Academia.eduAcademia.edu
Hume’s Science of Emotions: Feeling Theory without Tears Mark Collier Hume Studies Volume 37, Number 1 (2011), 3-18. Your use of the HUME STUDIES archive indicates your acceptance of HUME STUDIES’ Terms and Conditions of Use, available at http://www.humesociety.org/hs/about/terms.html. HUME STUDIES’ Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the HUME STUDIES archive only for your personal, non-commercial use. Each copy of any part of a HUME STUDIES transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. For more information on HUME STUDIES contact humestudies-info@humesociety.org http://www.humesociety.org/hs/ Hume Studies Volume 37, Number 1, 2011, pp. 3–18 Hume’s Science of Emotions: Feeling Theory without Tears MARK COLLIER Abstract: We must rethink the status of Hume’s science of emotions. Contemporary philosophers typically dismiss Hume’s account on the grounds that he mistakenly identiies emotions with feelings. But the traditional objections to Hume’s feeling theory are not as strong as commonly thought. Hume makes several important contributions, moreover, to our understanding of the operations of the emotions. His claims about the causal antecedents of the indirect passions receive support from studies in appraisal theory, for example, and his suggestions concerning the social dimensions of self-conscious emotions can help guide future research in this ield. His dual-component hypothesis concerning the processing of emotions, furthermore, suggests a compromise solution to a recalcitrant debate in cognitive science. Finally, Hume’s proposals concerning the motivational inluences of pride, and the conventional nature of emotional display rules, are vindicated by recent work in social psychology. 1. Introduction In Book 2, parts 1 and 2 of the Treatise, Hume attempts to understand agent-directed emotions such as pride and humility. What is their essential nature? Which situations elicit them? How do they inluence our behavior? Hume is conident that his science of human nature can make progress on these topics. Emotions are often experienced as tumultuous, but there is a discernible order beneath the surface. In fact, Hume claims to have discovered the “true system” of the indirect passions Mark Collier is Associate Professor of Philosophy, University of Minnesota, Morris, Morris MN 56267 USA. Email: mcollier@morris.umn.edu. 4 Mark Collier (T 2.1.5.5; SBN 286).1 And his conidence apparently did not wane over time: His psychological explanation of these emotions, he writes in the Dissertation on the Passions, must be acknowledged as “incontestable” (DP 2.13).2 Contemporary philosophers of emotion, however, offer a radically different assessment. Hume’s science of emotions, they maintain, is a complete muddle. Because Hume claims that emotions are species of impression,3 they take his analysis to be a version of feeling theory. But so-called feeling theories have, according to Robert Solomon, been “thoroughly discredited.”4 William Alston, reading Hume as identifying emotions with feelings, charges that Hume fails to recognize that thoughts are constitutive parts of emotions.5 This initial mistake, moreover, breeds confusion. It leads Hume to describe conceptual platitudes about emotions, for example, as if they were discovered through empirical inquiry.6 Things go wrong at the outset of his investigations, it seems, and thus we can safely ignore everything that follows. It is rare to ind consensus in the philosophy of emotions. This is the sole point on which there appears to be complete agreement. The consensus view, however, is open to criticism. This paper argues that while Hume does embrace, as his critics maintain, a feeling theory of emotions, the objections raised against this position are not as strong as commonly thought (part 2). It further argues that the standard dismissal of Hume’s account is unfortunate, since it directs attention away from his contributions to the science of emotions. Hume’s claims about the cognitive antecedents of emotions, for example, receive support from contemporary appraisal theory (part 3). His account of the psychological mechanisms involved in emotional processing, furthermore, offers a compromise solution to a recalcitrant debate in psychology (part 4). Finally, his proposals concerning the motivational role of pride, as well as the conventional basis of emotional display rules, have been corroborated by recent studies (part 5). 2. The Nature of Emotions Hume maintains that emotions such as pride and humility are “simple and uniform impressions” that cannot be reduced to any constituent parts (T 2.1.2.1; SBN 277). We should not be troubled, however, that these terms cannot be deined any further. Their meaning is evident, according to Hume, from “common feeling and experience” (T 2.2.1.1; SBN 329). Our complex idea of causation requires explication because its contents are obscure. But this is not the case with passions such as love and hatred. Anyone who is unfamiliar with their distinctive qualities, as one commentator puts it, has “simply never emoted.”7 Contemporary philosophers describe this analysis of the nature of emotions as “feeling theory.” A feeling theorist is one who identiies types of emotions with their unique phenomenal properties. There is something it is like to be in love, for example, and this sui generis qualitative experience differentiates love from other Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 5 emotions. Feeling theorists do not deny that emotions are caused by judgments or that they produce distinctive patterns of behavior. They merely deny that these causes and effects belong to the intrinsic nature of emotions. Emotions are essentially only feelings. Why do contemporary philosophers of emotion reject this position? We can distinguish four standard objections. Each of them, taken separately, is regarded as decisive. Together, they are thought to represent a complete demolishing of Hume’s account. Objection 1: We cannot identify emotions with feelings because distinct emotions often share one and the same feeling. We can easily distinguish between envy and jealousy, according to C. D. Broad, even though there is only a “shade of difference” between their phenomenal properties.8 Errol Bedford points out that the same is true of indignation and annoyance: these emotions “differ little, if at all,” in terms of their felt qualities.9 Objection 2: If emotions are essentially only feelings, their causes would be contingently related to them, and it would be logically possible for us to feel proud of the industry of Stone Age ants in Papua.10 But this is clearly absurd. Looking favorably upon either our own achievements or those of people close to us is part of what it means to say that we are proud.11 Objection 3: Emotions are intentional states; feelings are not. We cannot simply be angry; we must be angry at someone or about something. But feelings are sensations, like tickles and pangs, and thus lack direction.12 Emotions have a property, therefore, that feelings lack. It follows that they must be different. Objection 4: Emotions are subject to rational assessment. A man’s pride in his company’s record, for example, might be said to be unjustiied.13 Sometimes these evaluations involve types of emotions: joy is an unit response, for example, to the suffering of innocent persons. Other times these judgments are a matter of degree: intense fear is appropriate when we encounter a mad bull but not an angry cow.14 Feeling theorists cannot make room, however, for these normative considerations. If emotions are brute occurrences, like headaches, they cannot be said to be unreasonable.15 One might question whether these traditional objections manage to hit their target. After all, Hume appears to deny that emotions can be identiied with feelings when he refers to pride as “that agreeable impression, which arises in the mind, when the view either of our virtue, beauty, riches, or power makes us satisfy’d with ourselves” (T 2.1.7.8; SBN 297). It seems that Hume endorses, in this passage, a “hybrid” theory of emotions16 rather than a feeling theory. Emotions are individuated, on this proposal, according to their constitutive thoughts and feelings.17 Just as an inlammation of the skin is not sunburn unless it is produced by sunrays, a warm glow would not count as pride unless it is caused by thoughts of our superlative qualities. Volume 37, Number 1, 2011 6 Mark Collier A hybrid theory of emotions would enable Hume to make short work of the traditional objections. He could agree with his critics that envy and jealousy cannot be individuated according to their phenomenal properties. They can be distinguished, however, in terms of their causal etiologies. Envy is a disagreeable sensation produced by the belief that our rivals enjoy what we desire, for example, but we are only jealous when this enjoyment involves the attention of some third party. Hume could also acknowledge that the logical relation between emotion and thought is non-contingent, since a hybrid theorist holds that emotions are composed, at least in part, by cognitive states. Finally, he would be in a position to explain the fact that emotions are intentional and evaluable: they would be directed toward the world, and judged appropriate or not, according to the thoughts essential to them. Let us call this the conciliatory strategy. It acknowledges the force of the standard objections to Hume’s account but dodges these criticisms by locating a second deinition of emotion in his writings. The hybrid theory of emotions has independent theoretical plausibility,18 but ascribing it to Hume puts enormous strain on some of his texts. One must, at the very least, concede that Hume, as Davidson puts it, “can be quoted on both sides.”19 And the crucial passage cited in support of this reading is hardly conclusive. Hume proceeds in the very next sentence, after all, to refer to emotions as impressions (T 2.1.7.8; SBN 297). There are a number of places in the text, moreover, in which Hume unequivocally identiies emotions with feelings. He classiies emotions in general, for example, as “relective impressions” (T 2.1.1.1, T 1.1.2.1; SBN 275, 7–8). He often employs these terms, moreover, as if they were interchangeable (T 2.1.5.4, T 2.1.9.5, T 2.2.9.20; SBN 286, 305, 380).20 Hume makes it clear that “pride is a pleasant sensation” and that this feeling constitutes its “very being and essence” (T 2.1.5.4, T 2.1.2.1; SBN 286, 277). This does not entail, of course, that pride is only a sensation. (The mind is essentially a thinking thing, as Arnauld points out, but it might also be essentially embodied).21 But it is dificult to understand, given Hume’s theoretical commitments, what these further conditions might be. Hume’s oficial position is that causes are always separable from their effects (T 1.3.3.3; SBN 79–80). The thoughts that occasion emotions, therefore, cannot be essential parts of their nature. It is unnecessary to adopt the conciliatory strategy, moreover, and saddle Hume with two deinitions of emotion. One can simply embrace his version of feeling theory and show that the standard objections fail to refute it by replying to them as follows. Reply to Objection 1: Bedford and Broad understate the qualitative differences between individual emotions. Indignation does not feel the same as annoyance; the same is true of jealousy and envy. We often manage to identify our feelings, moreover, without awareness of the beliefs that produce them. We might notice Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 7 that we are angry or afraid, for example, without knowing why this is the case. This is why so many people seek therapy. Reply to Objection 2: If it was an analytic truth that we can only take pride in what is closely related to us, then it would be logically impossible to have this feeling in response to the achievements of Stone Age ants. But the proper deinition of emotion is precisely what is at issue. Common sense, moreover, supports Hume’s analysis.22 We do not ordinarily talk of emotions as being composed of thoughts. Our philosophical intuitions, once we have sorted them out, might suggest otherwise, but the burden of proof falls upon Hume’s critics to demonstrate that this is so. Reply to Objection 3: Emotions are not, strictly speaking, representational states. An emotion is an “original existence,” as Hume puts it, and thus cannot be true or false (T 2.3.3.5; SBN 415), but this does not prevent emotions from borrowing their intentionality from the beliefs that cause them. Hume would be in serious trouble if emotions could be shown to have original intentionality, but his opponents have not established that this is the case, and ordinary language is neutral on this issue. Reply to Objection 4: When we say that a man’s pride in his company’s records is unreasonable, according to Hume, we are asserting that his passion is based on a false or unjustiied belief (T 2.3.3.6; SBN 416). Let us suppose, for example, that he has cooked the books. His pride would be inappropriate, in the sense that it is caused by a belief in accomplishments which do not really exist. It is not the emotion, “properly speaking,” which is unreasonable (ibid.). The ittingness of emotions, like their intentionality, is derived from the cognitive states that elicit them. In summary, it is commonly thought that Hume’s feeling theory has been decisively refuted. Indeed, contemporary philosophers often take this as the starting point of their discussions, so that one rarely feels the need to argue for it anymore. But this orthodox view should be challenged. Hume has adequate resources, as we have seen, to respond to his critics. His feeling theory can account for the same desiderata as its rivals. It has the theoretical virtue, moreover, of putting phenomenal properties where they should be—at the heart of our analysis of emotions. 3. Causal Antecedents Hume describes his analysis of the essential nature of emotions as “preliminary” to the main project in Book 2 of the Treatise (T 2.1.2.1; SBN 277). This might strike contemporary philosophers as rather odd. Once emotions are identiied with feelings, after all, what explanatory work remains to be done? It would appear that Hume’s feeling theory leaves us with only one possible answer: introspectors must be trained to make careful observations about the contents of emotional consciousness.23 But even this task, according to Hume, would be superluous. We are Volume 37, Number 1, 2011 8 Mark Collier suficiently acquainted in everyday life with the qualitative aspects of emotions; pride and humility are the “most common” of any impressions, as he puts it, and “every one, of himself, will be able to form a just idea of them” (T 2.1.2.1; SBN 277). One hardly needs to be an expert in phenomenology, therefore, to know what it is like to experience these emotions. What remains to be understood is why we have the emotions that we do. What makes us proud of some qualities, for example, but mortiied by others? Why do we love some people but hate others? These are questions that we cannot hope to answer from the vantage point of common life. The general causes of our emotions, unlike their phenomenal properties, are not directly accessible to consciousness. In order to make progress on these topics, therefore, we must adopt a scientiic approach. We must carefully examine the comings and goings of emotions—as we would with any other natural phenomenon—by collecting data and performing experiments. The starting point of these investigations involves a description of the elicitors of emotions. Hume lists the “vast variety” of qualities, for example, which make us feel proud or ashamed: Every valuable quality of the mind, whether of the imagination, judgment, memory or disposition; wit, good-sense, learning, courage, justice, integrity; all these are causes of pride; and their opposites of humility. Nor are these passions conin’d to the mind, but extend their view to the body likewise. A man may be proud of his beauty, strength, agility, good mein, address in dancing, riding, fencing, and of his dexterity in any manual business or manufacture. But this is not all. The passions looking farther, comprehend whatever objects are in the least ally’d or related to us. Our country, family, children, relations, riches, houses, gardens, horses, dogs, cloaths; any of these may become a cause either of pride or humility. (T 2.1.2.5; SBN 279) The causal triggers of love are equally diverse: they include qualities of mind such as virtue and knowledge, “bodily accomplishments” like strength and beauty, as well as “external advantages” of family and possessions (T 2.2.1.4; SBN 330). There is a wildly disjunctive set of factors, then, which can prompt our emotional reactions. When we examine these lists, however, general patterns begin to emerge. The various things which produce feelings of pride, for example, are always agreeable and related to us (T 2.1.5.8; SBN 288). The elicitors of shame are, without exception, unpleasant and connected to ourselves (T 2.1.5.9; SBN 288–89). We experience love when we consider pleasing qualities that are related to other agents, and we feel hatred whenever another person’s characteristics make us uneasy (T 2.2.1.4; SBN 331). Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 9 Hume has isolated two main variables, then, in the causal antecedents of emotions. (A) Valence: Whether a quality is perceived to be pleasant or unpleasant. (B) Agency: Whether a quality is perceived to be related to ourselves or others. The inluence of these two dimensions becomes apparent when we consider Hume’s “square” of the indirect passions (T 2.2.2.3; SBN 333). Pleasant Unpleasant Self Pride Humility Other Love Hatred Figure 1. Hume’s Square of the Indirect Passions Our emotional responses lawfully depend, it seems, on how we perceive a situation. If we understood the agency and valence attributed to an object or event, then, we could in principle reliably predict which of these feeling would ensue. Hume has discovered a prima facie regularity in the operations of our emotions. But how could he establish that it is genuine rather than accidental? Hume acknowledges that he has not yet made this case. He devises a series of “new experiments” to prove, therefore, that our emotional reactions depend on these factors (T 2.2.2.1; SBN 332). Things which lack either valence (~V) or agency (~A), such as an ordinary stone, fail to elicit any of the indirect passions (T 2.2.2.5; SBN 334). Subjects without any valence (~V) but related to ourselves (A) also fail to do so (T 2.2.2.6; SBN 334). The same is true of qualities, moreover, which are perceived as pleasant or painful (V) but not in any way connected to us (~A) (T 2.2.2.7; SBN 334–35). Consideration of these crucial instances, then, should increase our conidence in his hypothesis; they are “undeniable proofs,” as Hume puts it, of its empirical adequacy (T 2.2.3.1; SBN 347). These experiments do not, of course, meet our contemporary standards of rigor. Hume does not make any quantitative measurements or identify his experimental subjects. Indeed, he appears to have performed these trials on himself. Although Hume did not rigorously test his hypothesis, we are currently in a position to do so. Social psychologists have devised an experimental paradigm that enables us to isolate the causal antecedents of emotions. Individuals in these experiments rate the situations that elicit particular emotions on a point scale along a variety of evaluative dimensions.24 This numerical data enables researchers to use sophisticated analytical techniques, such as factor analysis and structural models, to measure the statistical co-variation of appraisals and emotions. The results of these quantitative studies are consistent with Hume’s hypothesis concerning valence and agency. Participants report feelings of pride when they view Volume 37, Number 1, 2011 10 Mark Collier themselves as responsible for pleasant situations, for example, and they describe feelings of shame when they see themselves as the source of disagreeable ones.25 Subjects experience love when others are believed to be responsible for agreeable states of affairs, moreover, and they feel hatred when these agents are regarded as the source of undesirable actions or events.26 Recent work in appraisal theory, then, provides support for Hume’s account. But this should be viewed as a two-way street: Hume’s proposals can help guide future research in this area. Contemporary researchers often interpret the dimension of agency, for example, in a narrow manner that requires attributions of responsibility.27 But Hume teaches us that this condition is overly restrictive. We can feel proud of the noble deeds of our ancestors, for example, or the beautiful climate of our native lands (T 2.1.9.9–11; SBN 307–08). The same is true of feelings of humility. Lord Byron was apparently ashamed of his foot.28 We ought to prefer Hume’s wider speciication of agency, then, in terms of what is associatively connected to ourselves or others. This enables us to understand why our emotional responses often depend on features of ourselves, or of the groups with which we identify, that lie outside the scope of personal responsibility. Hume’s work on the passions also shows that social psychologists must expand their set of appraisal dimensions in order to properly understand the evaluative basis of self-conscious emotions. Consider the subtle reinements that Hume makes to his account in T 2.1.6. His oficial position is that valence and agency are not jointly suficient to produce pride and humility. Further conditions must be satisied. The irst is that the qualities of which we are proud or ashamed must be seen as comparatively unique. This explains why we do not normally take pride in our good health, for example, even though it is pleasant and connected to ourselves (T 2.1.6.2; SBN 292). These features must also be seconded by others: we cannot be ashamed about something which others do not recognize as a deiciency (T 2.1.6.6, T 2.1.8.9, T 2.1.11.1; SBN 292, 303, 316). We cannot understand emotions like pride and humility, then, solely in terms of the impact of events on our personal well-being; we must also take into consideration the circumstances and attitudes of others. These social dimensions are often overlooked, however, by contemporary appraisal theorists.29 4. Psychological Mechanisms Hume is not only interested in the antecedents of emotions. He also wants to understand the psychological mechanisms that transform these appraisals into feelings. And he is conident that experimental philosophy can make progress on this topic. “[I]n the production and conduct of the passions, there is a certain regular mechanism, which is susceptible of as accurate a disquisition, as the laws of motion, optics, hydrostatics, or any part of natural philosophy” (DP 6.19). Hume’s Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 11 investigations into the mechanisms by which the indirect passions are produced continue the naturalistic project of the Treatise. His goal is to discover the principles of human nature that are responsible for our emotional responses to the world. Hume’s positive account is radical and unlinching. Emotional processing can be fully explained, on this proposal, in terms of associative relations (T 2.1.4.2, T 2.1.5.11; SBN 283, 289–90).30 Hume maintains that philosophers have traditionally over-intellectualized the passions. He attempts to rectify this situation by showing that emotions are derived from psychological processes which are so simple and involve “so little relection and judgment” that they are shared with non-human animals (T 2.1.12.9, T 2.2.12.1; SBN 328, 397). Hume’s processing model involves two separate pathways that interact with one another to produce the indirect passions. Suppose that you own a beautiful house, for example, and you are currently admiring its distinctive characteristics. Hume maintains that this apprehension would give rise to two distinct psychological responses. The irst is affective: the qualities of the house, such as its form and function, would cause you to experience a pleasant sensation (T 2.1.8.2; SBN 299). The second is cognitive: the fact that you are its proprietor would cause you to think of yourself. Through a process of mutual reinforcement—or what Hume calls a “double impulse”—these responses would make you feel proud (T 2.1.1.4, T 2.1.5.5; SBN 284, 286–87). Hume is not entirely clear, unfortunately, about the details of this proposal. The qualities of the house produce a sensation of pleasure, but this agreeable impression would presumably resemble a number of positively valenced emotions. Why do they give rise to feelings of pride, then, rather than gratitude or joy? Hume suggests that this work is performed by the cognitive pathway in his model. The association of ideas serves to “assist and forward,” as he puts it, the association of impressions (T 2.1.4.4, T 2.1.9.5; SBN 284, 305). This remark does not give interpreters much to go on, but it appears that Hume has the following picture in mind. Our affective response to the house primes a variety of positive emotions. Pride is the member of this set, however, that is most closely related to the idea of ourselves. It is the fact that pride stands in a double association, then, which makes it “favoured beyond its fellows.”31 It must be admitted that Hume does not provide us with a clear blueprint of the interaction between affect and cognition. It might be charitably said on his behalf, however, that no one else has gotten much further on this issue. Contemporary researchers have only recently begun to focus on the psychological processes that underlie our emotional responses. There are several prominent models which assign a prominent role to both affective and cognitive elements.32 It is still not well understood, however, how these two components work together.33 The signiicance of Hume’s account of the mechanisms generating the indirect passions, then, does not lie with its details and speciications. Rather, his Volume 37, Number 1, 2011 12 Mark Collier crucial contribution was to put his inger on the right question: how do affective and cognitive processes work together to produce emotions? The importance of this insight cannot be overstated, moreover, since it is often overlooked in contemporary discussions. Consider the famous Zajonc-Lazarus controversy. Robert Zajonc argues that our affective preferences can be shaped without any cognitive processing.34 Richard Lazarus maintains that complex emotions are always based on cognitive appraisals.35 From a Humean perspective, the central problem with this debate is that one need not take sides. We can agree with Zajonc that our affective responses are immediate; we automatically ind some stimuli agreeable and others unpleasant. But we can also agree with Lazarus that emotions such as pride and humility are cognitively mediated. The question of whether affect or cognition is “primary” is not a helpful one.36 The main issue concerns the interaction between these two components. 5. Behavioral Effects Hume maintains that emotions are qualitative states. He does not deny, however, that they inluence behavior. Feelings of love produce a prima facie desire, for example, to promote the well-being of others (T 2.2.6.3; SBN 367). The crucial point is that this causal connection is logically contingent: “This order of things, abstractly consider’d, is not necessary. Love and hatred might have been unattended with any such desires, or their particular connexion might have been entirely revers’d. If nature had pleas’d, love might have had the same effect as hatred, and hatred with love. I see no contradiction in supposing a desire of producing misery annex’d to love, and of happiness to hatred” (T 2.2.6.6; SBN 368). It is a brute fact about human beings, in other words, that emotions generate the action tendencies they do. Hatred leads us, given our “original constitution,” to desire the misery of others (ibid.). But our psychological proile might have been different than it is. One might raise the objection, along with the logical behaviorists, that this is confused. Being disposed to help those in distress is not an effect of compassion; rather, it is part of the meaning of this concept. It is an analytic truth that one cannot have compassion unless one is practically disposed, all things being equal, to help those in distress. Hume must concede this point when it comes to character traits; we would not call an agent compassionate unless they were inclined to assist those in need. But he could reasonably deny, on the basis of his conceivability argument, that this behavioral disposition is a constitutive part of the emotion. We are not aware of any contradiction in imagining creatures who feel compassion but lack any prima facie desire to help. Such beings, as far as we can tell, are possible. Hume’s proposals about the effects of emotions, then, cannot be rejected on a priori grounds. They receive a good deal of support, moreover, from recent work in the science of emotions. Consider his account of the motivational inluence of Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 13 pride. Pride is a useful emotion; it gives us “conidence and assurance,” as he puts it, “in all our projects and enterprises” (T 3.3.2.8, T 3.3.2.14; SBN 597, 600). Hume’s oficial position is that pride is a “pure sensation” which does not “immediately” excite us to action (T 2.2.6.3; SBN 367), but this feeling can have an indirect effect on behavior by sustaining whatever activities cause it.37 For example, a person might be proud of his ine wardrobe, and this pleasurable sensation would reinforce the desire to appropriate such goods. Pride might not create new desires, then, but it can provide “additional force” to those that already exist (T 2.2.3.4; SBN 439). Hume’s proposal has been corroborated by recent experiments in social psychology.38 Participants in these studies were asked to solve dificult cognitive tasks. They were subsequently divided into groups: one was informed that they had performed exceptionally well (“Great job on that! That’s one of the highest scores we’ve seen so far!”) and received non-verbal cues (smiles and voice intonation) which indicated that they had impressed the experimenters; control groups did not get this type of positive feedback.39 The results of these manipulations were unequivocal. Those who received social acclaim reported heightened levels of pride and exhibited greater tenacity in solving future problems than those who did not. Indeed, researchers observed a direct relationship between the intensity of their pride and the extent of their perseverance.40 Hume recognizes that feelings of pride, however, can also have negative consequences. Pride is caused by thoughts of our exceptional qualities, which makes it a competitive and non-egalitarian emotion. Pride expressions are disagreeable to others, then, because they invite unlattering comparisons: “ ’Tis a trite observation in philosophy, and even in common life and conversation, that ’tis our own pride, which makes us so much displeas’d with the pride of other people, and that vanity becomes insupportable to us merely because we are vain” (T 3.3.2.7; SBN 596). Pride appears to be a zero sum game. Your addition requires my subtraction. Thus, if everyone were encouraged to display their sense of superiority, social life would quickly become intolerable. We manage to avoid this disagreeable situation, according to Hume, by learning to modulate our pride expressions. We might not be able to control our feelings, but we can “regulate our actions” (T 3.3.2.11; SBN 599). Sincerity, when it comes to pride, is not mandatory. In fact, “[s]ome disguise in this particular is absolutely requisite” (T 3.3.2.10; SBN 598). The obligation to appear modest is what Hume calls an artiicial duty. We are not naturally inclined to suppress our pride; we would prefer to boast about our distinguished qualities. But we adopt these social conventions because they prevent mutual indignation (T 3.3.2.10; SBN 597). Recent work in developmental psychology provides a good deal of support for this claim. Studies have shown that interpersonal norms for regulating emotional displays are acquired through peer reinforcement, imitation, and communicated expectations.41 Children who compete in games exhibit a greater capacity, over Volume 37, Number 1, 2011 14 Mark Collier time, to restrain or mask the pleasures of victory; older children show a greater tendency to minimize their smiles, for example, and keep their arms tucked at their sides.42 Children begin to inhibit their expressions of pride, moreover, as soon as they become aware of the relevant social conventions.43 We are taught at a young age, it seems, that social life requires dissimulation.44 Cross-cultural studies offer further evidence for this proposal. Paul Ekman and his colleagues have discovered that every culture has “display rules,” inculcated early in life, that regulate how emotions should be modulated in social situations.45 Ekman was concerned with basic emotions such as fear and surprise. But recent studies have extended this line of research to pride.46 Prototypical pride behavior involves bodily components (arms raised or akimbo, head tilted back) as well as facial ones (smile). Figure 2. Prototypical Pride Displays47 These pride displays are offensive to others because they suggest high status and rank.48 But this combination of bodily and facial expressions might have adaptive value by helping us to camoulage our feelings.49 We have greater voluntary Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 15 control over the muscles in our limbs than our faces. We might not be able to prevent ourselves from smiling when we think about our achievements, then, but we can surely prevent ourselves from expanding our chest or raising our hands over our heads. When these bodily postures are minimized, we are left with a relatively friendly gesture. Hume maintains that human beings are not the only animals who feel pride. The elaborate songs of the nightingales, he proposes, are “evident marks” of their vanity; the swagger and strut of a peacock, moreover, reveal the “high idea he has entertain’d of himself” (T 2.1.12.4; SBN 326). However, we might be unique in our capacity to suppress our emotional displays in the context of social interaction. Rousseau famously complained about the artiices of civilization. But Hume regards these conventions as indispensable. Human beings manage to get along with one another, on his account, by learning to mask their authentic feelings. 6. Conclusion Hume does not claim to have “exhausted” the subject of the emotions; it would be suficient for his purposes to have demonstrated, he maintains, that they can be approached from an experimental point of view (DP 6.19). And in this respect, he surely succeeded. Hume makes a number of important contributions to our understanding of the causal antecedents, psychological mechanisms, and behavioral effects of the indirect passions. It is time for us to rethink, then, the standard evaluation of his account. We should embrace Hume’s feeling theory—without tears. NOTES I would like to express gratitude to my colleagues at UMM, the editors of Hume Studies (Corliss Swain and Saul Traiger), and several anonymous referees for their valuable comments and suggestions. 1 References to the Treatise are to David Hume, A Treatise of Human Nature, ed. David Fate Norton and Mary J. Norton (New York: Oxford University Press, 2000), hereafter cited as “T” followed by Book, part, section, and paragraph numbers; and to A Treatise of Human Nature, ed. L. A. Selby-Bigge, revised by P. H. Nidditch, 2nd ed. (Oxford: Clarendon Press, 1978), hereafter cited as “SBN” followed by page number. 2 David Hume, A Dissertation on the Passions and the Natural History of Religion, ed. Tom L. Beauchamp (Oxford: Oxford University Press, 2007), hereafter cited as “DP” followed by section and paragraph numbers. 3 Anthony Kenny, Action, Emotion, and Will (London: Routledge & Kegan Paul, 1963), 20; see also George Pitcher, “Emotion,” Mind 74 (1965): 326–46, 326, and William Lyons, Emotion (Cambridge: Cambridge University Press, 1980), 10. Volume 37, Number 1, 2011 16 Mark Collier 4 Robert Solomon, “The Logic of Emotion,” Noûs 11 (1977): 41–49, 43; cf. John Deigh, “Cognitivism in the Theory of Emotions,” Ethics 104 (1994): 824–54, 824. 5 William Alston, “Moral Attitudes and Moral Judgments,” Noûs 2 (1968): 1–23, 14. 6 Patrick Gardiner, “Hume’s Theory of the Passions,” in David Hume: A Symposium, ed. David Pears (New York: St. Martin’s Press, 1966), 40. 7 Joel Marks, “A Theory of Emotions,” Philosophical Studies 42 (1982): 227–42, 228. 8 Charlie Dunbar Broad, “Emotion and Sentiment,” Journal of Aesthetics and Art Criticism 13 (1954): 203–14, 213. 9 Errol Bedford, “Emotions,” Proceedings of the Aristotelian Society 57 (1957): 281–304, 282–83. 10 Kenny, Action, Emotion, and Will, 24. 11 Irving Thalberg, “Emotion and Thought,” American Philosophical Quarterly 1 (1964): 45–55, 52. 12 Pitcher, “Emotion,” 327; see also Robert Solomon, “Emotions and Choice,” The Review of Metaphysics 27 (1974): 20–41, 21–22. 13 Bedford, “Emotions,” 299. 14 Broad, “Emotion and Sentiment,” 209. 15 Bedford, “Emotions,” 296; cf. Pitcher, “Emotion,” 329. 16 Jesse Prinz, Gut Reactions: A Perceptual Theory of Emotion (New York: Oxford University Press, 2004), 11. 17 Gabriele Taylor, “Pride,” in Explaining Emotions, ed. Amelie Rorty (Los Angeles: University of California Press, 1980), 393. See also Lilli Alanen, “The Powers and Mechanisms of the Passions,” in The Blackwell Guide to Hume’s Treatise, ed. Saul Traiger (Cornwall, UK: Blackwell, 2006), 190. 18 Irving Goldstein, “Are Emotions Feelings? A Further Look at Hedonic Theories of Emotions,” Consciousness and Emotion 3 (2002): 21–33; Eddy Zemach, “What is Emotion?” American Philosophical Quarterly 38 (2001): 197–206. 19 Donald Davidson, “Hume’s Cognitive Theory of Pride,” Journal of Philosophy 73 (1978): 744–57, 753. 20 It is tempting to respond that these impressions might be “hybrid” in the sense that they contain ideas as constituent parts. See Haruko Inoue, “The Origin of the Indirect Passions in the Treatise: An Analogy between Books 1 and 2,”Hume Studies 29 (2003): 205–21, 213; Alanen, “Powers and Mechanisms of the Passions,” 187; Rachel Cohon, “Hume’s Indirect Passions,” in A Companion to Hume, ed. Elizabeth Radcliffe (Malden, MA: Blackwell, 2008), 161 and 181n8. However, Hume makes it clear that “simple impressions” such as love and hatred are “without mixture or composition” (T 2.2.1.1; SBN 329). These impressions can mix or blend with each other, and thus form new impressions, but in so doing they do not retain their parts (T 2.2.6.1; SBN 366). Hume Studies Hume’s Science of Emotions: Feeling Theory without Tears 17 21 Antoine Arnauld, “Fourth Set of Objections,” in The Philosophical Writings of Descartes: Volume 2, ed. John Cottingham, Robert Stoothoff, and Dugald Murdoch (Cambridge: Cambridge University Press), 142–43. 22 Solomon, “Logic of Emotion,” 42; cf. Pitcher, “Emotion,” 326. 23 Kenny, Action, Emotion, and Will, 29. 24 Klaus Scherer, “Appraisal Theory,” in Handbook of Cognition and Emotion, ed. Tim Dalgleish and Mick Power (West Sussex: John Wiley & Sons, 1999), 640–42. 25 Craig Smith and Phoebe Ellsworth, “Patterns of Cognitive Appraisal in Emotion,” Journal of Personality and Social Psychology 48 (1985): 813–38, 829; Arif Butt and Jin Choi, “The Effects of Cognitive Appraisal and Emotion on Social Motive and Negotiation Behavior: The Crucial Role of Agency of Negotiator Emotion,” Human Performance 19 (2006): 305–25, 319. 26 Ira Roseman, Martin Spindel, and Paul Jose, “Appraisals of Emotion-Eliciting Events: Testing a Theory of Discrete Emotions,” Journal of Personality and Social Psychology 59 (1990): 899–915, 907. 27 Bernard Weiner, An Attribution Theory of Motivation and Emotion (New York: Springer-Verlag, 1986); Richard Lazarus, Emotion and Adaptation (New York: Oxford University Press, 1991). 28 Stephen Leighton, “Feelings and Emotion,” Review of Metaphysics 38 (1984): 303–20, 308. 29 Randolph Cornelius, The Science of Emotion: Research and Tradition in the Psychology of Emotions (New Jersey: Prentice Hall, 1996), 148. 30 The only wrinkle is that associations, in this case, involve impressions as well as ideas (T 2.1.4.3; SBN 283). 31 Norman Kemp Smith, The Philosophy of David Hume (London: St. Martin’s Press, 1960), 185n. 32 Leonard Berkowitz, “On the Formation and Regulation of Anger and Aggression: A Cognitive-Neoassociationistic Analysis,” in Emotions in Social Psychology, ed. W. Gerard Parrott (Philadelphia: Psychology Press, 2001), 328–29; Craig Smith and Leslie Kirby, “Towards Delivering on the Promise of Appraisal Theory,” in Appraisal Processes in Emotion: Theory, Methods, Research, ed. Klaus Scherer, Angela Schorr, and Tom Johnstone (New York: Oxford, 2001), 128–32. 33 Gerald Clore and Anthony Ortony, “Appraisal Theories: How Cognition Shapes Affect into Emotion,” in Handbook of Emotions, ed. Michal Lewis, Jeannette HavilandJones, and Lisa Barrett (New York: Guilford Press, 2008), 636–39. 34 Robert Zajonc, “Feeling and Thinking: Preferences Need No Inferences,” American Psychologist 35 (1980): 151–75, 151. 35 Richard Lazarus, “Thoughts on the Relations between Emotion and Cognition,” American Psychologist 37 (1982): 1019–24, 1019. Volume 37, Number 1, 2011 18 Mark Collier 36 Robert Zajonc, “On the Primacy of Affect,” American Psychologist 39 (1984): 117–23; Richard Lazarus, “On the Primacy of Cognition,” American Psychologist 39 (1984): 124–29. 37 Jane McIntyre, “Hume’s Passions: Direct and Indirect,” Hume Studies 26 (2000): 77–86, 81; Lorraine Besser-Jones, “Hume on Pride-in-Virtue: A Reliable Motive?” Hume Studies 36 (2010): 171–92, 187. 38 Lisa Williams and David DeSteno, “Pride and Perseverance: the Motivational Role of Pride,” Journal of Personality and Social Psychology 94 (2008): 1007–17. 39 Williams and DeSteno, “Pride and Perseverance,” 1010. 40 Ibid., 1012–13. 41 Carolyn Saarni and Maria von Salisch, “The Socialization of Emotional Dissemblance,” in Lying and Deception in Everyday Life, ed. Michael Lewis and Carolyn Saarni (New York: Guilford Press, 1993), 111–14. 42 Nadja Reissland and Paul Harris, “Children’s Use of Display Rules in Pride-Eliciting Situations,” British Journal of Developmental Psychology 9 (1991): 431–35, 433. 43 Reissland and Harris, “Children’s Use of Display Rules,” 434. 44 Pamela Cole, “Children’s Spontaneous Control of Facial Expression,” Child Development 57 (1986): 1309–21, 1320. 45 Paul Ekman, “Universals and Cultural Differences in Facial Expressions of Emotions,” in Nebraska Symposium on Motivation, ed. Juan Cole (Lincoln: University of Nebraska Press, 1971), 225. 46 The crucial piece of evidence comes from Jessica Tracy and David Matsumoto, “The Spontaneous Expression of Pride and Shame: Evidence for Biologically Innate Nonverbal Pride Displays,” Proceedings of the National Academy of Science 105 (2008): 11655–60, 11658. They discovered that sighted athletes—who presumably have greater awareness of social conventions for nonverbal expressions—are more likely to inhibit displays of shame than congenitally blind athletes. 47 Jessica Tracy and Richard Robins, “Show Your Pride: Evidence for a Discrete Emotion Expression,” Psychological Science 15 (2004): 194–97, 196. 48 Azim Shariff and Jessica Tracy, “Knowing Who’s Boss: Implicit Perceptions of Status from the Nonverbal Expression of Pride,” Emotion 9 (2009): 631–39; see also Christopher Oveis, E. J. Horberg, and Dacher Keltner, “Compassion, Pride, and Social Intuitions of Self-Other Similarity,” Journal of Personality and Social Psychology 98 (2010): 618–30. 49 Jessica Tracy and Richard Robins, “The Nature of Pride,” in The Self-Conscious Emotions: Theory and Research, ed. Jessica Tracy, Richard Robins, and June Tangney (New York: Guilford Press, 2007), 270–71. Hume Studies