting most of his assumptions. While Solomon in some places affirms and in others denies being a cognitivist, his views, at least those defended in The Passions: The Myth and Nature of Human Emotions (1976), seem existentialist in their account of the nature and formation of emotion, and emotivist in their metaethics. While Solomon’s account affords our emotional life the freedom needed to attribute responsibility, it contains at least three flaws: 1) his use of judgement as a model fails to establish the existence of that freedom; 2) his doctrine of emotions as evaluative judgements which are constitutive of value in the world fails to provide any way to distinguish between valid evaluative judgements and invalid ones. Such a distinction would have to be able to be made for there to be any ethical relevance to the question of whether one is responsible for their emotions; 3) Solomon’s “emotions as judgements” thesis fails to adequately account for the motivating force of emotions. It is the motivating force of emotions which makes them of special concern for ethics.

In critiquing Solomon, I use Greenspan’s arguments against judgementalism. Besides her concern that judgementalism, like some other cognitive theories, cannot account for cases of recalcitrant emotions, she charges that judgementalism overlooks the affective element of emotion which encourages one to action through states of comfort or discomfort. In addition to correcting this oversight of judgementalism, Greenspan’s theory offers the advantage over Solomon’s of introducing a standard by which emotions themselves may be judged, that of their instrumental value.

To contrast strategic views with the anti-cognitive theories with which I begin Part 2, I offer d’Arms and Jacobson’s argument against all forms of judgementalism, including the view Greenspan holds. They side with Griffiths’ view of emotion as non-cognitive environmental response, and posit an automatic system of emotive evaluation independent of any cognitive faculties. On their view, emotions may be judged “fitting” or not relative to the situations with which they are normally associated, but admit of no further justification beyond that.

In Part 4, I look at the different forms emotional assessment could take in the different theories discussed. I conclude that the kind of theory most amenable to the claim that one has moral responsibility for their emotions would be one which allowed both for a fair degree of emotional voluntarism, while retaining strong emphasis on the motivational force of emotions.

2. Degrees of Voluntarism in Theories of the Emotions

i) Biological Theories

My examples of what I take to be the most deterministic of emotional theories are those taken from some of the most rigorously scientific treatments of the topic. Paul Ekman (2003) and Paul Griffiths (2003a, 2003b) prefer physiological and sociological approaches to the topic of emotions. Both offer accounts that are unfriendly to the degree of emotional voluntarism required for attributions of responsibility.

Ekman thinks that the complexity and organisation of an emotional response requires that there be some sort of central direction to our overall response. He claims that this direction is afforded by what he calls an affect program. An affect program is a mechanism of undefined nature posited by Ekman as the storehouse of the patterns of response which are activated by stimuli in the environment (122). Such programs have a genetic basis but are also shaped by learning and by bodily development or damage (123). What happens in an emotional response is that an event acts as an elicitor that engages our appraisal system, which decides whether or not to activate an affect program, which motivates emotional responses. The responses Ekman takes to be paradigmatic of the emotions are “brief, often quick, complex, organized, and difficult to control” (121). Where what may be taken for an emotion fails to be episodic in the way described here, Ekman suggests that “mood” may be a more appropriate term for what is occurring.

Griffiths considers Ekman’s affect program theory a useful way of explaining emotional behaviour, particularly in terms of evolutionary adaptiveness (Griffiths 2003a, 289). Building on Ekman’s idea of an automatic appraisal system, Griffiths proposes that there are actually three kinds of mental process that occur under the rubric of “emotion” (2003b, 45).[1]

What can be gleaned from science on the topic is which sets of neurophysiological events accompany which sorts of emotional responses. Griffiths considers “basic” those emotions which occur most rapidly, resulting from the arrival of minimally processed information at the amygdala. These emotions, including fear, anger, disgust, joy, sadness, and surprise, exhibit the first level of mental process. The next level of emotional responses includes those which occur after information has passed through the sensory cortices. This information takes longer to arrive at the amygdala, where it is unified into a phenomenal experience. This second process, and the third, most complex, system of response, which involves various parts of the brain combining representations and memory in the amygdala, differ from the more automatic basic emotions in that they include cognitive and strategic elements (2003b, 45). That is to say, although Griffiths denies that there is an essentially cognitive component to emotion in general, what we think of as emotions with cognitive content or strategic intent are combinations of basic emotions and other mental processes.

In his focus on the biological underpinnings of emotions, Griffiths avoids discussion of one’s volitional involvement in their emotional responses, although he does cite evidence that much emotion appears to be effected strategically (59). For example, various animals, including humans, display different signs of emotion in the presence of conspecifics than when alone, suggesting that the emoter is either repressing or failing to experience an emotion in absence of the circumstances that would make it functional (55-57). One smiles more in the presence of others, and does not shake one’s fist when angered by a malfunctioning toaster. Griffiths’ preference is to explain such cases in terms of the survival-advantage of developing such habits combined with the social conditions that habituate one to behaving certain ways in certain situations (54).

Neither Ekman nor Griffiths gives weight to the fact that emotions tend to be taken to be “about” something. While the “aboutness” of emotions is reduced to a causal role in Ekman’s account, Griffiths flatly rejects any attempt to give a cognitive account of emotions. He accuses those who would describe emotions in terms of “propositional attitudes” of making overstatements in service of combatting the thesis that “emotions are feelings” (2003a, 285). Solomon in turn accuses Griffiths of misinterpreting cognitive theories of the emotions (Solomon 2003b, 1).

ii) Cognitive Theories

Given the nomological closure desired by those who tend to favour biological theories of emotions, it is hardly surprising that they would be hostile to the idea of voluntary emotions. But although cognitive theories are largely a response to this deterministic bias, many fall into passive characterisations of emotional response. Cognitive theories of the emotions admit of varying degrees of agency, diverging according to which element of emotion is cognitive. Depending on the particular cognitivist view, emotions may be or may contain any of the following: beliefs, thoughts, judgements, episodes of recognition, perceptions, or seeings-as.

The claim that emotions are or entail certain beliefs is at least as old as Aristotle, (Kenny 2003, 220) and finds variations in such modern theorists as Anthony Kenny and Gilbert Ryle (Calhoun 2003, 237). But Cheshire Calhoun proposes several difficulties for the thesis that emotions are or necessarily include beliefs. Different problems arise whether one’s cognitive theory is of the “patchwork” sort, in which emotions are taken to consist of sets of mental, physical, and behaviour elements which are not unique to the category of emotions, or of the “unity” sort in which emotions are those elements experienced in a distinctly emotional way (237-238). Where patchwork accounts may be criticised for reducing emotions to non-emotional constituents, unity accounts raise questions such as how one entity can simultaneously be both a belief and a feeling (238).

Further problems Calhoun raises for the identification of beliefs and emotions include what she takes to be common beliefs about their respective differences. Emotions are taken to be passive, involuntary, and non-rational; beliefs are active and articulate. She endorses this distinction and criticises those who deny the voluntariness of belief to make it more like emotions, and those, namely Solomon, who deny the passivity of emotion to make it more like belief (238). But I think Calhoun may be mistaken about how voluntary both belief and emotion are. I will argue in Part 3 below both that belief is less voluntary and emotions more voluntary than commonly thought.

Calhoun also uses examples of cases of emotion-belief conflicts, such as suffering arachnophobia despite believing that most spiders are harmless, or homophobia despite having rejected beliefs about the unnaturalness of gay relations to support her thesis that emotions do not necessarily follow upon certain beliefs, but certain emotions ought to (239). Cognitivists might offer several explanations for how emotions can conflict with beliefs yet still be somehow related. One could admit that there are non-cognitive emotions, but this admission compromises the cognitivist project (240). One could instead adopt the “inertia hypothesis” and claim that an emotional habit may in some cases outrun a change of belief, but this would only highlight the distinction between emotion and belief (241). The final option Calhoun explores is the possibility of explaining emotion-belief conflicts as belief-belief conflicts. Given that one may hold beliefs at various levels, which beliefs may turn out to be inconsistent, it may be the case that recalcitrant emotions attach to lingering beliefs that may not be explicit. The problem with this move is that it requires a complicated psychology of false consciousness and risks rendering the first-person ascription of beliefs less authoritative (241-242).

Rather than taking the position of many cognitivists that emotions necessarily correspond to certain kinds of belief, a view logico-linguistic or ontological in nature (237), Calhoun construes the relationship of emotions and beliefs normatively: there are certain emotions which ought to accompany certain beliefs. Beliefs held merely intellectually are to be considered defective (242). An example she gives of a merely intellectual belief is that of a logic student who believes that modus ponens is valid, but fails to grasp why (243). The missing part of understanding in this case is the phenomenal “Aha!” that one experiences of “getting” or grasping a concept. To be able to believe evidentially (holding beliefs which are confirmed in experience) what one holds intellectually, one must have this sort of experience of grasping. The provision of this experience of “seeing as” is the cognitive role played by the emotions (245). The cognitive set which brings out the features that constitutes a case of “seeing as” is not necessarily part of a belief system. Beliefs are only one part of our cognitive life, most of the rest being somewhat occult interpretive faculties (244). It is the operation of this cognitive set which is responsible for our emotions.

Although Calhoun says that emotional cognitive sets may be passively acquired (244), she doesn’t specify whether they must, or whether it is possible to alter this set intentionally. Given that she takes normal cases of belief modification to entail an emotional development, and belief may be changed voluntarily, it seems possible that emotion is revisable, but only by way of belief. Also, whether the emotional set adjusts to suit new beliefs seems to be a matter of how closely it follows belief in general. In the case of someone who formed a belief but did not have the expected adjustment of attitude, it is not clear whether this defect could be corrected and how that would take place.

Calhoun’s insistence on the voluntariness of belief is also undermined by her insistence on the passivity of emotions. This problem is avoided in “Cognitive Emotions?” by her focus on cases where one has formed a correct belief and failed to see its object in the right way. However, there are also cases where one’s inability to believe evidentially prevents their coming to believe intellectually. That is, our emotional cognitive sets sometimes prevent the problem of emotion-belief conflict by barring the way to belief. I have in mind such cases of emotionally-invested evidence refusals as: “Michael Moore is full of lies,” “Not our Johnny,” and, “God loves you.” In many of the occasions where such statements are uttered, one is unable to accept the counter-evidence presented to them because of a deep emotional investment that would be threatened were a particular belief to be formed. One may feel pride in their country and so see all evidence as either confirming the appropriateness of this opinion or as suspect if it does not. Parents may cherish their child’s innocence so that they fail to note its loss.

Ronald de Sousa offers a perception analogy to explain the cognitive content of emotions, which he takes to be a better guarantor of the objectivity of emotions than seeings-as (1987, 149-158). Emotions on his account belong to the broader class of attitudes, which resemble both beliefs and perceptions. Like beliefs they are not attached to any single organ, but like perceptions they are perspectival and involuntary. Emotions, like perceptions, cannot be hypothetically adopted--they are as real as they are experienced (156).

To the question, “If emotions are like perception, what do they (quasi-) perceive?” de Sousa provides a complex account of the facets of emotional objects, and formal conditions of emotional rationality. If the proper object of sight is some visible quality, the analogous correlate of emotions would be some tertiary quality such as fearsomeness or lovabilty (152). But the list of the sorts of qualities that might be apprehended by emotion includes many which do not seem to have the objective status of the secondary qualities we perceive. Although there may be some variance in people’s perceptions of secondary qualities, there is a norm from which we do not expect to stray far. Deviation from perceptual norms usually raises the suspicion that the deviant perceiver suffers from some defect in his or her sensory apparatus. The individual who insists the chair is brown when all others insist it is green is charged with colour-blindness. And indeed we would wonder how someone could claim that the horde charging towards the village was not fearsome. But whether or not someone or something is lovable is a much more difficult issue to settle if it is resolvable at all.

De Sousa draws a parallel between emotive qualitities, such as lovable and fearsome, and enjoyable aesthetic qualities such as beauty and grace (151). While aesthetic qualities also tend to admit of varied recognition, there are more and less appropriate ascriptions of such qualities. One familiar with music theory, who has sensitive ears and a generally interested heart is somewhat more authoritative in claiming that a piece of music is beautiful than one who is tone deaf and moved only because of the piece’s association with a fond memory (155). De Sousa points out that similar concerns factor into the reliability of perception. Variations in perceptions may be caused by: variations in the primary properties of the object; environmental conditions; the physiology of the perceivers; experience, beliefs, and desires; or social and ideological factors (152). The difference with emotion and aesthetic sensibility is in terms of which of these factors tend to have the greater influence.

But de Sousa’s theory gets even more fine-grained than this, and only key elements will be explained here. Such properties as fearsomeness are the formal objects of emotions (122). The objective property of the emotion’s target (that or them at which the emotion is directed) which is thought by the emoter to be the cause of the emotion is a focal property. In a standard case, a paradigm scenario (explained below), if a focal property is: 1) a real property of the target, and 2) also the cause of the emotion, it counts as a motivating aspect (116-117, 120). The formal object is what makes the motivating aspect intelligible. If I am afraid of the oncoming horde (the target of my fear), because they are violent and armed (focal properties), and not because I suffer from hippophobia (the irrational fear of horses), and it is normally the case that violent, armed hordes are considered fearsome, then these properties are the motivating aspects of my fear.

De Sousa explains the establishment of what is to be considered a normal emotional response in terms of paradigm scenarios. Such scenarios consist of characteristic objects and characteristic responses established biologically and culturally (182). Children have genetic predispositions to respond certain ways to certain situations. As they are taught to identify and name such responses in the context of the scenarios in which they occur, they develop a vocabulary of emotions (183). As one develops into maturity, one’s emotional vocabulary, as well as their stock of scenario-response associations increase in size, complexity, and subtlety (183-184). Since deviations from normal response tend to be discouraged or denied uptake[2], one is trained into typical emotional behaviour.

Although emotional responses, according to this account, are biologically seeded and socially nurtured, they are not automated reactions (182). It seems to me that if one could not attend to one’s responses and to some extent curb or encourage them, one could not learn appropriate emotional behaviour.[3] De Sousa refers to cases of using one’s beliefs to induce other beliefs, or emotions to induce other emotions as “bootstrapping” (238). Some of the ways in which this occurs are through conscious efforts to shift one’s focus of attention (243) or through the adoption of emotion-specific behaviour which can lead to the development of the emotion itself (241).

3. Strategic theories and their critics.

i) Existential judgmentalism.

Jean-Paul Sartre holds that self-deceptive feigning of emotion, such as de Sousa describes being employed in the service of bootstrapping (de Sousa, 241) is characteristic of most emotional expression, and that most cases of emotional response can be explained in strategic terms. Strategic views of emotion purport to offer the widest allowance for voluntarism, but I shall argue that they are riddled with similar limitations to cognitive theories.

In his Equisse d’une théorie des émotions,[4] Sartre argues that emotions are strategies for creating meaning in an indifferent world and for avoiding confrontations with one’s self and with others. Emotions transform the world to frame events in ways which one can accept (Sartre, 43). People faced with difficult assignments might change their attitude to the task in such a way as to render its goal undesirable. People faced with situations that call for decisive action may “suffer” a breakdown which would excuse them from duty (46).

Sartre explains that such activities are not planned in the sense of having been considered in advance—rather, they are a form of conduite-irréfléchi, or unreflective action (40). Such actions are not unconscious, but are structures of consciousness, ways of being conscious of their objects. He uses the example of the act of writing as a structure of consciousness, not conscious of itself, but of the statement being recorded to paper (40). Similarly, one does not reflectively engage in emotional behaviour, but unreflectively takes on an emotion as a way of being conscious of an object (39) (counting things, states and events all as potential objects). In passive fear one faints to remove the situation from consciousness (by removing consciousness from the situation). In active fear one does not run to get away, but apprehends something as to-be-run-from because there is no way to make the object go away itself (46). In passive sadness, one sulks over loss to put off the challenging task of finding new meaning in the world (47). In active sadness, one breaks into sobs and hysteria as a way of making oneself conscious of a problem as overwhelming so one may deny responsibility for having to face the problem (48).

Robert Solomon, influenced by Sartre, develops the idea of emotions as strategies and ways of seeing (seeing-as). He claims that emotions place value in the world. If the world that would exist unobserved can be called “reality,” the world we know, which includes beauty and fearsomeness, is “surreality” (Solomon 1976, 67). His choice of this unfortunately confusing term (he is not referring to baguettes in the sky) is motivated by its “super-reality” or “reality-plus” etymological meaning. But although he is a subjectivist about value, he also seems committed to there being at least one universal value: the maximisation of personal dignity and self-esteem. This maximisation, he claims, is the ultimate end of all emotional strategies (190).

Solomon expresses concern in “What is a ‘Cognitive Theory’ of the Emotions?” that the term “cognition” is too often taken to mean “information processing” (Solomon 2003b, 1). Narrow views of what counts as cognition motivate incredulousness at the idea of cognitive emotions. In fact, cognitivists diverge widely on what counts as cognition and which features of emotion are cognitive (or which features of cognition are emotive). He agrees with Calhoun that “belief” fails to fit with all cases, but adds that this is in part because beliefs are not experiences[5] (8). He also rejects the possibility that emotions could be thoughts on the grounds that thoughts are too episodic, and emotions tend not to be so intellectual (5-6) (This is in interesting contrast to Ekman who thought moods were disqualified from being emotions for failing to be episodic).

Solomon’s choice of cognitive faculty upon which to model emotion is judgement. He prefers judgement to belief because the former captures both the aboutness and the evaluative tone of emotion. He’s sometimes careful to make clear that judgements needn’t be conscious or articulate but may be reactionary and dogmatic (1976, 201). In some of his work, Solomon’s model of judgement is much more deliberative. Judgements can be criticised and argued against, which is part of what makes emotions revisable (2003a, 231). As judgements, they may be refuted.

It is not clear, however, if the characterisation of emotion as cognitive helps the case for emotional voluntarism as much as Solomon seems to think it does. What he assumes in each case, in common with Calhoun, is that our non-emotional cognitive faculties are voluntary. Unlike Calhoun, who contrasts the voluntary cognitive with the involuntary nature of emotions, Solomon uses a voluntaristic account of cognition to support his claim that emotions are voluntary. Solomon may be correct that we have enough power over our emotions to render us responsible for them, and he has some plausible psychological stories of how such control works in many situations. But he is wrong to suppose that this ability and responsibility can be derived from the equation that emotions are judgements.

It is debatable how much volition is involved in a judgement. In a conscious deliberative judgement where one chooses the matter to be pronounced upon and weighs evidence, there is freedom in judging insofar as one chooses which topic is to be considered, which facts to consider, and the method of arriving at a conclusion. If the evidence (or argument) is genuinely convincing it seems that the conclusion is from that point forced. The rational deliberator is not expected to be able to deny a conclusion that follows logically from premises or empirically from observations. However, in many cases the conclusion is underdetermined by the evidence and a decision is required to make a final pronouncement. In such a case a judgement is somewhat voluntary (one has room for doubt), though not entirely arbitrary.

But many judgements occur without conscious deliberation, and we would here need a stretched idea of volition. Perceptual, aesthetic, and intuitive judgements seem more to be matters of sensitivity than deliberative processes. The things in the mirror seem farther than they are; I cannot stomach creamed corn; this salesperson gives me the creeps. Now, it may be the case that I have had some hand in the development of the constitution that forces these judgements. I haven’t trained myself to correct for illusions; I haven’t given a dish a chance to grow on me; I have involved myself in a community that tends to distrust certain groups, which influences my own attitudes. But it would be hard to argue for deliberate occurrent control in these cases. So if Solomon would have it that one is responsible for perceptual, aesthetic, or intuitive judgements (the latter two being the most clearly tied to the emotions) it would have to be on the ground of establishing their complicity in the development of the apparatuses that judge in these respective cases.

Solomon provides examples of emotions which are consciously fuelled, such as ruminating on a wrong done to oneself in order to intensify one’s hatred toward the wrongdoer, and examples of emotions less consciously arrived at (1976, 122). It is unclear to me whether he will accept truly unconscious emotional judgements, as he has a complex theory of self-deception which seems to be able to render any mental process conscious in some sense or another.

Regardless of whether emotional judgements are made consciously or not, they all follow the same logic, according to Solomon. He uses a courtroom analogy to explain his model. According to him, a court creates guilt as a constitutive speech-act. The phrase, “This court finds the defendant guilty,” is, according to Solomon, a sort of responsibility-avoiding rationalisation in which the verdict is made to appear as a discovery rather than as a decision. This process is analogous to what Solomon calls “the myth of the passions” (9) whereby we describe our emotions in passive terms to deflect responsibility for their creation. On this model, guilt is not a predicate of the defendant prior to their being constituted as such by the legal process which includes laws, courts, trials, and finally a verdict (195-196). Solomon’s model seems to entail that the commission of a crime is just the first step toward guilt, but guilt itself is not achieved until the final act of the judge’s pronouncement takes place.

At one point in this analogy, Solomon refers to the judge as someone who acts as the final stamp on a chain of settled institutions and free agents. By admitting the law and fixed procedures of jurisprudence into his model of judgement, Solomon may be seen to be compromising the view that judgements are voluntary. On the other hand, this model may be useful for showing the hybrid nature of judgement as involving both decision and constraints. Presumably, in a clear-cut court case involving competent lawyers and a rational, virtuous judge, the facts combined with the laws and the rules of jurisprudence would essentially determine a verdict. However, we do argue and deliberate over matters of innocence and guilt, and the reason we are concerned with the competence and virtue of legal representatives is that cases are seldom clear-cut or completely anticipated by the rules. So there must be room for the discretion of the representatives. But not just any decision is possible. The rules that exist are being interpreted as they are being applied. This is an apt analogy for the emotions because there are at least two levels of (first-person) interpretation involved in emotional expression:

1) The physiological symptoms accompanying an emotion may be interpreted in different ways (Schachter & Singer 2003, 111). As further knowledge of the circumstances surrounding a case will affect the assessment and assignment of blame, someone’s interpretation of what may feel like anger will depend on whether the person has just received an offense, or is underslept and overcaffinated (Solomon 1976, 155-156).

2) Again using anger as an example, the fact of whether or not one has received an offense is discovered in part by interpreting the emotion. Assuming I have interpreted my symptoms as anger, I still need to decide whether this anger is justified, that is, whether the perceived offense really is an offense. I may initially have taken something as an offense, as indicated to me by my anger, and then realised that I had misunderstood what was being said.

On Solomon’s account the very taking of offense constitutes the offense as offensive. This seems to result in a very relativistic picture of emotional judgement. It does not seem to allow for an emotion itself being assessed as appropriate or inappropriate, though he allows certain emotions to be rational or irrational insofar as they serve the end of promoting dignity and self-esteem.

I believe the appeal of these existentialist theories of emotion lies in their direct opposition to those who take passion and the primacy of desires as evidence for psychological determinism. On a classic sort of determinist account: biology determines one’s desires; desires manifest in our emotions; our emotions determine our intentions. But on the Sartrean model, our intentions effect[6] our emotions, and desire is just a mode of intending. What is lost by this reversal, however, is the grounding that more deterministic theories have. While the assumption of determinism may seem to render assessment of emotions or any sort of conduct pointless,[7] appeal to evolution and biology provides the possibility of objectively desirable targets, through measures of survival or fitness. These objective desires could then anchor the chain of causation leading to an emotional episode, state, or disposition, which are the sorts of thing that determine action.

But assessment of such a chain would be mere diagnostics, and “responsibility” a figurative term. “Swann was responsible for his torture because he obsessed on Odette,” would be metaphysically equivalent to, “the chain in the tank is responsible for the failure of the toilet to flush because it snapped.” It may not be the case that morality and physics need apply to metaphysically distinct objects in order to justify some sort of moral discussion and evaluation, but it does seem that reduction of the moral to the physical or biological would be a much more tempting project on the assumption of determinism. Such a reduction could undermine our belief in moral responsibility and that very well could spell disaster for morality generally. Because emotions bring out salient features of situations (de Sousa, above), and provide motivation (Greenspan, below), if we lacked the appropriate emotions in situations of moral choice, it would be very hard to understand the imperative nature of the moral choice over any other possible choice. And if the cognitivist intuition is right that beliefs and emotions intertwine, then loss of belief in the relevance of moral choice could lead to the loss of the moral sentiments themselves.

My main disagreement with Sartre and Solomon, which I expect is shared by many others, is over the sweeping nature of their theories. There seems nothing one can do for which one could not be considered responsible. If we were responsible for every emotion, discussions of responsibility would be as pointless as they would be were we never responsible for them.

Also, much of the significance of many emotions lies precisely in that they are unchosen. If one could just “snap out” of a depression, it would undermine the demand of many sufferers to be treated as worthy of medical help. If we could choose whom to love, we could all settle into the most rational relationships possible. Such rational matchmaking might lack some of the sense of luck which we value in our discovery of a match. Also, if emotions are evaluative, it seems that part of their accuracy should rely on their being receptive, measuring what’s in the world, rather than projective, placing things in the world.

ii) Greenspan’s attack on judgmentalism.

While emotions may be used strategically, it is not always the case that they are. And while they may imply beliefs or intentions, it is not always the case that they do. It is for this that reason that Patricia Greenspan objects to the classification of emotions as judgements. She grants that emotions often include evaluative judgements, but not all emotions entail judgements, and the reduction of emotion to a type of judgement obscures the role of emotion in motivation and mistakes questions of emotional justification for questions of epistemological justification.

Greenspan grants the judgementalist view that emotions often imply beliefs—judgements that certain states of affairs may obtain. For example, fear often includes a judgement that the situation I am in is dangerous. This is the factual component of the judgement, but what makes fear an appropriate correlate to a belief that there is danger in this example is its evaluative element—that danger, the possibility of harm to one’s self or what one holds dear, is to be regarded as an evil (Greenspan 1988, 3). Greenspan’s concern with the judgementalist position is that it seems to entail that the appropriateness of emotions can only be spoken of in terms of adequate reasons for belief (in this case, whether or not there really is any danger) or in terms of the evaluations accompanying them (whether danger really is an evil) (5). However, there are other ways in which an emotion may be inappropriate or appropriate. On one hand, it may rest on grounds which would be inadequate for belief justification, but still be perfectly appropriate. On the other, it may be reasonable by judgemental standards, but inappropriate in terms of its adaptiveness.[8]

The adaptiveness of an emotion is its instrumental value (9). Because emotions serve as spurs to action, they may be assessed in practical terms as well as in the epistemological and evaluative terms applied by the judgementalist. To capture this aspect of emotional justification, Greenspan focuses on the role of affect. On her view, an emotion is a compound of an affective state of comfort or discomfort and the sorts of evaluative propositions referred to by judgementalism. These propositions are the objects of the affect (3). “There is danger” is the object of fear, “there is potential” the object of hope. The affective component gives emotions a role in motivation that is not dependent on whatever judgement may be involved—it is the comfort or discomfort of the emotion which gives its content the evaluative connotation it has, whether it is repulsive or desirable. This affective evaluation provides an extrajudgemental reason for action.

In fact the reasons provided by affect are justified differently than the reasons provided by the belief content of an emotion. It could be the case that there are not adequate grounds for belief which would warrant the action which is motivated by affect. For example, I may not have adequate reason to believe that a slasher lurks in the darkened alleyway, but my fear of the passage is quite sufficient to motivate me to seek another route home, and that fear needn’t rest on a settled belief that danger actually lurks. This is one element of emotion that Greenspan takes as evidence that they cannot be beliefs: one may experience the emotion whether or not they hold any particular beliefs about the matter.

If emotions are often in conflict with belief, this may bode ill for the prospects of a volitional theory of emotions. Since what is consciously willed tends to be paradigmatic of what is free, an emotion which implied propositional content contrary to one’s stated beliefs would seem to run against one’s will. But this assumes will to be conscious. If we allow that there may be unconscious intentions, it could be the case that one wills to respond with a different emotion despite considering such response inappropriate in light of their conscious beliefs. It may also be the case that such desires are quite independent of beliefs, answering to a more primitive region of psyche (Freud’s “id”). Alternately, one’s stated beliefs could be a complicated self-deception where one’s way of being conscious of their emotional intentions is in the mode of denial (Sartre’s “bad faith”).

Greenspan’s strategic characterisation of emotion seems to suggest a fair bit of control over emotional response. In “Emotions, Rationality, and Mind/Body” (2003) she describes a scenario where someone gets angry at a store clerk regarding a consumer complaint. The consumer has options available for emotional reaction, and reasons for taking either course. She may feel sympathy for the clerk or not have time now to pursue the issue, and so suppress her anger. But assenting to the anger may be useful in getting what she wants of the clerk (2003, 119). The verbal phrases in her story for these choices are “letting it go” and “letting it happen,” suggesting that emotion begins somewhat automatically and can from there be taken up strategically, although the emoter may also have a hand in setting up the conditions for the emotion’s activation (119, 125). But since Greenspan allows biology and socialisation to figure into the patterns of one’s emotional reasoning, it may not be the case that one is considering the strategic value of their emotion (124). The emotion could still be adaptive, and this may be why it is in the cultural repertoire, but one's reasons for assenting to or suppressing an emotion may not include considerations of adaptivity. One may refuse to anger, for instance, because it violates a religious precept, despite judging that anger may be very effective in the situation.

In “Emotional Strategies and Rationality” (2000) Greenspan states that emotional strategies may be subliminal, and are typically unconscious (2000, 471). In fact, it is generally necessary that they be unconscious as too conscious an emotional response would render it feigned rather than genuine according to our common understanding of “genuine” emotions.[9] The way in which one does bring about an emotional state is through the taking of a particular (judgemental) perspective on a situation, for example, to regard something as an offense and focus on its offensive aspect until anger develops (472). This intimacy with judgement does not entail that emotions are judgements of some sort, but that many sorts of judgement, in this case moral judgements, do typically result in characteristic emotion-types.

While Greenspan acknowledges many cases where it seems to be that one is in some regard choosing their emotional response, she remains generally agnostic as to how voluntary emotion usually is, insisting that emotions can be assessed in rational terms in any case (2003, 125).

iii) D’Arms and Jacobson’s “Anti-quasijudgementalism”

Justin d’Arms and Daniel Jacobson share Greenspan’s suspicion of the characterisation of emotions as judgements, but consider her alternative burdened with the same problems. In fact, they refer to her view as a form of “quasijudgementalism” because she continues to place emphasis on the propositional content of emotions[10] (d’Arms & Jacobson 2003, 130). Their objection is to the individuation of emotions in terms of constitutive thoughts (133). This type of individuation allows emotions to be constructed simply by combining a thought with an affect, resulting in excessively fine-grained distinctions between similar emotions. Their concern seems to be motivated by an assumption that it is preferable to have a few basic categories in one’s ontology of emotion, with variations falling clearly under one of the major categories.[11] Such a methodological preference is what motivates Griffiths’ insistence that talk of emotions is generally unfruitful because emotions do not constitute natural kinds (Griffiths 2003a, 290). Such a reductionist preference is certainly not shared by all philosophers of emotion—some are even willing to allow as many emotions as there are ways of expressing them (Campbell 1997).

D’Arms and Jacobson also side with Griffiths in terms of how emotions tend to come about. Proposing a small list of natural emotions whose existence can be explained by reference to evolutionary factors which would make such affective tendencies survival-promoting, d’Arms and Jacobson suggest that emotional episodes are activated by environmental cues operating on a response system which works independently of belief (2003, 138-139). This hypothesis, they contend, makes better sense of recalcitrance than Greenspan’s theory. In addition to emotion operating independently of cognition, they also claim that emotional responses motivate the evaluations often associated with them. They thus place themselves in the sentimentalist tradition, which holds that moral judgement is based on sentiment, rather than sentiment being the result of a judgement (144-145).

One way in which d’Arms and Jacobson’s account might seem to make better sense of recalcitrance is that recalcitrant emotions tend to be those which would have had an evolutionary advantage (141). Their paradigm is phobia, and the most common phobias include those of snakes, falling, and spiders. Fear and jealousy also count as natural emotions on their view, and are commonly recalcitrant. They contrast these with more conceptually-oriented emotions such as homesickness and religious awe which they claim never occur in opposition to their typical beliefs.[12] A further point they make against judgementalism is that if thoughts and emotions are both types of cognitive entities, it would seem to follow that cases of belief-belief conflict should be as common as emotion-belief conflict. They do admit that belief-belief conflict occurs, but presume that it is usually settled as soon as the conflicting beliefs are made explicit, which does not occur in the case of emotion-belief conflict (142).

Although at base emotions may appear essentially reactive, d’Arms and Jacobson admit that because of human emotion’s complex interaction with belief and judgement, they are subject to a degree of rational control not possible in other animals (144).[13] However, they deny that it makes sense to talk of justifying particular emotions. And since emotions tend to precede judgement or strategic thinking, rationality is not a standard by which they ought to be evaluated. Rather, the closest we can come to evaluating emotions is in terms of their fittingness as defined by the conditions with which one would normally associate them (145).

4. Justification

In this final discussion, I wish to look more closely at the possibility of evaluating emotions. If emoters are to be assigned responsibility for their emotions, there must be better or worse emotions to experience in different circumstances. The different theorists above hold different positions on how and whether an emotion can be evaluated, based on their theory of what an emotion is and how much control one has over it.

Both the most and least voluntaristic theories I have looked at have little to say about justification. The ultimate unjustifiability of everything underpins all of Sartre’s work, and the closest he comes to evaluation in Equisse is to take a slightly moralistic tone in regard to self-deceivers.[14] On Solomon’s account emotions set up rather than follow an evaluative framework, and so in one sense cannot be held to standards of appropriateness (Solomon 1976, 200). But there is a sense in which appropriateness could be considered as a relationship between the ideology normally associated with the emotion and one’s judgement of the circumstances in which it occurs (229). Finally, since Griffiths sees emotions as evolved and functional, there is a sense in which certain emotions may be more or less “fit,” but this measure will be made by the biologist, not by the ethical theorist or the rational decision theorist.

Calhoun holds that certain emotions ought to accompany certain beliefs, but refers to their disunity as cases of defective belief (142). According to her, emotion, rather than acting as an independent system of some sort, is the completion of a belief, its experiential component. Seeing something as X, where seeing-as is a characteristic way of feeling-towards, would be appropriate where something is intellectually believed to be X. But it only seems to be this failure of matching which would count as a case of inappropriate emotion.

De Sousa assesses emotions in multiple ways (the most explicitly ethical of which will be examined in Chapter Three). One is in terms of rationality. Rational assessment imposes on emotions the constraints that they be consistent and arrived at in an appropriate manner (de Sousa 1987, 162). Since rationality is a teleological concept, an emotion’s rationality will depend on its success. A successful emotion will apprehend its formal object (that which would make the emotion intelligible) in its target (158). The forms of rationality de Sousa considers as models for emotion are cognitive and strategic. Cognitive rational success would involve some kind of matching between representation and world. This criterion of matching is basically the same as Calhoun’s, and applied to beliefs and judgements. But although emotions are like beliefs and judgements in some ways, they are also like actions and wants (163). These tend to be judged in terms of strategic rationality, their success is dependent on how they bring about effects beneficial to the agent (164). With emotions, strategic success would depend on what sorts of results an emotion seems to aim at. Fear should help us avoid danger. Anger calls for justice. These emotions are strategically faulty when they bring about results opposed to their intent.

Finding that emotions have some resemblance to beliefs and judgements, some to actions and wants, but that they cannot be exhaustively assessed by the corresponding measures of cognitive and strategic rationality, de Sousa invokes a third measure, that of axiological rationality (169). As axiological, ethical discourse deals with what is worthy or valuable, the axiological success of an emotion is its success in bringing out the aspects of a situation and the options for action which are most appropriate according to a paradigm scenario (171-172). Take a scenario of a brutally beaten young woman being pushed out of a moving car. Assuming that my society values sympathy for the harmed, I should be horrified and saddened by what I see. I have failed if I see humour or opportunity to take advantage of this person’s vulnerability in this situation.[15]

Before registering my concern with this measure of success, I will mention two related views. One is Mary Warnock’s, who holds that emotions are justifiable in terms of the worthiness of their objects. Like de Sousa’s paradigm scenarios, worthiness seems to be a basically socially-agreed upon set of valued objects. As a consequence, emotions which have no name in a society, or which have abnormal objects, cannot be criticised or justified (Warnock 1967, 56). The other view I have in mind which depends on social norms in the assessment of emotion is that of d’Arms and Jacobson, described above. Since fittingness is a matter of normal association, the concept of a socially inappropriate, yet correct, rational, or laudable emotion is disallowed by their view.

Why restricting the assessment of emotions to standards of social convention is a defect of these theories can be seen by attending to Allison Jaggar’s piece, “Love and Knowledge: Emotion in Feminist Epistemology” (1989). Jaggar agrees that emotions are partly biologically, and partly socially constructed. She also allows for a degree of agency in one’s emotional response, comparing emotions to habits which one may be more or less successful in breaking (1989, 138).[16] Jaggar also agrees with de Sousa that there are social paradigms defining what counts as an “appropriate” emotional response (135). However, rather than ending with society as the final judge of an emotion’s worth, Jaggar goes on to stress that there is a political agenda being served by the establishment of emotional paradigms.

Those who do not respond to paradigmatic situations “appropriately” are said by Jaggar to experience “outlaw” emotions (144). Often such emotions, rather than representing a defect, suggest an insight into the flawed values upon which paradigms are based. That a woman is made uncomfortable by a comment intended as a compliment may be an indicator that such socially-sanctioned “compliments” play a role in subordinating their targets (144-145). In such a case an outlaw emotion actually helps one see through the official story of social reality and should be cultivated in the name of consciousness-raising.

The epistemic and political uses of outlaw emotions renders them amenable to being judged in terms of Greenspan’s “adaptivity,” a strategic measure. Greenspan also considers emotions in cognitive terms and in terms of the worth of their objects (1998, 119). But besides these, and the egoist strategies that seem to be the objects of adaptive measurements, Greenspan’s attention to the motivational nature of emotion allows for a more forward-looking assessment than other theories. An emotion may not just be maladaptive, but potentially (a potential she doesn’t bother to exploit) blameworthy as the first step toward a harmful action (153). This aspect of emotional assessment and the related topic of forward-looking emotional duties are the interest of A. C. Ewing who holds that emotions are not themselves the objects of our appraisal. Rather, what we judge is the active policy associated with these emotions, as well as the cultivation of emotional habits (1967, 71).

5. Summary

Biological theories have the benefit of grounding study of the emotions in observables and objective standards of fitness, but at the cost of denying the conscious participation of agents in emotional genesis. Emotions are postulated to operate through a system which responds more quickly than cognition, and therefore they cannot be thought of as chosen. This has the advantage of scientifically explaining recalcitrance, but is seems to call into question our tendencies to assess emotional response in moral terms.

Cognitive theories bring emotions back into the mental, but tend to have difficulties with recalcitrance, and are ambiguous about how much volition is involved in emotional response. There is also a tendency to leave out the phenomenology of the emotional episode, and with that much of what gives emotions their motivating force. It is this motivating force that Greenspan incorporates into her modified cognitive theory, giving her theory a way to combine a standard of assessment akin to the evolutionary standards invoked by Griffiths, and a story about how our beliefs are involved with our emotions.

Most of the theorists so far have limited themselves to the emotional agent in situation, or the static character, who may have a history, but is closed to future change. In the next chapter, I wish to look at the agent in development—how one comes to have the disposition one has, how disposition changes, how it may be judged. I wish to show that it is the relationship between our capacity for emotional control and its limits which creates the problem and the possibility of forward-looking emotional responsibility.

Chapter two: Character

1. Introduction

There are many phrases for describing the connection of one’s character to their action: “That is in his nature;” “From bad seeds bad fruit springs;” “Character is destiny.” There are also phrases which assume various kinds of links between inheritance or environment and character development: “the fruit doesn’t fall far from the tree;” “born criminal;” “product of her upbringing.” In this chapter, I examine some common assumptions concerning the relations between character and action, and concerning the nature of character development and change. I argue that while social and biological factors do tend to have a major impact on the development of character traits, they do so not in terms of impressing a sort of scar on one’s mental slate, but in terms of setting limitations and making certain choices more salient than others. To explain how this works, I use an analysis of identity formation and agency from Amélie Rorty and David Wong (1990). I also contend that where character is commonly conceived of by those with determinist leanings as a set of laws or internal forces (desires, passion, instinct) which propel one’s action, there is good reason to think that actions play a large role in developing a character. Support for this view comes from the existentialist-influenced theories of Robert Solomon (1976) and Michele Moody-Adams (1990).[17]

The term “character” is used variously, and I should clarify which among its meanings are relevant to my account. Character as integrity, reliability in a morally laudable sense is not my concern, although this sort “character” may be a trait of a person’s character. I also distinguish character from temperament, holding that temperament is an element of one’s character. I take character to be, as Moody-Adams describes, “an interlocking set of relatively consistent patterns of thought, emotion, and action” (Moody-Adams 1990, 128). While I adopt this part of Moody-Adams’ definition of character, her explanation that it “develops and persists because of the role it plays in sustaining one in a way of life,” is, I shall argue, only part of the real story.

In this chapter, I reject deterministic theories of character as some sort of force or structure that causes action. I agree with Solomon and Moody-Adams that character follows from action, rather than the other way around. However, I will temper their strong voluntarism with an explanation, mostly from Amélie Rorty, of how it is that character, while developing from choices, constrains and frames future choices in ways that tend to favour characteristic behaviour. Since it is emotion which ties together the various elements of character, the possibilities for emotional control described in Chapter One entail possibilities for control over character.

2. The Formation of Character

i) Social and Genetic Determinism

Given what has been learned, or at least what has been hypothesised, in the fields of sociology, psychology, and genetics in the last 150 years about the various determinants of character trait development, one may be tempted to deny the possibility that one could be considered responsible for their character. This deterministic worry has prompted scepticism about moral responsibility, and by implication threatens my own theory.

The inspiration for scientifically driven deterministic theories lies in the 19^th Century works of Charles Darwin on the biological side and Sigmund Freud on the psychological. Darwin holds that emotional tendencies are more or less fixed by inheritance (Darwin 2003, 58). While some habits may be developed during one’s life, perhaps by acts of will, the majority are innate (62-63). There are at least two problems with theories of genetic predisposition to certain character traits: 1) The idea of inherited character fails to explain the deviance of offspring from their progenitors. It is at least sometimes the case that two foul characters will parent a virtuous child; 2) A theory of “inherited temperament” has obvious negative political implications, being easily appropriated in the service of racial and sexual stereotyping.[18]

Freud, like Darwin, holds a theory of instinctual emotion (Freud 2003, 103), but expands the role of early-life development of particular emotional habits. These two giants are the major contributors to popular (folk) deterministic claims such as “it’s all instinct.” While Darwin’s belief that habit could be genetically inherited gave some hope that patterns of social behaviour could be changed, scientists no longer believe that we can inherit a learned habit.

More recently, the discovery of the neurochemical substrate of emotions has motivated a different brand of determinism. This trend is exemplified in the following quote from a text entitled Mind, Brain, and Society: Toward a Neurosociology of Emotion:

Humanists have sometimes argued that what privileged human life above nature was thinking—a matter, at base, of cognitive activity, but that…extends to concepts of soul and its secularized variants in related notions of self. From the viewpoint of modern neuroscience, such claims are inconsistent [my emphasis] with emerging understandings about the brain, and especially about the neurological foundations of emotion, reason, and memory… Naturalizing the human entails understanding emotion and cognition…within the same framework of material forces at work in our approach to understanding the physical world. (Smith and Franks 1999, 13)

If emotional tendencies motivate behavioural tendencies, and emotional tendencies are results of general chemical constitution, then behaviour is an effect of chemistry. Character in this case would refer to the kind of chemical make-up one happened to have. Of course, chemistry fluctuates. If emotions are environmentally responsive, the chemical change correlated to the emotional change would be a result of the environment, not just a neurophysiological accident.

A volitional model of emotion could accommodate the facts of our neurochemistry as well. If emotions are correlated with neurochemical changes, the changes in both are going to occur at the same time whether such changes are caused or willed. If one allows for a notion of unconscious will, then the fact that such changes occur below consciousness (or faster than cognition--see the discussion on Griffiths, Chapter One) does not show they cannot be voluntary.[19]

Social accounts of character and emotional development manage to avoid biological determinism by explaining development in terms of class and environmental conditions, and these conditions themselves as historically contingent. The ways in which someone responds emotionally to situations is based on that person’s acquisition of culturally appropriate attitudes (Armon-Jones 1986, 33). But though advantages are gained by this social constructionist approach, particularly in terms of empirical adequacy and its ability to explain differences between culturally different yet biologically similar humans, the explanation of the role of individual reasoning and feeling is left out. This can lead one to believe (along with the application of law-like statements to human populations) that development occurs through historical forces insurmountable by any individual.

ii) Identity and Agency

Acknowledgement of the influence of biology and society need not prevent one from also being able to explain agency in the development of character. Amélie Rorty and David Wong provide a taxonomy of the factors influencing the development of character traits. Their interest is mainly in terms of such traits as make up a person’s identity. The difference between identity and character, as I see it, is that “character” includes both objective descriptions of my tendencies and temperament, and the ways in which I am perceived or judged by others; whereas “identity” refers (in Rorty and Wong’s sense) to something that primarily consists of my perceptions of myself as well as others’ perception of me. So to describe my character, one must refer to things I actually do, and feel, while my identity could be radically disjoined from my actual behaviour. I may, for instance, identify as bisexual while never having indulged, or been particularly inclined to indulge in any sort of sexual practice with one of the two sexes to which I am supposedly attracted.

Why identity is so important to character is that people tend to try to act in ways consistent with their self-identity, and are encouraged to act in ways consistent with their attributed identity. If I identify as Christian I may attempt to live up to Christianity’s demanding moral precepts. If I am part of a Christian community, such striving will be encouraged and transgressions condemned. Whereas emotions frame choices, making some more appealing and easier than others, self-conception and how one is identified by others similarly frame emotions. To a large extent my feelings will derive from who I think I am and ought to be. I may feel things as injustices based on my understanding of the rights of a person like me. My image of myself may enable or debilitate me through opening or closing possible choices (Dillon 1997, 226). If I do not conceive of myself as the sort of individual capable or worthy of attaining goals, I will not be likely to form the emotional intentions required to try attaining goals. I shy away where I should pursue, I express envy rather than congratulation, I get mad when I should get even. Because the ways in which I think of myself and am thought of have this sort of influence on my emotional habits, identity is has an important influence on the development of character. Amélie Rorty and DavidWong (1990) describe six aspects of identity which influence character development:

a) Physical Traits

Among major contributors to our identity are our physical traits. Somatic traits such as body type, proprioceptive traits such as clumsiness or grace, and kinaesthetic traits such as agility or sluggishness all feed into images of a person and set limits on development (Rorty and Wong 1990, 21). The agile and graceful individual is able to excel in sports or dance, winning attention which boosts ego and confidence. Confidence itself breeds confidence, since one becomes confident through success, and can only succeed through trying, but won’t try without confidence. The clumsy individual is similarly channelled into the nerd identity. Social preconceptions of body types have just as much influence: the child with “shifty eyes” is treated with suspicion and so fails to form bonds of trust; the man who grows a beard discovers, though treated as a weak-chinned nerd in youth, that he now commands respect which makes it possible for him to act in confident ways and be taken seriously. The roles in which one is likely to be taken seriously play a major part in any decisions one is likely to make. If I know I won’t gain the respect of an audience as a dancer, because of my obesity, I may take on the role of clown and use my obesity to comic advantage. If my small stature and feminine voice are barriers to my being taken seriously in argument, I may have to use coquetry and manipulation to achieve my goals. In these cases, the identity I have developed due to my physical features has limited the ways I could effectively pursue my ends in the world such that I am steered into adopting certain strategies which eventually become characteristic.

b) Temperamental Traits

Shyness and confidence, as well as irascibility, generosity, and the like, are temperamental traits (22). To some degree, such traits may be based on a genetic predisposition, that is, one may inherit the tendency to produce more of one hormone than another. But the same physiological events can support different emotions,[20] so even if one has certain hormonal tendencies, the story of how these physiological occurrences manifest in the emotions they do needs to be completed by social and psychological stories. As with physical traits, temperamental traits carry with them a social mythology that tends to encourage development in one direction or another. Rorty and Wong put the point, “Because they standardly and widely elicit specific patterns of social responses, temperamental traits often tend to ramify to form clusters of mutually reinforcing dispositions” (22).

c) Social Roles

Another major contributor to identity is the existence of social roles (22). Not only are there expected roles to be played by those of certain body-types (for example, gendered roles), but there are also roles one is expected to play into or grow into given the wishes of one’s family, the contingencies of one’s life (for example, becoming a parent, or choosing a profession), or even one’s physical and temperamental traits (“you’ve got a friendly manner, ever thought about sales?”).

With each role one chooses or is channelled into comes a set of expectations about the proper character for that role, and hence patterns of salience and possibilities for recognition. My choices and sense of self will in part be influenced by the duties and capacities I see myself as having given my roles. These same roles will also influence how I am treated by others and how I am expected to respond. If I am a waiter, for example, I am expected to be quite gracious, and would not be able to respond to a rude customer as might otherwise be appropriate. In other circumstances, an angry response might be taken seriously and perhaps elicit an apology, but here it will be taken as impudence on the part of one whose job it is to cow-tow to customers. Similarly a wife might be expected to be consistently supportive, a grandparent generous, or a grown man brave. These expectations and those associated with socially-defined groups (below), have political implications which will be considered in Chapter Three.

d) Socially-Defined Groups

Related to social roles are socially defined groups which may have a set of roles associated with them. Socially defined groups include ethnic, gender, and class categories. Given one’s membership in each of these categories, there are certain attitudes one is expected to have, certain roles one is expected to be more suited for, and various stereotypes limiting one’s options of self-expression and self-realisation (23).[21] Again, if I am not seen as an appropriate candidate for certain forms of expression or living, my attempts to have my emotions, claims, and efforts taken seriously will fail. Such political and philosophical efforts as affirmative action and attempts to theoretically undermine racial essentialism are moves to eliminate the barriers to personal development imposed by group identities.

e) Ideal Identity

So far, most of the categories of identity have been characterised as somewhat given and socially imposed, but it is important to stress that it is not the merely the case that roles, social group, or cultural stereotypes limit one’s options for self-expression, they also make certain ways of being more appealing than others. The social role of parent, for example, is surrounded with a certain mythos which provides one who adopts the role with a sense of direction, accomplishment and honour. The ideological part of identity which sets standards for development is what Rorty and Wong call ideal identity (23). One’s ideal identity may consist of emulated role models (24) or certain highly valued traits, such as rationality or empathy, which one is striving to exemplify in one’s character (25). It is in striving toward ideals that one is most conscious and active in the development of their character. If I am striving toward confidence, I will undertake risky projects (the risk may range from rejection to death) whose success will bolster my confidence and bring me closer to my ideal.

One’s identity is clearly an important factor in determining one’s options for growth and change, as well as in providing motivations and a particular concern regarding decisions reflecting on one’s character. But the visceral imperative of choosing some options over others isn’t entirely explained by reference to conceptions of identity. Another aspect of character which plays into the patterns which establish character traits is one’s temperament.

iii) Temperament and Agency

Gregory Trianosky supports the view that character revision is possible and voluntary, but holds that the motivations one has for undergoing such changes are largely non-voluntary (1990, 99). One’s character, according to Trianosky, consists in their standing desires, attitudes, values, commitments, emotions, and their abilities and capacities for action (97). But he distinguishes character from temperament which is a brute fact about one, and is operative prior to any voluntary participation in character development. Qualities of temperament are too spontaneous to be considered emotions, attitudes, or motives (100). They include such things as tendencies to be easily influenced, natural sympathy, or hyperactivity. While the elements of character are maintained to some extent wilfully, through practice and perseverance (97), temperament is usually settled by the end of one’s formative years, or if it changes, it does so due to the effects of age or trauma.

I will return to Trianosky’s account in Chapter Three. For now I would like to question his description of the relationship between character and temperament. The fact that he places emotions in the category of character traits and opposes this category to temperament is a little confusing, since we often think of temperament as just being one’s emotional habits. Trianosky’s distinction can stand up if one defines emotions as having objects and cognitive content, and temperament as the feeling-tendencies which may guide the sorts of emotions one is likely to form. But temperament itself is developed in much the same way as the initial formation of character traits. It may have some physiological underpinning, but one’s experiences are as formative of temperament as they are of attitudes and emotions. A childhood of bullying may give me a defensive and insecure temperament, as well as an attitude of distrust in the existence of justice and goodness. It is true that attitudes are more easily revisable given new evidence, whereas the defensive temper may continue to announce itself in the form of emotional recalcitrance. But I do not see why, given that we know many of the factors that form temperament, it should be unrevisable.[22] If the sorts of conditions under which we change our characters are the same as those under which our temperament tends to change, it seems plausible that the latter may be occurring somewhat voluntarily.

iv) Temperament as Agency

Robert Solomon acknowledges that there is a regular connection between one’s emotions, in the sense of temperamental dispositions, and their characteristic behaviour, but that this empirical connection does not support the metaphysical thesis of determinism (165). Solomon feels that the different ways in which the same emotion or feeling may manifest in behaviour is best explained by appeal to the actor’s intentions—what it is one hopes to accomplish by the adoption of such behaviour (166-170). And since emotions themselves are strategies adopted for the maximisation of personal dignity and self-esteem (393), one’s temperamental tendencies can be thought of as a pattern of pre-reflective choices which then gets reflectively articulated in the habits of expression one adopts.

“Character,” then, is for Solomon (and Ryle and Sartre) a descriptive category consisting in the sorts of attitudes, emotions and behaviour one characteristically adopts. While Solomon allows that one may initially fall unreflectively into habits[23] which become characteristic, as soon as one recognises the sort of person one is, any future actions consistent with that character are endorsements of that character (20-23). Once reflection enters in, one’s self is consciously chosen. This is not to say that pre-reflective action is unchosen, just that reflection brings with it new possibilities for change and another level of responsibility (413).

Solomon’s suggestion that all character traits can be traced back to some sort of choice is either completely implausible or requires a stretched sense of choice. The defensiveness of the bullied person, the self-hate of the abused, the arrogance of the spoiled, and diffidence of the outcast all seem to be cases where one has developed a disposition quite despite themselves. In each of these cases but arrogance, it would seem that the kinds of emotional “strategies” that have been adopted have nothing to do with the maximisation of personal dignity and self-esteem. However, Solomon’s insight that character follows action, rather than causing it, is very useful. A less radical version of his view may allow that certain temperamental dispositions are quite involuntarily developed, but that these may be changed once brought to consciousness. In fact, Solomon has an account of how such changes occur, which I will describe in Part 4, below.

3. Character Creation and Violation

i) Magnetised Dispositions.

I agree with Solomon that one has some control over the development of one’s character, but I don’t find a satisfactory explanation in his account of how it is that certain undesirable patterns come to be established as character traits. Because he also holds that emotions constitute the meaningful world, he will allow that some behavioural options will occur to one more readily than others, but his description of this as some sort of instrumental reasoning does not explain the visceral imperative of emoting characteristically.

Amélie Rorty, in the process of trying to explain recalcitrant emotional habits, provides a complex and coherent account of the development of emotional patterns that explains the interaction between temperament and intention. Rorty proposes a principle of charity in the attempt to explain another’s emotions: we should assume that there is some rationale for why someone is responding the way he or she is (110). It may not be the case that what one is doing is rational or even strategic, but attention to the intentional component of an emotion will provide a useful guide for understanding its causes.

Rorty conisders three closely related factors as part of her analysis of the intentional component of emotion. These are: 1) Formative psychological events and “the development of patterns of intentional focusing and salience, habits of thought and response” (105); 2) The social and cultural stock of emotional behaviour and expression; 3) Genetically fixed sensitivities and patterns of response (105). While these factors have some value in the explanation of emotions, they fail on their own to explain the cognitive features of an emotion required for a moral psychological analysis. For this, we need to look more closely at the intentional component of an emotion.

The intentional component of an emotion is a person’s description of the situation motivating the emotion (107). So if I believe that my former lover is now in bed with my former best friend, this belief is the intentional component of my feelings of envy and betrayal. My intentional set is my disposition to respond this way in these circumstances. In cases where such sets persist despite changes in beliefs (such as belief in the appropriateness of such responses) or where responses are activated on inadequate evidence to hold the justifying beliefs, Rorty suggests that the causal history of the emotion may include the development of a magnetised disposition (108).

A disposition becomes magnetised through the development of mutually reinforcing habits of discrimination, intention, interpretation, and behaviour (113). These are also influenced by social and genetic factors, but what makes the disposition self-perpetuating is its intentional component and the tendency we have to fall into habits. What characterises a magnetised disposition is not only its persistence, but the tendency one has to look for occasions to activate particular intentional states (106). The jealous individual not only feels pained on discovery that their admired is enamoured of another, but constantly looks for evidence that this may be the case. The young romantic inclined to fall in love with each beautiful acquaintance falls easily and somewhat willingly.

Emotions tend to come with justifying intentional components. There is an object or aspect which one usually takes to be the (appropriate) cause of the emotion. If I am angry, I usually find some fault in those at whom I am angry. If I feel that everybody picks on me, I tend to cite instances which show that this is not paranoid. However, it is possible to start to feel unjustified in one’s habits. Such a feeling may be due to a belief. For example, my jealousy may be at odds with my belief that all people are perfectly free to choose who they sleep with, or even that I am not the most worthy object of my admired’s affections. My vengeful attitude may be at odds with my commitment to pacifism. One may also experience secondary emotions (121) which cause unease at one’s habitual responses, such as disgust at the effects of jealously on my relationships or shame at the ill-will I harbour toward others.

Because emotional patterns tend to self-perpetuate (due to their tendency to feel justified, and to their tendency to colour reality), even in instances where one’s disposition is not significantly magnetised, it can be difficult to change a pattern merely in light of new beliefs or values (121). This is simply an acknowledgement of the nature of habit. Two things are implied by claiming that something is a habit. One is that someone does something with a degree of regularity, the other is that this person tends to do so without a lot of conscious attention to what she or he is doing, somewhat like a reflex. Once an emotional disposition is habituated, it may require a wilful act of paying attention to abort or stifle habitual responses. But because emotional dispositions are like habits, and habits may be broken or established, emotional dispositions should be subject to the same sorts of origin and dissolution. The problem with explaining how this can occur is precisely that emotional responses, particularly magnetised ones, feel involuntary, and because of their persistence the first few times one attempts to respond uncharacteristically are likely to be frustrating experiences. However, people do often succeed in changing their habits, so the bridge between one’s current intentional set and one’s ideal intentional set (the way one would like to think and emote) must be traversable.

ii) Acting In and Out of Character

Neil Levy assumes that it is a mark of good character that one is able to see their flaws and act to change them. Conversely, he holds that if one has a bad character, it is because one is unable to see one’s own flaws, and so it is unrealistic to expect such a person to be able to change (120). It seems to me that this account results in everyone having static characters: bad characters do not change because they cannot; good characters seem to have no reason to change. In this case, it would be trivial to ascribe freedom to the good character, since both good and bad characters are functionally static. This theory, then, can be falsified on empirical grounds, for people do, in fact, act out of character.

Michele Moody-Adams takes examples of otherwise vicious people having moments of virtue as evidence that it is possible to act “out of character” and hence that character is not destiny. She admits the existence of constitutive luck, the unchosen influences and impressions on one’s initial character development, but contends that the existence of such circumstances is insufficient to undermine a belief in the ability to do otherwise than characteristically expected (112).

According to Moody-Adams, there is no “internal” cause of types of behaviour. It does not make sense to attribute characteristics which are not manifested. Since character traits are tendencies to feel and behave in particular ways under similar conditions, it is only through one’s behaviour in particular contexts that we can begin to make claims about their character. Since it is one’s actions which call upon one attributions of characteristics, it is under the category of action that character may be judged (114). This point is made in response to Thomas Nagel’s claim that it is irrational to judge character (explained in Chapter Three). If character were some internal engine pumping out actions, moral assessment of character might make no more sense than moral assessment of plumbing. However, since there is no evidence for anything which would count as a character trait outside the realm of action, character assessment just follows from the assessment of character’s constitutive actions: “A moral assessment of what a person is really like cannot rightfully be detached from what a person really does” (115).

Moody-Adams presents several difficulties for the conception of character as something internal and unrevisable. One is the metaphysical problem of what character is thought to endure in (116). One might meet her concern from a physicalist position: if persons or minds are identical to brains, character can be reduced to those neurological structures that cause the various behaviours manifesting the character. Moody-Adams might counter that this still requires the activation of particular neural events which leaves the question open: where is this character when these events are not in action? The physicalist might point to certain persistent structures which are what make the characteristic events more likely to occur than others. As character changes, old chemical bonds dissolve and new ones form, facilitating different sorts of behaviour. The best response to this might be just to point out that it is far too full of empirical claims to be considered seriously without some hard experimental data to back it up.

Another problem with the internal view is that it is based on the mistake that character trait terms are meant to pick out things. Rather, statements concerning character are “fundamentally elaborate inductive hypotheses rooted in the observation of patterns of behaviour, and of the circumstances in which such behaviour is displayed” (117). One may object to this that this description could apply to any sort of induction, and we do take many inductive conclusions to entail the existence of causally involved entities (such as molecular structures) persisting through the observed events. There are two responses to this: 1) there is not enough consistency in behaviour to treat it as one treats the objects of physics; 2) our understanding of moral agency requires that decisions be the sort of thing that is not strictly forced by antecedent causes.

This second point is why Moody-Adams thinks it is intuitively plausible that one’s character is not beyond the control of the will (117). Indeed, it seems to me that our condemnation and approbation of people’s character traits is not merely an irrational habit, but is conducted in part because it influences others to change or maintain these traits. Blame typically creates guilt; guilt serves as a secondary emotion with one’s actions, emotions, and character as its objects; this assessment motivates efforts to change. Were it not the case that we had some evidence that this chain can work, we should long ago have abandoned efforts at scolding, punishing, or reforming, and adopted the more prudent option of sequestering, or medicating, anyone who’d demonstrated a socially malignant character.

The problem of predictability is also invoked by Moody-Adams. The counter-evidence she employs against law-like generalisations of human behaviour is that people can and do act out of character. While she concedes that one generally is predictable in their behaviour, it is not because of any psychological laws or forces. Rather, life simply runs more smoothly when we can somewhat count on knowing what others will do in situations we may face together (118). While someone of any sort of character is likely to be able to find companionship of some kind, if only a partner in crime, those who eschew all manner of consistency will quickly find that they have no friends and may face the wrath of society’s fear of insanity.

By pointing out the rationality of consistent behaviour, Moody-Adams shows that consistency is in no way incompatible with the freedom to choose anew each time. She also insists that the social construction of possible ends and identities, and positioning of more or less appealing alternatives is not a problem for choice. Society provides the materials which are required for choices to be made. Its effect on character is that of providing the interactions and scenarios which teach the skills and ways of thinking that allow one to determine one’s course (127).

In Chapter Three, I will explain the consequences of Moody-Adams’ view of character for attributions of moral responsibility. For the remainder of this chapter, I wish to develop my own views on the ways in which a person may act out of character and how this may result in character change.

iii) The “Who” that is in Control

I’m not sure I agree with Moody-Adams’ adamant denial of something persistent in traits of character. I share her suspicion of the image of character as a sort of stamp on a substance, but I would not deny the utility of metaphysical concepts such as “self” and “will” in the analysis of moral psychology. Below I develop three key concepts in the analysis of uncharacteristic action.

a) One’s Right Self

Where one makes judgements and performs actions which are generally uncharacteristic, and this is due to some short-term extenuating factors, I will say that this person is not his or her “right self.” I do not mean this to imply something as strong as temporary insanity, as one may be quite in control in an immediate sense. The question is one of who is in control. “I wasn’t myself” may be a sensible statement if “myself” is taken to mean, “what I normally am like (when not faced with exceptional circumstances, when not on novel drugs, etc.).” Anita Allen points out that there is a certain level of provocation beyond which anyone would be expected to act and emote uncharacteristically (1997, 110). However, to say I am not my “right self,” such episodes would have to be infrequent. If I repeatedly perform these supposed exceptions, then after a while it is clear that I am not acting out of character, but that I am not who I was or thought I had been (113).

One’s self, I agree with Solomon and Moody-Adams, is largely constituted by what one does. But one’s character is not just a list of what one has happened to do so far, and one’s self is caught up in temporal issues (such as memory and future-oriented intentionality) which one’s actions, at the level of physical description, are not. Actions leave impressions of a sort. Confidence develops from repeated success, diffidence from failure. What I have succeeded in doing in the past I will be less reluctant to attempt in the future. Eventually, I may become so used to the unproblematic execution of particular sorts of action that there is no longer any reason to doubt or hesitate. This lack of hesitation gives a feeling of naturalness to what one does, making it seem to “flow” from their character. This growing confidence is where habituation can develop. If I have mastered a skill (be it sales, singing, or staying calm) to the point where I needn’t reflect on the way in which I do it, it may become the case that I will do it without thinking. For instance, the mannerisms and psychological tactics involved in being a good salesperson can become so natural that I become unwittingly manipulative in all my relationships.

Exceptions to characteristic action are therefore not only statistically less in character than one’s normal behaviour, but they lack the psychological habituation of one’s “real” traits. So while there is no essential self, there is at least a stable enough core set of dispositions to count as a self over periods of time. The patterns of behaviour and temperament we display in realising these dispositions, when attended to, form the themes of our personal narratives, and become part of our identities (Schectman 1996, 95). In distinguishing “in character” from “out of character” behaviour, this stability should be adequately long to contrast against uncharacteristic episodes such that it is easy to distinguish the norm from the exception.

Uncharacteristic behaviour takes two general forms: voluntary and involuntary. This distinction will be important for my discussion of responsibility in Chapter Three. The voluntary exception is easy to explain if one is at all influenced by the arguments of Solomon and Moody-Adams. One decides on this occasion not to do what one ordinarily does. But given the magnetic nature and motivating power of emotions, there seems to be a little more explanation than this required. What would motivate one to act out of character?

b) Knowing Oneself

One’s self-understanding is an important factor in both one’s ability/vulnerability to act out of character (in both good and bad ways), and one’s ability to change or develop character. The most obvious reason why people would choose to be good despite bad character is that they know what their character is like and don’t like it. In these cases knowledge of one’s tendencies is a condition of the possibility of modification. Knowledge of one’s personal history may provide explanations of how one has come to be the person one is. This itself provides clues for how to change.

Acting badly out of good character may result from a lack of preparation for dealing with the emotions which arise in a novel situation. It may also be part of a lapse in self-attention ranging from a mild case of forgetting oneself, as in slips of tongue and short flashes of temper to a more intense loss of control as in full-blown rage or paroxysmal weeping.

An example of a lapse which contains elements of both insufficient knowledge and a forgetting: I have been happily married for a decade, and am experiencing my first extended period away from my spouse due to business. We’ve demonstrated consistent loyalty and honesty, and expect our separation to be painful, but of no threat to our vows. I, however, not having been away from my spouse before, have not had occasion to be forced to reflect on the nature of my loyalty. It may be the case that I lose all the visceral imperative of my love after eighty-seven hours away from the beloved due to a need to rely on constant reinforcement of commitments. The first affair I have when I am away may be in part attributed to my lack of this knowledge of myself. The first time I allow myself to be overcome by lust may be a case of forgetting myself in Allen’s sense. Forgetting oneself will be explored further in under Point C, below.

An example of self-knowledge motivating an effort to change may be seen by continuing the story: after the delight of the first affair, I turn into a regular Don Juan, all the while lying to my spouse back home. Eventually, conscience catches up with me and I can’t escape my own judgement of myself as a bad, bad, person. This manifests either in compensatory behaviours, such as transferred acts of goodness, or in avoidance behaviours such as self-deception or self-annihilation (perhaps through alcohol). Eventually the situation becomes intolerable, the cheater decides a change is in order and starts by turning down the next opportunity to cheat. Or, perhaps if I don’t trust my willpower, I avoid situations of temptation. The ways in which changes in behaviour lead to changes of character is discussed in Part 4, below.

c) Forgetting Oneself, Losing Oneself

Not all uncharacteristic actions are voluntary. There are times where one really does feel overwhelmed by a passion. What it is like to lose oneself is to temporarily neglect all considerations but those directly connected with performing (and perhaps rationalising) a particular act. In more extreme cases, there simply is no rational thought in play, whether one is ecstatically gyrating on a dance floor, or savagely throwing their fists against an object of anger. Losing oneself is actually the goal of some practices, those aimed at the cessation of desires, as in meditation, or aimed at artistic expression, such as in musical performance.

Anita Allen describes losing oneself as a longer-term version of forgetting oneself (112). To forget oneself is to “inadvertently and temporarily…abandon the manners or morals—the communal norms of decorum or decency—that generally sustain both social approval and self-esteem” (104). It is what occurs in cases of violating etiquette and in extreme cases, it can become a part of what are termed “crimes of passion,” although Allen reserves the term “forgetting” for the less extreme cases (113). This is because “I forgot myself” is a kind of excuse, and not an adequate one for major crimes or outrageous outbursts. To say I forgot myself is in part to distance myself from what I have just done, to make clear that it should not be taken as a reflection of my character. I may be culpable for a lapse, but that it is a lapse is a fact that ought to soften others’ judgement of me (113).

The fact that it is possible to have such lapses generates a responsibility to pay attention to oneself (see Chapter Three on responsibility), and to watch both one’s feelings, and the situations to which one exposes oneself. Because of the magnetising tendencies of some emotions, each lapse puts one at risk of establishing a habit.

4. Character Change

i) Actions may only be considered to be “out of character” for as long as they are exceptional in occurrence. At some point, any habitual action comes to be “in character”. Both consistency and regularity are required for a trait to accrue to one’s character. If “loses temper in traffic” is a to be a trait,[24] it must be the case that its possessor: 1) tends often to lose their temper in traffic; 2) is in traffic enough to make such events somewhat common in their life. If a mild-mannered rural resident had only once been to the city and upon finding themselves in a traffic jam gave to shouting insults and honking, this would not suffice to attribute to them the trait “traffic-irritable.” However, if they began making weekly trips into the city and found the same thing occurred each time, they would have to admit (or flee from) the fact that they are indeed a “traffic-irritable” person. It is not that they have discovered a latent tendency, but have developed a new habit through a combination of a new regular circumstance with which they deal (for whatever reasons) differently from others.

ii) Character may also be changed intentionally (not just as a consequence of action which does not have character change as its object). Knowledge of self includes knowledge of what circumstances tend to bring out which aspects of one’s character. One may use this knowledge to alter the circumstances of their life such that they foster different habits.

Given what I have said above about the habituation of emotion and behaviour, and what has been revealed in Rorty’s account of magnetised dispositions, and Solomon’s account of “working oneself into” various emotional states, we should be able to draw some insights into how character change is possible. In individual episodes of emoting, I have such advice as, “count to ten before you get angry” and “don’t obsess” to go on. Given the influence of physiology on emotion, I can do such things as regulate my breathing in a single instance, or for a more general stress-out disposition, I could find outlets for this stress, such as working out, to diffuse the physical states which are contributing to the emotion. This does not guarantee any change in attitude, but it is much easier to change one’s attitude if the visceral force is taken out of it. This point was made many years ago by William James (2003, 74).[25]

Since beliefs also contribute to emotion, reflection on the facts to which I am responding can help, assuming there is a way in which I could see my emotion as unjustified. Of course this is by no means easy. Since emotions create patterns of salience, I am more likely to attend to those aspects of a situation which support the judgemental component of an emotion. Anger, resentment, jealousy, love, and self-deprecating attitudes all tend to motivate a search for confirming instances of the emotion’s appropriateness. It is precisely when I am already down on myself that I will pick on myself. It is when I resent someone that I am more likely to see insult in that person’s behaviour toward me.

These episodic interventions in one’s habitual tendencies can, with time, be entrenched as new habits. But given that one is, in these cases, often acting against what one’s heart (“heart” being a metaphor for the interaction of physiology and intellect) motivates, change will require a strong commitment and perhaps the adoption of a new ideology (Solomon 1976, 420). And since emotional patterns arise from complex interactions of belief, hormones, and society (among other factors), there is no quick-fix way to become a different person. Problems of undesirable emotional habits must be dealt with one episode at a time, and as the motto of many support groups says, “one day at a time.”[26]

5. Summary

What I have tried to show in this chapter is that character is not just a given base from which one launches into their behaviour in the world. It is formed through one’s behaviour, while tending to guide future behaviour. It is much like a path in the woods. A path is established because it has been walked. Given that it is established, it becomes the most salient route through the woods. However, it is still open to someone to forge a new path elsewhere, but it will take time to trample down the vegetation which is in the way.

If someone has developed emotional habits which are resulting in bad ethical choices or in self-destructive behaviour, it is insufficient to respond to criticism with “that’s just how I am.” While it may be easy to understand why someone responds in certain ways given the history of their development, the history does not justify the behaviour. Rather it shows that the answer to changing the behaviour lies not just in a simple decision not to do the same in the future, but a deeper revision of one’s dispositions themselves.[27]

In the next chapter, I apply these conclusions regarding character to the problem of attributions of emotional responsibility.

Chapter Three: Responsibility

1. Introduction

If it is true that we have control over the development of our emotional habits or episodes, we may be said to be responsible for them. Typically, we understand the causal sense of responsibility as one’s responsibility for a state or event that one has participated in bringing it about with some knowledge of its probability and a reasonable temporal distance between one’s contribution and the outcome. In the case of emotions one’s causal responsibility extends as far as one’s choices have contributed to one’s emotions and emotional tendencies thus far. I am interested in the ethical question of responsibility which takes two forms: one is the question of whether one is causally responsible in the sense described above with an eye to blame or praise; the other is the question of whether one has any future-oriented responsibilities to do or avoid praiseworthy or blameworthy actions or consequences. If one has participated in bringing oneself into a state of rage, and this state is counter-productive or harmful, there is a question of whether one ought to be blamed for this condition. The question of future-oriented emotional responsibilities is that of what sort of temperament we ought to cultivate. There are several reasons why I feel that it is important to be able hold people morally accountable for their emotions:

1) Since emotions motivate action through comfort and discomfort, and frame the range of choices we are likely to make through patterns of attention and salience, holding people accountable for the emotions themselves can steer them away from actually making morally wrong choices;

2) Since our readings of each other’s emotions are crucial to communication, understanding and trust in relationships, it is important that we have ways to assess whether people’s attitudes are consistent with their authentic participation in these relationships. It is also sometimes the case that some relationships are what they are by virtue of the emotions between the parties, while some emotions are not appropriate for certain kinds of relationship. Erotic love ought to be maintained toward one’s lover (considered in Part 6 below); it ought never be cultivated toward minors.

3) Since emotions often include cognitive content, it is important to scrutinise them to see whether the things they perceive in/believe of the world are really in the world. The cognitive dimension of emotion is of ethical interest particularly where the content of an emotion is a moral judgement. Since some emotions, such as anger take a blaming tone toward their objects, an erroneous judgement is akin to an erroneous accusation. If I am angry and you have done nothing wrong, it is unjust that you should experience my hostility. However, it is important to be especially careful passing judgement on an emotion in cases where the emotion conflicts with the common sense apprehension of a situation. Such emotions may be providing insights into an unacknowledged harm or value in the world. Since they have this potential, there is also an ethical dimension to the ways in which we interpret and evaluate the emotions of others.

In this chapter, I draw on insights from several different theorists to show how an account of emotional responsibility might be articulated—the factors it would need to consider (such as the relation of the agent to the outcome), and the terms it would have to employ, the terms of virtue theory. Part 2 deals with Claudia Card’s analysis of responsibility (1996) which I adapt to the case of emotions. Part 3 looks at W. George Turski’s (1994) and Edward Sankowski’s (1977) respective interpretations of emotional responsibility to establish the conditions that would have to be present for attributions of emotional responsibility. Part 4 asks which mode of ethical judgement emotions should fall under. I endorse, in this section, Ronald de Sousa’s division of the topic into four modes: emotions may be evaluated according to their motivational tendencies, their logical structure, their role in personal relationships, and as intrinsically valuable experiences (de Sousa 1987, 306). Part 5 looks at a few particularly ethically suspect emotions, and Part 6 raises dangers of condemning certain emotions in some cases.

2. Responsibilities

My investigation of our control over character and temperament in Chapter Two is a response to the problem of “constitutive moral luck” which supposedly poses difficulties for attributions of responsibility. There are other responses to the problem, some of which I survey after explaining constitutive luck. While I argue that control over, hence responsibility for, self-constitution in the realm of character and attitude is sufficient to avoid the problem, other responses will be useful for generating an approach to cases where one has lost control. In these cases of loss of control it may be appropriate to temper our attribution of causal responsibility; nevertheless there still remains the question of future responsibility. I will use Claudia Card’s response to moral luck as a resource for ascribing future-oriented responsibilities to emotional agents.

Thomas Nagel assumes not only that temperament is beyond one’s control, but that most other aspects of character are as well. Our habits, knowledge, inclinations, aptitudes, and attitudes are all a matter of our constitutive moral luck, the circumstances, upbringing, and biological factors that have made us who we are (Nagel 1979, 28). The realisation that constitutive luck and the circumstantial luck that places one in one’s ethical trials are largely beyond the control of the agent raises sceptical questions about the rationality of holding people responsible for their character and possibly even their actions (38). While my response to this concern has been to stress how much control we actually have over our own development, Claudia Card incorporates the problem of moral luck into an account of responsibility which provides an ideal model for forward-looking responsibility for one’s emotions.

Card’s discussion of responsibility and moral luck starts by distingushing forward from backward-regarding attributions of responsibility (Card 1996, 25). Backward-looking or retrospective responsibility is concerned with causal origins, blame, punishment, excuse and mitigation. Applied to emotions and the development character, a retrospective attribution of responsibility would point to an agent’s choices involved in the development of character and emotional responses. Forward-looking responsibility is the sort that is undertaken.[28] Responsibilities may be committed to directly and explicitly. Responsibilities are also entailed by certain roles. Insofar as one has undertaken the role, one has assumed those responsibilities, whether or not one has endorsed these responsibilities individually. Forward-looking responsibility accepts as given what has gone before, including those formative aspects of character that have been out of one’s control, and from that base commits to future action.[29]

Card acknowledges that people are given unequal opportunities to develop and express themselves morally. Rather than threatening attributions of responsibility, the presence of this type of contingency complicates the meaning of responsibility. It becomes a crucial factor in our attributions and calls for “humility and mercy” in our assessments (22). Given a history of bad moral luck, it may not be the case at some point that we are responsible for most of who we are thus far, but it is up to us to take responsibility for who we are in order to be able to take responsibility for our future (24). On this view, luck is not a threat to agency, but an element of it (39) which needs to be appreciated in order to deepen our understanding of responsibility (21).

Card specifies four senses of taking responsibility: To undertake administrative or managerial responsibility is to prioritise and strategize one’s possibilities. To be responsible in the sense of accountable is to accept that one is answerable for something and then being prepared to actually answer. Care-taking responsibility involves backing something and seeing it through. The credit sense of responsibility refers to backward-looking blame and praise (28).

The connections between the first three, forward-looking, senses of taking responsibility to the fourth are: 1) that it is on the assumption that one has accepted or been assigned responsibilities in the future sense that one’s failure or success in meeting these invokes attributions of credit; 2) that our practices of praise and blame serve as spurs to encourage others to take responsibility in the future senses (29).

The scenarios in which one takes responsibility are assumed to include many features outside of one’s control. It is because I have, whether through my own choices or not, found my life in a state of disarray that the possibility arises of taking managerial responsibility over my affairs. Applied to emotions, it is because my emotional disposition has rendered aspects of my life unmanageable that I need to take responsibility for them.[30]

I have offered Card’s response to the problem of moral luck to elucidate the senses of “responsibility” with which I am working. One further sense she mentions is that of a capacity—a responsible person as a person who is able to take responsibility (27). If responsibility has been lacking due to one’s suffering at the hands of one’s circumstances, it can be achieved by attempting to exceed one’s apparent abilities, and succeeding:

We develop responsibility as a virtue by first taking responsibility in ways that outrun our apparent present worthiness to do so and then carrying through successfully. Luck is involved in both our motivation to take responsibility and our ability to carry through. (27)

The establishment of an improved capacity for the exercise of certain moral skills is a kind of character development. Card’s model applies to responsibility for emotions in the following way: One takes administrative responsibility when one considers which present emotional habits are problematic, which deficient traits are worth trying to attain, and how these may be altered or pursued. One undertakes care-taking responsibility both in one’s emotional relationships with others, and in terms of one’s own “moral hygiene” (explained in Part 3 below). One is accountable for one’s emotions when one is prepared to answer for them.[31] When one is blameworthy or praiseworthy for one’s emotions will be explored in the next two sections.

3. Conditions of Responsibility

W. George Turski proposes that emotions need to be considered in terms of responsibility because of their motivating power, ability to frame intentions, and importance to selfhood and identity (Turski 1994, 99). He is unconvinced by neurophysiological or metaphysical deterministic arguments that would characterise emotion as wholly passive, charging that most such arguments are loaded with question-begging theoretical assumptions, such as definitions of emotions which rule out the possibility of describing their cognitive content (101). He also admits, though, that traditional accounts of voluntarism are too simple and prone to counterexample (102). In particular, he considers the internal cause/external cause model of framing the issue of retrospective responsibility incapable of accommodating internal involuntary causes, such as neurosis, and the problem of some external contingencies being involved in voluntary actions.

Turski proposes instead that we base attributions of responsibility on the nature of the relationship of the agent to the outcome (103). The conditions under which an agent has a certain response or forms a certain habit, and his/her response to those conditions, are relevant to the determination of their responsibility. Under conditions of extreme pressure, we would not expect an agent to behave ideally. There is a point of teasing after which anyone can be expected to lash out at the culprit, and a point of grief where one is not expected to stoically meet all the duties of one’s life, at least for a while. But when someone grieves because s/he broke an emotional rule of the scientific professional and named a lab rat which now, as planned, is dying of cancer, we might accuse that person of having been the author of his/her own emotional hardship.[32]

Turski is interested in striking a balance between the fact that we have a large degree of control over our emotions and the fact that for some emotions to retain their value, they cannot be simply chosen. Grief, for example, must be something we are somewhat passively undergoing if we wish to secure the sympathy of others (111). Love would lose its lustre lacking luck. Could we fall for just anyone, the sense of being blessed would fall out of love’s experience.

Emotions, on Turski’s account, are alterable in all the ways mentioned in Chapter Two, but our responsibility for the future of our emotions is based in our nature as agents with the capacity to apply distinctions of worth to our desires and actions (112). According to Turski, articulation or assessment of an emotion alters the emotion in some way, and generates second-order desires in relation to it (113). Because of this, the act of assessing one’s own emotion is a way to keep emotions in check.[33] Self-evaluation on this model becomes a kind of moral hygiene, a responsibility to keep oneself in good character, rather than an after-the-fact trial of guilt.

The relationship between assessing an emotion and changing an emotion is an important one. Our psychic life is a constellation of nodes which are constantly emitting waves. The result of the interaction of so many frequencies may appear cacophonic, but knowledge of which elements are responsible for which frequencies may help to produce some harmony. If I am generous, but not to a fault, mild-mannered, hard to offend, but extremely impetuous, I may find myself in trouble that could not be attributed to any of those traits except impetuousness. Knowing this, I can keep an eye on it, “It really seems like a good idea for me to go to this party, but I always think it’s a good idea to do whatever gets suggested first. Is it really a good idea? How will I feel about not getting my work done yet another night?”

Edward Sankowski supports the idea that there is a link between self-knowledge and self-control:

…if persons are responsible for a sufficiently wide range of behaviour, and since their behaviour can be oriented in intelligent ways to regulate certain of their emotions in the light of the rational assessment of emotions that they can make—it is quite defensible sometimes to apportion responsibility to a person for an emotion which he [sic] feels. (Sankowski 1977, 835)

Any attribution of responsibility for an emotion or for future emoting will require that the agent have the capacity to assess their own emotions, knowledge of how to regulate their emotions, and the means for regulation (835). The regulation of emotion is the way in which one modifies, prevents, eliminates, or cultivates an emotion or emotional tendency (836). The means available for regulation include things like possibilities of expression and a lack of harsh or coercive conditions (837).

Sankowski distinguishes attribution of responsibility from assessment of worth. An emotion may be bad, but not blameworthy because it is out of the agent’s control. A character trait may be commendable, but not credit-worthy because the agent did no work to establish it (838). Such cases of luck fall outside the scope of responsibility.[34]

An additional condition of responsibility I wish to propose has to do with being able to distinguish between one’s usual character and an exceptional period of time where one is not able to rely on their emotional control. This can be under conditions of extreme pressure as mentioned above or result from a combination of external or self-induced factors that lead one into a “bad headspace.” States like paranoia and jealousy which seek confirming instances can be particularly prone to self-perpetuation. Gloom results in lowered self-esteem which turns one’s thoughts to one’s own shortcomings which deepens the gloom. Based on what I said in Chapter Two about the formation of character—that emotional episodes and attitudes which become habitual result in a character trait—these sorts of exceptional states should be considered out of character, at least the first time they appear. Blameworthy emotional states which are exceptional in this way may be considered to be part of one’s background moral luck.

However, once one has experienced a loss of control of the kind described in Chapter Two, one now has new knowledge about oneself—that one is capable of such episodes under such conditions. This creates the possibility of taking responsibility for avoiding future incidents.

4. Attributing Responsibility

Since backward responsibility is concerned with external assessments of an agent’s conduct (as opposed to the internal assessments of planning and moral inventory) (Card 1996, 26), the terms of assessment involved have the character of praise and blame. In Chapter One, I mentioned some of the different ways in which emotions may be evaluated, and in Chapter Two, I explored the various influences involved in the development of emotional habits and distinguished entrenched from episodic emotions. I wish now to frame the different types of evaluation described in Chapter One in terms of their measure of concern with future or retrospective responsibility.

As seen in the first chapter, Chesire Calhoun, Mary Warnock, Justin d’Arms, and Daniel Jacobson, insofar as they evaluate emotions, do so retrospectively, though this evaluation may not imply responsibility per se. Calhoun’s concern is that an emotion be appropriate to its accompanied belief, but she describes the failure of this as a cognitive defect rather than a sort of moral failing (Calhoun 2003, 242). Warnock would have an emotion be appropriate to its object (Warnock 1957, 54). D’Arms and Jacobson would assess emotions in terms of evolutionary fitness (d’Arms and Jacobson 2003, 145). They allow (along with Griffiths) that emotions may be assessed according to social norms, but deny any further normative sense in which they may be judged, and would not consider “responsibility” a suitable term to apply to what are essentially unwilled physiological events.

Patricia Greenspan evaluates emotions in terms of rationality and adaptivity. The rationality of emotions is evaluated retrospectively, in terms of their adaptivity in anticipation of present and future scenarios (Greenspan 1988, 83-85). The question of what I ought to be feeling may be answered in terms of cognitive fit--for example, “Should I be afraid?” admits of the same answer as “Is there something to be afraid of?”--but the question of adaptivity requires a look at possible outcomes (160). Since emotions act as spurs to action (152), the most adept emotion will be that which encourages the best course of action in a scenario (160).

It is not entirely clear how Greenspan’s strategic account translates morally, but it would seem to justify some form of virtue ethics. The excess and deficiency of bravery is a good example. Rashness is maladaptive because it can get one killed, or start unnecessary battles. Cowardice is maladaptive because it fails to spur one to required action. Often the paralysing fear of cowardliness leaves one vulnerable to being actually harmed.

Greenspan’s strategic account allows emotions to be evaluated morally in the same way as one would assess character traits,[35] but is there a way to assess them as one would actions or decisions? A. C. Ewing says it is the active policy associated with emotions which concerns us morally. He grants emotions a motivating power and allows us control over them via our ability to control their expression and adopt habits of thinking. From this control he suggests that there may be a duty to try to make ourselves feel certain ways in certain situations. However, he also stresses that what we usually assess is not the formation of an emotion, but the action to which it leads (Ewing 1957, 71). It is only secondarily that we associate emotions with ethics, because of tendencies of certain emotions to result in certain acts.

Ronald de Sousa argues that emotions can be judged morally, and that there are four distinct principles by which judgement may be made. I find his account well-suited to my theory because it allows the consequences that follow from the fact that we have emotional responsibilities to be considered in the language of several different major ethical traditions: deontology, consequentialism, care, and teleology. Although he doesn’t explicitly state so, I find that each of the four modes of assessment he describes—motivational aspect, universalisability, role in relationships, and whether some emotions are intrinsically worth cultivating—neatly correspond to the above four ethical traditions. By adopting de Sousa’s multi-faceted approach, my theory of emotional responsibility will be articulable through any of these ethical languages.

The first mode of assessment de Sousa describes, similar to Ewing’s standard, is to judge an emotion by its motivational aspect. That which tends to motivate moral behaviour would be the more moral emotion (de Sousa 1987, 306). Altruistic emotions would be paradigmatically moral, but de Sousa also argues that personal feelings, such as admiration for another’s wit or pride in one’s own, should also sometimes be counted as morally relevant. Altruistic emotions play in decisions of what one ought to do, but personal feelings have to do with human thriving, which is just as much an ethical concern (308). The personal feelings also maintain our relationships (309).

Given that emotions create frameworks of salience, the development of an inappropriate emotion will skew the apprehension of a situation such that the correct action is less likely (319). As well, we will fail to discover the emotion’s inappropriateness. If we take into account both what Solomon has said about emotion’s role in creating value in the world and the theoretical consensus that emotions are partly evaluative, the initiation of an emotion increases the likelihood that the emoter will continue to see the response as appropriate. This tendency is due to the double effect of finding a moral property and accentuating (or sometimes projecting) moral properties. Such properties could be violations of principles of justice, personal slights, or, for de Sousa, axiological paradigms. In any case, since emotions carry their fuel in this regard, it is especially important that the emotionally virtuous person avoid hasty emoting.

The second way to approach moral justification of emotions is to put them to a Kantian test and check for universalisability (309). If one has a second-order desire that a first-order emotion occur under relevantly similar conditions, this supports its claim to moral appropriateness (311). If I am slighted, for example by a co-worker’s failure to say good morning and treat her coldly for the remainder of the day, I could reflect on whether such momentary lapses in manners (especially common in a hurried work environment) should always warrant such persistently mean response. It occurs to me that if all breaches of etiquette were met with such hostility, the workplace would becoming a tense and unnaturally formal environment, which cannot be willed since the whole point of saying “good morning” is to foster a friendly, relaxed atmosphere.

The link between emotion and character becomes clear when we apply a principle of universalisability in an emotional assessment. If I feel comfortable with the idea that I will respond this way in all relevantly similar situations (note, de Sousa’s principle is agent-relative, unlike Kant’s (310)), I can be said to have endorsed a disposition to such response as a trait of my character. This endorsement is no guarantee of its appropriateness, however. I may feel as prepared to endorse a negative trait as a positive. But this test does help make sense of whom we are to consider the “real self” of the agent in question (see Chapter Two, Part 3, Section IV). At least the kind of actions an agent tends to endorse counts toward our assessment of that person’s motivations. Of course, one may endorse a principle of behaviour and fail to exemplify it, in which case the nature of our judgement changes slightly. If one were to endorse a pattern of irascibility, we would judge that person bad. Were one to endorse patience and regularly lose one’s temper, we would judge that person weak or deluded—the deficiency is not in basic moral attitude but emotional control.

The third moral measure of emotions involves their role in various sorts of relationship (312). Certain emotions are not only expected between members of certain sorts of relationship, but some emotions are necessary to make these relationships work at all. An obvious example is between lovers, but all friendships involve some sort of expectation of feelings. I may have friends who are very sympathetic and some who are very self-absorbed, but if these friends were to suddenly reverse attitudes, I would no longer be sure how to interact with them. The sympathetic friend who turned egoistic, in particular might cause me a lot of hurt. This consistency is part of what constitutes integrity (Card 32). Integrity is that aspect upon whose basis judgements of reliability are assessed, both by others and by ourselves (33).

Our personal relationships are also dependent on reliability in our moral sentiments. Considerations of justice that might make anger appropriate in most circumstances might compete with the need to maintain good relations with others. It is a different thing for a bank to hire a collection agency to harass a debtor than it is for a friend to collect from a close friend. Given the rarity of perfectly compatible comrades, more lenience in some areas may be required if we are to enjoy the pleasures of our friends’ better qualities. There are also attitudes of trust, authority, and varieties of respect that make particular relationships function properly, specifically those with purposes such as caregiving, managing, educating, and so on.

De Sousa’s fourth moral measure invokes the concept of human thriving to argue that some emotions are valuable just for the experience of them (de Sousa 314). For instance, joy is not good based on any function is serves, it is something at which we aim. Of course, due to other considerations, a given instance of joy may be very bad. One should not feel joy at the tragedies of others, for example.

In sum, on De Sousa’s view, emotions may be assessed consequentially in their role as motivations, deontologically as universalisable, in a care perspective as the cement of interpersonal relationships, and teleologically as constituents of a good life. Combined with his bootstrapping theory of how emotions may be controlled and modified (see Chapter One), De Sousa’s account allows for the ethical assessment of emotions in backward and forward looking terms and provides guidelines for responsibility and the determination of imperatives.

De Sousa’s attention to the intrinsically valuable experience of emotion suggests an imperative to avoid unpleasant emotions (314). Of course, some unpleasantries will be necessary according to other principles. Guilt is not pleasant, but it is a spur to avoid repeated transgressions, and it is the mark of a conscientious character. Also some emotions while technically unpleasant are integrated into the allover experience of some of our most emotionally rich projects. Frustration in pursuit of an artistic skill sweetens the experience of its mastery. Make-ups make up for the fights in our love affairs. Success presupposes doubt. But some emotions are “wholly nasty” (315). According to De Sousa, some emotions, like envy and despair serve no ethically worthwhile role. I will consider why he may hold this in my discussion of dangerous emotions, in Part 5 below.

5. Nasty Emotions, Dangerous Emotions

Given that emotions are often appropriate or inappropriate given very particular circumstances, it would be hard to clearly sort them into groups as good or bad (de Sousa 317). Nonetheless some emotions almost always fail on any scale of measurement, and some are so deeply nasty that no way in which it might be successful or positive could outweigh its badness. An example de Sousa uses is racist sentiment, which is reality-distorting, negatively motivating, and if somehow or another the racist takes delight in his/her emotion, this experience increases rather than lessens its viciousness (316). Other examples include envy and despair.

I have used the example of jealousy as a nasty emotion several times throughout this thesis. That is because I find jealousy a particularly salient example. It is politically suspect—it implies the treatment of others as property. It is decidedly unpleasant to the emoter and only pleasant to cruel others. It is strategically irrational, since it aims to hold a particular sort of relationship in place, yet tends through its offensiveness to drive relationships apart. It tends to magnetise since it seeks confirming instances of the suspicion it inspires. But perhaps its deepest danger lies in its close connection with the riskiest of emotions: those associated with romantic love.[36]

The joys of infatuation are lauded in verse and song. It’s character as a sort of discovery makes pursuit the most salient of options. “Falling in love” is so exciting and full of possibilities (the first several times), and surrounded with myth, that what control one has over entry into this emotional space[37] is used to propel oneself all the faster. Though love which turns out successfully may be one of the most wonderful of experiences, one of its characteristics is a sort of personal investment that opens one up to feeling the visceral imperative of each love-related emotion at an intensity one may not otherwise experience. As loss of a beloved is so much more significant than loss of a lesser-loved, grief can be devastating, and jealousy can arise to remind one of the possibility of loss. Betrayal by a beloved also cuts more deeply than any other kind. The “crime of passion” which is often a result of the sort of loss of control discussed in Chapter Two is typically thought to occur between lovers, or love rivals.

But there is risk involved in any relationship (Card 30). There are also sometimes reasons to indulge in the worse emotions, at least on occasion. And sometimes the seeming inappropriateness of an emotion is due to a faulty apparatus of expression and interpretation. In the next section, I explore Peter Goldie’s defence of jealousy, which shows that commonly vilified emotions may have discreet values. I also respond to some concerns raised by Allison Jaggar, Elizabeth Spelman, Sue Campbell, and Jacqueline Zita regarding the dangers of labelling certain emotions improper.

6. Dangers of Labelling

Care needs to be taken in labelling an emotion inappropriate, dangerous or “nasty”. The virtues of an emotion may be hidden by prejudice or conservative interpretation. Because emotions play a role in agency and because they contain judgements, it is incumbent on us to we treat the emotional expressions of others seriously. One must assume appropriateness until a clear critical examination is complete. Otherwise there are the dangers of missing a potentially valuable insight, of repressing political speech (Spelman 1989, 272), and of undermining the confidence others to have their emotional expression understood as they intend (Campbell 1997, 172).

As an example of an emotion which is often prematurely judged as inappropriate, I offer Peter Goldie’s defence of the classically derided emotion of jealousy. He agrees with its critics that it is an emotion prone to excess, due to its vividness and sense of powerlessness (Goldie 2000, 231). However, jealousy can be an exciting element of a relationship, giving it experiential value, and it is not the only emotion capable of excess.

If two people are in a relationship where each has committed to do what is possible to remain together monogamously, Goldie feels jealousy is an appropriate response to perceived threats to this commitment (237). In fact, I would agree that if it is part of a couple’s agreement to be exclusive that jealousy may be a perfectly appropriate emotion, though still capable of being excessive. If a commitment has been made to stay together in an exclusive relationship, not only might jealousy be appropriate as a judgement of a breach of the agreement, but each party may also be said to have a number of future-oriented emotional responsibilities.[38]

Another emotion that is often taken to be unjustified is anger. Sometimes the reason why anger may appear unjustified is that the angry person has difficulty naming the object of her anger. This is often the case with people in subordinated social positions, whose ability to name the object of their anger is limited by the ways in which the roots of injustice are hidden in an oppressive society (Spelman 1989, 267).

Alison Jaggar stresses the role of emotions in shedding light on hidden unjust aspects of social reality. Often women and other people in subordinate positions experience anger and discomfort toward scenarios which are not taken to be the appropriate objects of such emotions according to social paradigms (Jaggar 1989, 144). These “outlaw emotions” may sometimes be insights into social harms which are rendered invisible by official accounts. Since the power to designate what counts as appropriate or not lies in the hands of the dominant, the subordinate may lack the resources to explain the experiences to which only they, from their marginal position, have access (142). Furthermore, outlaw emotions may be felt by anyone in situations where a subversive sentiment challenges a received view. In my example of the laboratory rat-lover above, it appears that the scientist is blameworthy for her emotion, but it may be the case that she has discovered an injustice in the subjugation of other species to human research.

Still, Jaggar agrees that emotions are assessable. The measure of appropriateness she suggests is in terms of a sort of thriving:

[…] emotions are appropriate if they are characteristic of a society in which all humans (and perhaps some nonhuman life too) thrive, or if they are conducive to establishing such a society. (147)

Note that this principle is both teleological and deontological, and can be derived from de Sousa’s second principle, that of universalisability, and his fourth, that of the pursuit of intrinsically valuable emotions, except that here thriving is general and presumably includes such emotions.

Among the emotions which are both outlaw and justifiable on the above principles, is anger. While anger is generally accepted in cases where one has been wronged, the wrong done is not always obvious to the community of interpreters and many legitimate cases of anger get dismissed. Elizabeth Spelman (1989) points out that anger is seldom regarded as legitimate when it is the anger of someone in a socially subordinate position toward groups or members of dominant groups (264). This is because to be angry at someone is to pass judgement on him, which implies an attitude of at least moral equality on the part of the judge (266). This is threatening to the socially dominant, and denial of the right to be angry is a means of excluding subordinates from the category of moral agency.

Spelman suggests that there may in fact be an obligation on the part of those in oppressed groups to get angry, but cedes that the discouragement that exists makes it understandable that people would be afraid to express their anger (269-270). Sue Campbell (1997) looks closely at the barriers placed on emotional expression by those in a position to interpret an emotion. The denial of anger to certain groups does not just occur through the agent’s anger being judged inappropriate, but also through refusals to recognise anger as anger (166). Rather than being recognised as a moral emotion with cognitive content, anger is sometimes interpreted as rage (Spelman 271) or as bitterness (Campbell, 166).

Bitterness is taken to be a sort of faulty emotion, or in less voluntaristic terms, a “disease of the affections” (166). So bitterness is not just a term for an illegitimate form of anger, but is more of a character-trait term which posits the origin of the correlated emotional response within the emoter instead of within the environment. Anger is a response to a harm or offense; bitterness is taken to be the state of a person who won’t get over anger.

Campbell argues that the effect of uses of the label “bitter” is to silence the grievances of those who are not in the privileged position to say which grievances get to be heard. It also characterises the emotional lives of women and oppressed groups as unhealthy (167) and thus deserving of treatment rather than uptake. This problem hits close to my theory, since I have been recommending self-improvement to deal with disruptive emotions. What these and the following example show is that the sphere of emotional responsibility needs to be expanded to include our responsibilities toward our treatment and interpretation of the emotions of others.[39]

Jacquelyn Zita delves further into the problematic use of “self-improvement” to deal with one’s character and temperament. She agrees with Jaggar that many emotions, rather than being neurotic or ill can be indicators of social problems (Zita 1998, 75-76). In her essay, “Prozac Feminism,” she focuses on the rise of pharmaceutical solutions to emotional problems. In particular, she shows how drug companies have co-opted the language of feminism in advertising aimed at selling temperament-altering drugs, such as Prozac. Marketed as a solution to the problems of trying to be a “successful” woman in the face of strategically faulty emotions, such as depression and compassion, Prozac is claimed to have the interests of feminism in mind, while in fact promoting conformity to hegemonic norms (62-63). The bold, quick-witted, multi-tasking libidinous self promised by Prozac is precisely the model of womanhood desired by the business world and straight male consumers of sex (62). A further danger of this pharmaceutical trend is the medicalisation of personality. Subversive emotions are now not just outlaw, but sick (69).

I see several reasons to be concerned with the medicalisation of personality which are compatible with my view. One is that treating emotions as illnesses undermines their voluntarism, and medicating them gives up the control one ought to be cultivating. Another is that it denies the cognitive content of emotions, treating those which motivate anything less than endorsement of the present social order as kinds of disease.

Another reason that a pharmaceutical response to my argument for emotional responsibility is incompatible with my view is that the illness model fails to distinguish between moral and immoral emotions. “Health” is measured essentially in terms of fitness for participation in the status quo. Such a standard offers no possibility of progress and potentially licenses intuitively deplorable results. For instance, if it is permissible to drug oneself to fit into a system, where this drugging may remove one’s ability to spot the moral shortcomings of that system, or if it is permissible to prescribe this route, what are the acceptable limits of drugged social participation? Would any social system be successful assuming the drugs to make it livable were made available to all? This is not to say that it is never appropriate to take drugs for the sake of temperament. But it cannot be the first answer or the final solution in all cases.

I think Jaggar, Spelman and Campbell successfully show problems with assessing emotions hastily or without attention to bias, and Zita gives reason to be concerned about how one is prepared to deal with problematic emotional tendencies. Rather than posing a threat to my insistence on our capacity to be held responsible for our emotions and future emotional duties, these authors, along with Goldie’s defense of jealousy, have posed problems which either allowed me to articulate the limits of my view, or show how it encompasses their concerns.

7. Summary

Again in this chapter, I have stressed the motivating power of emotions as one of their most important moral features, but we have seen that they admit of assessment along many other axes. Some emotional responsibilities are implied by roles, some by their logical structure, some by considerations of thriving. In the assessment of an emotion by one of these standards, it is insufficient to look superficially or faithfully rely on the received paradigm. An emotion may be conducive to thriving, though it functions disruptively in the short term, as in cases where inarticulable anger aims at indicating a social barrier to thriving. Role duties may be undermined by a deeper ethical insight, as in the case where a researcher feels guilt for inflicting a disease on a lab animal.

Given that we are malleable, and can change who we are through conscious and self-deceptive means, spiritual or chemical, we have both the responsibility to take charge of our emotional lives in positively motivating, endorsable, meaningful ways, and to be reasonably critical in our judgements of our own and others’ emotional behaviour.

[1] The fuzziness of the borders between “emotions” and “moods” and related concepts is one of the difficulties in philosophy of emotions that motivates Paul Griffiths to argue that “emotion” is a philosophically useless category. According to Griffiths, the best conceptual analysis can reveal about the emotions is what people currently think about the emotions (2003a, 287). The larger question of what emotions are could only be answered if emotions formed a natural kind and were therefore analysable through science. But the fact that no cognitive, environmental, physiological, or behaviouristic constants hold between all cases of what are taken to be emotions shows that there can be no objective ground for distinguishing an emotion from other psychic or social phenomena.

[2] The legitimating acknowledgement of an emotional expression. Explained in Chapter Three, below.

[3] A memorable example from my childhood was learning not to cry. I couldn’t for the life of me understand how being accused of being a sissy would do anything but make me cry more, but eventually I caught on and learned to curb what I had assumed was uncontrollable.

[4] Translated into Engilsh as Sketch of a Theory of the Emotions, or The Emotions: Sketch of a Theory.

[5] Calhoun’s interest in the emotive element of coming-to-understand suggests that she is committed to some beliefs having a phenomenal quality, which would make them experiences on one level of description.

[6] Not “affect,” but “effect” as in “make be.”

[7] It has often been argued that this needn’t be the case, perhaps most successfully by P. F. Strawson in “Freedom and Resentment” (Strawson 1974). I agree that it would still make sense to assign praise or blame in a deterministic world, but I also believe that the two are incompatible, which is to say they can not make sense at the same time. What I draw from this is that determinism could not make sense, even if it were true.

[8] Despite the evolutionary ring of the term, Greenspan is referring to an emotions value in particular scenarios, not in terms of its species-survival promotion.

[9] The success of the emotion would likely also be undermined by its being spotted as affected, rather than sincere.

[10] They also suggest that Robert Solomon be considered a quasijudgementalist because in 1988 he admitted that beliefs and emotions can conflict (130).

[11] Another objection to judgementalism, one with which I am more sympathetic and would apply against cognitive theories generally, is that it seems to exclude infants and dumb animals from the category of emoters. This consequence does not appear to be one of large concern in the literature, though, and I suspect that what I might want to call “emotion” in the dog and child would be reduced to “feeling” by most theorists. I think the cat may be angry that someone stole his favourite seat. To make sense of this I could either accept the propositional theory of belief and deny the cat beliefs, forcing me to reject many cognitive theories, or I could say that the cat has a non-propositional belief that his territory has been violated, allowing for a dumb-animal-and-baby-friendly account of the cognitive element of emotions.

[12] I regard this particular claim as somewhat dubious, having long defended the view that the emotion I atheistically feel watching the sun set from Wreck Beach or listening to Beethoven’s Ninth is the same emotion worshippers feel in their temples. I have also long claimed that my sense of longing for the West Coast while living on the prairies where I grew up has a quality that I can only call homesickness. But it may be in this latter case that my concept of “home” is out of step with the norm.

[13] Would a case of a dog suppressing her whimpering count as a case of canine control over emotions? Perhaps on an expression theory of emotions that differentiated an actual case of whimpering from a desire to whimper which is not expressed.

[14] Whether authenticity is to be considered a moral value, a virtue, or some other sort of norm is up for debate. It seems to be used as a normative concept, but how normative concepts fit into the value-free ontology of existentialism is a sticky problem.

[15] If the car from which this person has been pushed bears markings of certain notorious gangs in my community, fear and a desire to mind my own business may also be appropriate, and possibly paradigmatic, given the likelihood that this is how I have always seen others respond under such circumstances. The possibility of such socially-sanctioned indifference is why I deny that paradigm scenarios define the measure of an emotion’s moral assessment, and prefer to adopt de Sousa’s four-part measure of motivation, universalisability, relationship maintenance, and experiential value.

[16] Amélie Rorty’s attention to the habitual nature of emotion (described in Chapter Two) provides an insight into the problem of recalcitrance not achieved by the theorists dealt with so far.

[17] Aristotle holds a similar view on this and many other points throughout this essay. It is considerations of length and breadth that motivate my omission of pre-Darwinian thinkers. I do however intend to study Aristotle’s theories of character and emotion in future work.

[18] Or class stereotyping, as in concerns about “good birth” and royal blood. It has been pointed out by critics of Dickens that Oliver Twist suffers from a sort of genetic classism implied by the fact that the only thing which differentiated the incorruptible Oliver from his rascal associates was that (as it turned out) his mother was from a higher station. He’d been a gentleman all along. He even inherited perfect grammar.

[19] This is my response to all determinist theories that rely on neuroscientific “evidence” against free will: what goes on in the nervous system will look the same whether the activity is willed freely or determined. There is a genetic fallacy involved in the move from describing the workings of the nervous system to showing that the system somehow is not a free being. I suspect that it is the Cartesian search for a non-corporeal source of the will that leads some to assume that the absence of such a source proves determinism to be true. In fact, if the mind really is the brain, then study of the brain is study of the activities of a free agent.

[20] Walter B. Cannon argues this, pointing out that fear and rage exhibit the same physiological symptoms, but do not feel alike (2003, 80).

[21] Since one is seldom a member of just one socially-defined group, these expectations can often be inconsistent and a source of personal confusion and marginalisation. Membership in one group may bar someone from the community they should enjoy as members of another.

[22] I suspect that it is the association between temperament and physiology which leads some to suppose that the former is fixed when the latter ceases the rapid changes it undergoes throughout childhood. The “mellowing” one is expected to undergo in old age could be explained by the physical changes of aging. However, it is equally plausible to explain the mellowing of age in terms of one’s conformity to the behaviours associated with the social category of the elderly. Or, it could be explained by the changes in beliefs and attitude that come with years of life experience.

[23] How habit develops on his radically voluntaristic view is not clear.

[24] I’ll allow a trait to be as general (“short-tempered”) or specific (“loses patience with video store clerks”) as one likes. An advantage of allowing such specific traits is that it enables us to capture cases where one is consistent and regular in a very particular way. It wouldn’t make sense to say I was short-tempered, when there is only a very particular set of circumstances that sets off my temper. But when I respond to these circumstances consistently, we should be able to say I have a corresponding trait.

[25]Many pot smokers report that it is impossible to stay mad once one has gotten high.

[26] An example of such a program is Emotions Anonymous (EA), a Twelve-Step program, modelled after Alcoholics Anonymous. In it, people who have had difficulty controlling their emotions discuss with each other the misfortunes that have followed and help support each other in their efforts to change. While it aims at emotional stability, it sends a mixed message about emotional control. The language of the program suggests that the members are actually powerless over their emotions and are able to maintain control through the aid of the community, program, and a higher power (Emotions Anonymous 1994).

[27] Sometimes it is not the person who needs to change, but larger elements of society. This will be considered in Part 6 of Chapter Three.

[28] I would not want to rule out that it also makes sense to say “you have a future responsibility to x” without that person accepting the responsibility, but here I may be confusing responsibility with duty.

[29] This is similar to Sartre and Solomon’s position that one’s responsibility is established by becoming aware of what one is now.

[30] Step One of the Emotions Anonymous recovery program is to admit that one is powerless over one’s emotions and that one’s life has become unmanageable (EA 41).

[31] The ambiguities of these terms can be misleading here. In one sense one is “accountable” whether or not they have agreed to be. But the context here is that of taking responsibility.

[32] Or we might laud their progress to a higher level of moral awareness, one which undermines the paradigm scenario of not feeling for lab animals. This possibility is explored in Part 6 below.

[33] Steps four and ten of EA are concerned with creating and maintaining a moral inventory (Emotions Anonymous 1994, 53, 71).

[34] Gregory Trianosky also distinguishes between having virtues and being creditable for them. If a person has a character trait because of his or her active discrimination, and not just because she or he is easily moulded by society and upbringing, this person may be considered responsible and creditable (94-96). What Trianosky denies responsibility for is one’s capacity to be influenced or decisive. These propensities he sees as part of one’s temperament, which he feels is outside of one’s control (98-99). My reasons for rejecting the claim that temperament is outside of one’s control are explained in Chapter Two, but the claim widely enough believed have had a significant impact on the discussion of responsibility.

[35] The relationship between emotions and character traits is explained in Chapter Two.

[36] Although all loves admit the possibility of jealousy. The affections of one’s children, friends, parents, students, and audience may all be jealously guarded.

[37] I am being wary not to identify “love” as one emotion or even by any specific division (e.g. eros, agape, filia) as any experience of “love” tends to consist in a set of other emotions with some added unifying feature, which is not just “attraction.” I know not what this extra feature actually is and I sometimes wonder whether it is just a unifying myth for a particular emotional history in relation to some other person. Or perhaps it is a social category based on family resemblances of various sorts of orientation toward others.

[38] It is certainly understandable that feelings change and so with them relationships, or relationships’ authenticity. Furthermore, it seems that the imperative to try to experience intrinsically valuable emotions would motivate the choice to pursue each new infatuation when an old relationship has become a chore. However, insofar as we are aware of the conditions under which the tender emotions develop and wane, a commitment to stay in an authentic loving relationship with someone includes a commitment to do what is in our power to maintain the sentiment appropriate to the relationship and to avoid putting ourselves in situations which try our limits of control. Dates are for more than courtship and phone calls are for more than information. Flirtation is only to a certain degree innocent. Although it is to some degree possible to willfully sustain an affective attachment to another, we must be cautioned against demanding of anyone too much bootstrapping for the sake of maintaining a commitment. There must be a limit to what is reasonable to ask to stick to emotionally. Obviously something like love has deal-breakers, violence for example. Also, given that something like love has as part of its meaningfulness that it is to some degree unchosen, it cannot be the case that its maintenance become the whole goal of all time spent together. Furthermore, it may not always be the case that an obvious deal-breaking act has occurred. There may be very legitimate reasons for turning one’s back on a love of which one may not be explicitly aware. Sometimes the emotions are the very indicator of these hidden problems.

[39] The articulation of this dimension of emotional responsibility will have to await a future work. It is something that has already been anticipated by these theorists, and explicitly in by Margaret Walker in “Ineluctable Feelings and Moral Recognition” (1998).