EMOTION

Over the centuries, the emotions have proven to be a notoriously recalcitrant philosophical subject, defying easy classification and stubbornly straddling accepted philosophical distinctions. With changing conceptions of the mind and its powers, categories such as emotion, desire, appetite, passion, feeling, and sentiment come and go. The general term the emotions is a relatively recent arrival to the English language, first gaining prominence in the nineteenth century, long after terms such as fear, shame, and joy were in common use. Its introduction was an attempt to clump together states that were supposedly marked by a degree of "emotion," a metaphorical extension of the original sense of the word, namely, agitated motion, or turbulence. Only the vagueness of the metaphor allows it to stretch far enough to cover typically quiescent "emotions" such as being pleased or sad about something.

Probably one influence on the extension of the term is the older category of "passions," in the sense of ways of being acted upon. In many languages nearly all emotion adjectives are derived from participles: for example, the English words amused, annoyed, ashamed, astonished, delighted, depressed, embarrassed, excited, frightened, horrified, irritated, pleased, terrified, surprised, upset, and worried—and even sad (from "sated"). When people are, for example, frightened, something acts on them, that is, frightens them: typically, something of which they are aware. However, even if the terms commonly used for the various emotions suggest that the notion of passivity is central to the ordinary concept of emotion, that notion seems irreparably vague, at best reflective of a prescientific picture of a person (or, for that matter, a physical object) as acting and acted on, as doing and "suffering." Indeed, it is not obvious that the states we call emotions have anything interesting or important in common that distinguishes them from all other mental states. Some philosophers and scientists have argued that what we call "the emotions" do not belong to a "natural kind" or class, and even that the concept of emotion should be banished entirely, at least from scientific discourse. These issues will be taken up in a later section.

The Philosophical Tradition

Since William Alston published his seminal article on "Emotion and Feeling" in the first edition of this Encyclopedia, philosophical scholarship in the area has undergone tremendous growth and variation. Among the major catalysts for change in philosophical thinking about the emotions have been new developments in psychology and neuroscience. However, the medium within which this ferment has largely been taking place is linguistic and conceptual analysis. Although analytic philosophers of emotion use relatively sophisticated logical and linguistic tools, their task has not been much different from that of the many classical philosophers who attempted definitions of various emotions: for example, Aristotle in the Rhetoric, Descartes in The Passions of the Soul, Hobbes in the Leviathan, Spinoza in his Ethics, and Hume in A Treatise on Human Nature. Moreover, the most important outcome of the analytic thrust was a view that had been at least implicit in traditional accounts, namely, cognitivism. Although there are several varieties of cognitivism, perhaps the most influential versions hold that the various emotions are distinguished in part by the types of situation that evoke them; or, more exactly, by the types of situation the awareness of which evokes them; more exactly still, by the content of the beliefs and other propositional attitudes that cause them. Note the importance of situational and cognitive features in Spinoza's definitions, for example:

Fear: an inconstant pain arising from the idea of something past or future, whereof we to a certain extent doubt the issue.

Regret: the desire or appetite to possess something, kept alive by the remembrance of the said thing, and at the same time constrained by the remembrance of other things that exclude the existence of it.

The classical philosophers contributed more than definitions, of course. For example, some declared certain emotions to be the primary or basic emotions. However, the philosophers remained armchair theorists, putting forward at best introspective or anecdotal data. The scientific advances of the nineteenth century, particularly in biology, made it possible to move beyond this.

Bodily Responses and Feelings: Darwin and James

In The Expression of Emotions in Man and Animals, published in 1872, Charles Darwin investigated the various, mostly involuntary physiological changes, especially in the facial and skeletal muscles, which constitute the so-called "expressions" of emotions (1998 [1872]). Others broadened the investigation to include the internal visceral phenomena associated with various emotions. Still, these were thought to be investigations into mere manifestations or accompaniments of emotions. As John Dewey pointed out, "The very phrase 'expression of emotion,' … begs the question of the relation of emotion to organic peripheral action, in that it assumes the former as prior and the latter as secondary" (p. 553). It was left to the introspectionist psychologists, most notably Wilhelm Wundt and Edward Titchener, to offer a systematic account of what they regarded as "the emotions themselves," namely the subjective feeling qualities characteristic of the various emotions, an account that relied heavily on what subjects reported.

To William James, these descriptive studies of "what it is like" to feel the various emotions afforded no insight or understanding. Turning instead to the causes of these feelings, he argued, in his classic 1884 paper, "What is an Emotion?" that they were actually the felt awareness of precisely those physiological "manifestations" of emotion that Darwin and the biologists had been studying. Thus, according to James's theory (also known as the James-Lange theory), an emotion is the felt awareness of reverberations of the "bodily sounding-board," that is, of bodily reactions to something perceived or thought: reactions such as trembling, quickening of pulse, crying, running, or striking someone. It is this perception of one's own bodily responses that endows each type of emotion, such as fear, anger, and joy, with its special feeling quality. From this premise James drew the radical conclusion that emotions or emotional states were effects rather than causes of these bodily reactions. Thus common sense has it backwards: The truth is that "we feel sorry because we cry, angry because we strike, afraid because we tremble, and not that we cry, strike, or tremble, because we are sorry, angry, or fearful" (James 1884, p. 190). However, this conclusion drew on the further assumption, which James inherited from the introspectionists and from Darwin himself, that the various emotions or types of emotional state were nothing but particular feeling qualities. That is, if the emotions are just a subclass of feeling qualities, and if these feeling qualities are caused by the bodily reactions that commonsense regards as manifestations or expressions of emotion, then common sense has it backwards.

Whatever the merits of his arguments, the influence of James brought about a major shift in philosophical and scientific thinking about the emotions. Most important, the study of emotional feelings could no longer be regarded as a special introspective science, insulated from our general theory of human beings as biological organisms.

Philosophical Cognitivism

Opposition to the first of the Jamesian premises, which treats the various emotions as just so many feeling qualities, was a major impetus to the cognitive turn in the philosophy of emotion. However, it was above all the intentionality of emotions that put the cognitive revolution on its positive course. Starting with Anthony Kenny (1963), various authors endeavored to show that, unlike the brute physiological feeling states celebrated by James, emotions and their associated feelings had the characteristic of being about things and events. Thus, people sometimes are (and feel) scared of snakes or angry about the fact that their car was stolen. In this respect, emotions were thought to differ from mere bodily feelings, which, if they are about anything at all, are about bodily phenomena, rather than snakes or car thefts.

The intentionality of emotions also distinguishes them from moods, which are general response templates that are not about anything in particular, even though they may have been precipitated by the awareness of particular facts or events. In many languages the same term may be used for both a mood and an emotion: one may be sad (or: depressed, euphoric) about something, or simply in a sad (depressed, euphoric) mood or frame of mind; or both at once. How deep the distinction between emotion and mood goes is debatable, as many so-called emotions tend to spill over from one category to the other: Initially about their original precipitant ("He's angry about the theft of his car."), they develop into a general response template ("Don't go near him, he's in an angry mood!").

Among cognitive theorists, the notion that the content of emotions takes a propositional form assumed special importance. Suppose I believe that John stole my car. I may say that my anger is about the car, about the car theft, or about what John did. However, fully parsed, my anger attribution can be logically reformulated by the phrase: I am angry about the fact that "John stole my car." Because propositions are the primary vehicles of logic and cognition, the propositional nature of emotion and its intentional objects made it easy to think of emotions in both cognitive and logical terms. It was now possible to articulate and debate what were termed the logic and structure of the various emotions and even their inferential ties to one another.

An early and forthright propositional theory is due to Robert Solomon (1976), who, with a strong emphasis on phenomenology, revived the Stoic view that emotions were themselves simply normative judgments of an urgent kind. Ronald de Sousa (1987) argued that emotions are better assimilated to perceptions. Emotions of a given type, such as fright, represent what they are about as having the corresponding property—for example, as being frightening. They also impose "determinate patterns of salience" on our thought processes: guiding our attention, our lines of inquiry, and our inferential strategies. De Sousa's view in some ways anticipates Jesse Prinz's "embodied appraisal" theory, described in the section, "Somatic Wisdom" (2004). Robert Gordon (1987) argued that most emotions are propositional attitudes that are identified by their causal relations to other propositional attitudes, especially beliefs and wishes. Most emotions are "factive," that is, about a fact (or what the subject takes to be a fact) that frustrates or satisfies a wish; others, such as being afraid or hopeful, are uncertainty emotions.

Some critics argue that propositional accounts would exclude animals and infants lacking language. This criticism would seem committed to the controversial thesis that animals without language do not have any propositional attitudes, including desires or beliefs. Nonetheless, it would be a mistake to try to force all emotions into a propositional framework. It is hard to think of a that -clause that describes what love or hate is about. In some cases, what is called the same emotion (or emotion type) has both propositional and nonpropositional forms. Although one may be startled (to discover) that something is the case, one may also be startled by a sound—with no associated proposition or cognition at all. Even in the case of an emotion about a fact, it is not obvious that its content is exhausted by its propositional content.

Finally, even where propositionality is not in dispute, one may not be able to explain it in terms of standard states such as beliefs and desires. For example, to sustain the claim that fearing (hoping) that p depends on being uncertain whether p, one needs to allow for compartmentalization, for example, to distinguish between emotional and intellectual certainty; otherwise one could not account for cases where the fear (hope) that p persists despite a belief that it is not at all possible (epistemically) that p. To make such a distinction with any clarity, however, may be beyond the competency of analytic philosophy. It may require reference to the underlying neural architecture. For example, Joseph LeDoux (1998) discovered that there are distinct pathways by which the amygdala may be activated, a cortical "high road" that is cognitive, and a "low road" that bypasses the cortex and is strictly perceptual. This hypothesis nicely complements the claim that some emotional states and processes might be modular, that is, "hardwired" in a manner that makes them impenetrable by changes in beliefs and desires. The "quick and dirty" low road often alerts us to emergencies that our cortex "knows" do not exist. These examples suggest an analogy with perceptual illusions, which a correct belief sometimes fails to dispel.

Valence

Emotions are often classified by their valence. Theorists and laypeople tend to readily agree that emotions, or most of them, are either positive or negative. The agreement evaporates, however, as soon as they are asked, "In what respect?" One point of disagreement concerns what is being evaluated: Is it what the emotion is characteristically about that makes it positive or negative (intentional valence), or is it one's having or experiencing the emotion (experiential valence)? What it is about may be good, or something the subject appraises favorably or would wish to be the case, as in pride, delight, and hope; or it may be bad, or something the subject appraises unfavorably or would wish not to be the case, as in shame, regret, and fear. Having or experiencing the emotion might also be judged positive or negative in any of several respects. It might be characteristically positive or negative in affect (i.e., pleasant or unpleasant), or even unconsciously aversive or attractive, and it might be beneficial or harmful, or morally good or bad. Because of such disagreements, some argue that the idea of emotional valence is of dubious value and should be abandoned.

However, it may be an important feature of emotions that they have multiple dimensions of valence. If an emotion's experiential valence is of the same sign (positive or negative) as its intentional valence—for example, an aversive emotion that is about something that is bad for you or goes against your wishes—then it is likely to promote rational decision-making and action. The actions people take to alleviate the unpleasantness or aversiveness of fear (a negative aspect of having or experiencing the emotion) tend to reduce the risk of bad things happening (a negative aspect of what the emotion is about): for example, fear of a flood leads the inhabitants to retreat to high ground, thereby averting disaster. (There are of course thrill-seekers for whom the very aversiveness of fear has a second-order attractiveness, and, within the safe confines of dramatic art, many people can enjoy the fear or "as-if" fear they empathetically experience.)

Likewise, the possible negative consequences of a decision tend to be amplified in our minds by our anticipation of regret and remorse: For example, if I buy this appealing but unreliable car, I may kick myself if anything goes wrong with it. These premonitory influences may on the whole guide us to useful behavior, in roughly the way that hunger, thirst, and sexual feelings lead us, wittingly or not, to behavior that is conducive to biological fitness. Add to this theme of doubly valenced emotions a revival of James's second premise, that emotional feelings are perceptions of bodily reactions, and we are led to the topic of the next section.

Somatic Wisdom

From Plato onward, European and North American philosophers have thought the regulation of emotion essential to a rational life, and a similar view was promoted even earlier in Buddhism and other Asian religions. The underlying supposition was that unregulated emotions are impediments to the rational life. However, this is compatible with the thesis held by a number of philosophers that emotions, or at least some of them, make a positive and possible indispensable contribution to rational decision-making. According to De Sousa, for example, emotions are indispensable for guiding our attention, our lines of inquiry, and our inferential strategies.

It was suggested earlier in this article that if an emotion's experiential valence matches its intentional valence, then it is likely to promote rational decision-making. A similar view has received support from findings in neurology and neuroscience, most prominently by the cognitive neuroscientist Antonio Damasio (1994) and his coworkers. Damasio's somatic marker hypothesis holds that successful and unsuccessful decision outcomes produce differing bodily responses—for example, as measured by skin conductance—and the accumulation of such responses over time leads to anticipatory bodily responses that guide future decision-making. One need not be aware (phenomenally conscious) of these responses in order for them to influence decision-making. However, a part of one's frontal cortex (functionally, the somatosensory or body-sensing cortex) must keep track of them. Most of the supporting data have come from observations of decision-making deficits in people with prefrontal damage and comparison with normal subjects in experimental gambling tasks. Additional data suggest that the capacity to recognize and to name certain emotions in others on the basis of their facial expressions also depends on the capacity to monitor one's bodily responses when observing them. Damasio's theory goes far beyond this, and some of it is controversial; but this brief statement makes it clear why Damasio thinks the aversiveness or attractiveness of undergoing certain emotions can be a premonitory influence that sometimes "knows" better than pure reason does which decision paths are likely to lead to preferred outcomes.

Prinz is probably the first philosopher to build a general theory of emotion on a broad and richly detailed account of empirical research. Although sympathetic to the somatic theories of James and Damasio, he argues that our emotional "gut reactions," unlike pains, tickles, and feelings of fatigue, are representational states. Applying Dretske's thesis that a state may be representational in virtue of having an evolved function of carrying a certain class of information, he argues that these bodily changes constitute perceptual appraisals or evaluations of our relationship to the environment with respect to well-being. He calls his view a non-propositional appraisal theory, because he holds that emotions need not involve propositional attitudes such as belief, judgments, and desires (2004).

The Nature of Things

The classical definitions of emotions were answers to questions of the traditional Socratic form: "What is regret?" "What is fear?" and so forth. The aim was not to capture the nuances of ordinary usage, but rather to be telling us something about ourselves: to explain, as Spinoza said, "not the meaning of words, but the nature of things" (1883 [1677], p. 178). However, if this is the ambition of the philosophy of emotion, then some philosophers would reply that it is up to science and not philosophy to tell us about the nature of things. In particular, we have to look beyond the terms of ordinary language and the concepts embedded in our everyday "folk" psychology, beyond even the best philosophical attempts to regiment these terms and concepts, if we are to discover what the emotions really are. This appears to be a special application of a more general view in philosophy of mind, that of eliminative materialism. However, whatever the merits of that general indictment of everyday psychology and any philosophy that attempts to build on it, there may be special reasons to be skeptical of traditional philosophical thinking about the emotions in particular.

Paul Griffiths maintains that we should use biological evolutionary principles of classification to determine what emotions really are. Following Paul Ekman (1992), a leading innovator on the role of facial expression in emotion, Griffiths posits surprise, anger, fear, disgust, sadness, and joy as the basic emotions. These adaptive responses are evolutionary homologues, discrete genetically ordained behavioral syndromes that are a legacy of our shared mammalian heritage. Appearing in all cultures, these adaptive responses are associated with the same facial expressions in each culture. The classification here is by descent and homology, rather than by resemblance and analogy, which is more typical of analytic approaches. The special evolutionary status of these basic emotions is reflected in Griffith's philosophical declaration that they are natural kinds. That is, they are projectible kinds: They share causal properties that are sufficiently well correlated to sustain generalizations from known to unknown cases. However, the term emotions does not designate a natural class of kind, for it would serve no scientific purpose to group them with the so-called higher cognitive emotions, such as envy, regret, and shame. Predictably, this thesis has sparked controversy, just as the general thesis of eliminative materialism did two decades earlier. Prinz counters that all emotions are valenced appraisals that exploit common aversive or appetitive mechanisms. Louis Charland (2002) suggests that there is a natural kind of organism that might be called an emoter, in virtue of having a brain that meets certain criteria of functional organization.

Normativity and Culture

Evidently all cultures have implicit rules governing at least some emotions: not just whether and how they should be expressed, but also whether and under what conditions and in what degree one should have them. In European and North American cultures, at least, emotional responses are commonly measured by standards of rationality, appropriateness, and morality. A particular instance of an emotion may be thought irrational if it is based on an irrational belief or desire. However, it is common to think an instance of emotion may be irrational even if it is not based on an irrational belief or desire; typically, because it is not suited to what it is about. We also judge instances of emotions as too little or too much, for example, in the case of grief or remorse.

It was suggested earlier that the notion of passivity, of being acted on, may be an important feature of the ordinary concept of emotion. However, it is widely assumed that people have some control over how the environment acts on them and are to some degree responsible, not only for the expression of emotion, but for their having the emotion. Aside from regulating one's exposure to eliciting situations, it is supposed that one can in many cases alter the course of the emotion—for example, by intervening cognitively to reappraise the eliciting situation. Indeed, attending to one's emotional state and labeling it may alter the state. It is plausible that when we use emotion labels in giving expression to our emotion, as in, "I'm angry!" Or "I'm in love!" We are shaping as well as describing our emotional state. Emotion kinds would thus be what Ian Hacking (1995) calls "interactive kinds," like race, ethnicity, and gender: To classify one's own state as of a particular interactive kind, or to be so classified by others, tends to alter the state and to influence one's feelings and behavior accordingly.

Social constructionists would emphasize that we are shaping our emotions to fit it into an acceptable cultural mold. The psychologist James Averill argued that the various emotion concepts are merely cultural creations that shape our assessment of certain transitory syndromes. While pretending to be passively moved to behave in certain ways, people are actively adjusting their behavior to fit these cultural categories. Although this theory is a valuable counterbalance to the widely held assumption that our emotions simply "are what they are," it would be extreme to assert that our emotion categories simply create our emotions ex nihilo or to deny that the categories themselves are, perhaps in some societies more than in others, flexible and open to change (Reddy 2003).

Emotions seem a particularly nuanced category, varying in uncharted ways from instance to instance. They also vary in the course they follow from moment to moment and day to day. For reasons such as these, as Iris Murdoch (1970), Martha Nussbaum (2001), and Jon Elster (1999) have emphasized, often the best way of defining an emotion type is by reference to literary examples. Literary examples also make it clear that conceptions of emotion vary over time as well as from one present-day culture to another.

The issues addressed in this section may seem hopelessly tender-minded to philosophers who prefer to focus on biological mechanisms and natural kinds. In turn, philosophers drawn to the issues of this section may find the naturalistic focus excessively narrow. What is chiefly at issue is the proper equipment to bring to philosophical thinking about emotions. Should we allow ourselves to conceive human organisms as people and to employ the full panoply of concepts, learned or biologically preordained, that appear to be indispensable for everyday social perception and understanding? Or should we lay aside these concepts and steadfastly conceive human beings only as complex biological systems?

Retaining our everyday tools of social perception, we will find normative questions, matters of passivity and freedom, and the richness and perspectivity of narrative understanding coming to the fore. Laying these tools aside, we can focus on purely naturalistic explanations of emotional phenomena and the natural kinds that enter into these explanations. Partisans of the naturalistic approach may be tempted to assert that only by laying aside the accustomed tools can we discover what emotions "really" are. Partisans of the other approach might argue that to lay down these tools of social perception is precisely to forego understanding people. One important challenge task for the philosophy of emotion will be to determine whether and how to reconcile these two approaches.

See also Alston, William P.; James, William.

Bibliography

Ben-Ze'ev, Aaron. The Subtlety of Emotions. Cambridge, MA: MIT Press, 2000.

Charland, Louis C. "The Natural Kind Status of Emotion." British Journal for the Philosophy of Science 53 (2002): 511–537.

Damasio, Antonio. Descartes' Error: Emotion, Reason, and the Human Brain. New York: Putnam, 1994.

D'Arms, J., and D. Jacobson. "Expressivism, Morality, and the Emotions." Ethics 104 (1994): 739–763.

Darwin, Charles. The Expression of the Emotions in Man and Animals. London: John Murray, 1872.

Descartes, Rene. The Passions of the Soul. 1649. In The Philosophical Writings of Descartes. Vol. 1. Translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch. Cambridge, U.K.: Cambridge University Press, 1984.

de Sousa, Ronald. The Rationality of Emotion. Cambridge, MA: MIT Press, 1987.

Dewey, John. "The Theory of Emotion. (I) Emotional Attitudes." Psychological Review 1 (1894): 553–569.

Dixon, Thomas. From Passions to Emotions: The Creation of a Secular Category. Cambridge, U.K.: Cambridge University Press, 2003.

Ekman, Paul. "An Argument for Basic Emotions." Cognition and Emotion 6 (1992): 169–200.

Elster, Jon. Alchemies of the Mind: Rationality and the Emotions. Cambridge, U.K.: Cambridge University Press, 1999.

Goldie, Peter. The Emotions: A Philosophical Exploration. Oxford: Oxford University Press, 2000.

Gordon, Robert. The Structure of Emotions. Cambridge, U.K.: Cambridge University Press, 1987.

Greenspan, Patricia. Emotions and Reasons: An Inquiry into Emotional Justification. New York: Routledge, 1989.

Griffiths, Paul E. What Emotions Really Are: The Problem of Psychological Categories. Chicago: University of Chicago Press, 1997.

Hacking, Ian. Rewriting the Soul: Multiple Personality and the Sciences of Memory. Cambridge, MA: Harvard University Press, 1995.

Jaggar, Alison. "Love and Knowledge: Emotion in Feminist Epistemology." Inquiry 32 (1989): 151–176.

James, William. "What is an Emotion?" Mind 9 (1884): 188–205.

Kenny, Anthony. Action, Emotion and Will. London: Routledge, 1963.

Lazarus, Richard. "The Cognition-Emotion Debate: A Bit of History." In Handbook of Cognition and Emotion, edited by Tim Dalgleish and Mick J. Power. Chichester, NY: John Wiley and Sons, 1999.

LeDoux, Joseph. The Emotional Brain: The Mysterious Underpinnings of Emotional Life. New York: Simon and Schuster, 1998.

Lyons, William. Emotion. Cambridge, U.K.: Cambridge University Press, 1980.

Murdoch, Iris. The Sovereignty of Good. London: Routledge, 1970.

Neu, Jerome. A Tear is an Intellectual Thing: The Meanings of Emotion. New York: Oxford University Press, 2000.

Nussbaum, Martha. Upheavals of Thought: The Intelligence of Emotions. Cambridge, U.K.: Cambridge University Press, 2001.

Oatley, Keith. Emotions: A Brief History. Oxford: Blackwell, 2004.

Ortony, A., and T. J. Turner. "What's Basic About Basic Emotions?" Psychological Review 97 (3) (1990): 315–331.

Panksepp, Jaak. Affective Neuroscience: The Foundations of Human and Animal Emotions. New York: Oxford University Press, 1998.

Prinz, Jesse. Gut Reactions: A Perceptual Theory of Emotion. Oxford: Oxford University Press, 2004.

Reddy, William. The Navigation of Feeling: A Framework for the History of Emotions. Cambridge, U.K.: Cambridge University Press, 2003.

Roberts, Robert C. Emotions: An Essay in Aid of Moral Psychology. Cambridge, U.K.: Cambridge University Press, 2003.

Robinson, Jenefer. "Startle." The Journal of Philosophy 92 (1995): 53–74.

Rorty, Amelie. "Explaining Emotions." In Explaining Emotions, edited by Amélie Rorty, 103–126. Los Angeles: University of California Press, 1980.

Russell, James A. "Core Affect and the Psychological Construction of Emotion." Psychological Review 110 (1) (2003): 145–172.

Solomon, Robert. The Passions. New York: Doubleday, 1976.

Sorabji, Richard. Emotion and Peace of Mind: From Stoic Agitation to Christian Temptation. Oxford: Oxford University Press, 2000.

Spinoza, Benedict. Ethics. Translated by R.H.M. Elwes. London: George Bell & Sons, 1883.

Wollheim, Richard. On the Emotions. New Haven, CT: Yale University Press, 1999.

Zajonc, Robert, B. "The Interaction of Affect and Cognition." In Approaches to Emotion, edited by Klaus Scherer and Paul Ekman. Hillsdale, NJ: Erlbaum, 1984.

Louis C. Charland (2005)

Robert M. Gordon (2005)