Most of the arguments about specific aspects of emotion can be traced to a more fundamental difference between what may be called the organismic and the interactional view­points. Before I summarize these viewpoints and state my own position, it will be useful to acknowledge two barriers to any serious inquiry on this matter: first, the practice among social scientists of ignoring emotion or subsuming it under other categories; and second, the acceptance of several ideas about emotion that confuse any discussion of it.

Some theorists have gone so far as to deny that emotion is a tenable concept. Thus the psychologist Elizabeth Duffy, after distinguishing between longitudinal concepts (which describe phenomena that occur sequentially) and cross-sectional con­cepts (which describe phenomena such as perception, thought, and emotion, which occur simultaneously) argued for dispensing with cross-sectional concepts altogether. She was correct to point out that they represent loose and over­lapping categories of phenomena (1941, p. 184). Unfortu­nately, her alternative simply eliminates the complexity we ought to be trying to describe. The same objection applies to social psychologists who believe that the exquisite care they take to avoid discussing feeling, in order to focus ever more intently and narrowly on cognition, increases the scientific character of their work. A content analysis of their own per­sonal speech habits over an average week would certainly

show that emotion is more central to life as they live it than to life as they study it.

Many social psychologists give emotion short shrift by subsuming it under some conceptual umbrella. For example, in an otherwise informative study of soldiers’ attitudes to­ward the Women’s Army Corps in 1950, Suchman and col­leagues subsume emotion under the concept of affect: “Af­fect toward an object can be very generally classified as either positive or negative. For our purposes, however, an­noyance, anger, distrust, and fright are all shadings of nega­tive affect, and these shadings we shall ignore” (cited in New­comb et al. 1965, p. 48). When emotion is subsumed in this way, the interesting dimension of emotion becomes the “how much.” What precisely there is “a lot” of or “a little” of is unclear. We lose the distinction between a fearful dislike of the Women’s Army Corps and an angry dislike of it. We lose a wealth of clues about the various definitions of reality that people apply when adopting an attitude. We lose the idea that emotions reflect the individual’s sense of the self-rele­vance of a perceived situation. We lose an appreciation of what the language of emotion can tell us.*

For those who do not deny or subsume emotion, two other ideas sometimes obstruct our clear understanding of emotion. These are: (1) The idea that an emotion, like anger or jeal­ousy, can have an independent presence or identity within a person through time. (2) The idea that when possessed by emotion we are led to act irrationally and see distortedly. Be­cause these notions are sometimes applied by writers in both the organismic and the interactional camps, we should exam­* There is loss when emotion is conceptually ripped away from the situation to which it is attached. When Aristotle discusses his fifteen emotions, Descartes his six, Hobbes his seven, Spinoza his three (with forty-eight derivatives), McDougall his seven, and Tomkins his eight, the immediate relation of emotion to viewpoint or frame is lost. This is also a problem with Joel Davitz’s otherwise interesting attempt to formulate a dictionary of emotions (1969). Just as modern linguists now examine language as it is used in social context, so emotion, another sort of language, is best understood in relation to its social context.

ine their content before turning to the assumptions that di­vide the organismic and interactional theorists.

Does emotion have a presence or identity independent of the person it is “in”? We talk as if it did. We commonly speak of “expressing,” “storing,” “getting in touch with,” or even “spreading” an emotion. We speak of guilt as something that “haunts” us, or fear as something that “grips,” “strikes,” “be­trays,” “paralyzes,” or “overwhelms” us. Fear, as we talk about it, is something that can lurk, hide, creep, look up, or attack. Love is something we fall into or out of. Anger is something that overtakes or overwhelms us. In this way of talking we use the fiction of some independent, outside agency in order to describe a contrasting inner state.

As Roy Schafer points out in A New Language for Psycho­analysis (1976), the very way we normally talk about an emo­tion, our very use of nouns such as “anxiety” or “love” or “anger” suggests entity. Even verbals such as “fearing” or “dreading” (we can’t speak of “anxiousing”) are themselves abstractions and carry the same implications as the nouns they replace. Schafer proposes a new action language as a substitute for common parlance. He would remove expres­sions like “to fear” or “fearing” because they refer abstractly to a number of separate actions and modes of action; thus “to fear may subsume to flee, to avoid, to act timidly, or pla­catingly” (p. 275). Though Schafer is perceptive in identify­ing common expressions that embody problematic assump­tions, his action language seems to me too simple an apparatus for coping with the complexity of everyday emo­tional life.

Commonly we find ourselves speaking of emotion as if it had a location or residence. When we speak of love as residing in the heart and envy in the bile, the heart and the bile are put in place of the person. The speaker personifies an organ or portrays emotion as “a substance or quantity of energy of a certain kind.” We also speak of emotions as having some sort

of continuous identity, as when we say an emotion is “stored” or “accumulated,” or when we refer to an “old” emotion.

Metaphors that suggest agency, residence, and continuity through time often convey with uncanny precision just what it feels like to experience an emotion; they enjoy a poetic ac­curacy. But they can get in the way of understanding how emotion works.

A second idea that impedes our understanding of emotion is that the inner state of emotion is always associated with outer action that is irrational. This is sometimes the case and sometimes not. A man who feels fear at the sight of a rattle­snake moving toward him may run to safety. He may act ra­tionally. Were he not afraid, he might not run, which in the absence of other forms of protection, would be irrational. Again, a mother may, with the feeling of love, reach out to hug her child. Here, too, the feeling and act seem consonant and “rational” in the sense that what a person does, under the in­fluence of feeling, gets people where they want to go as much if not more than what a person would do if not under the influence of feeling. They only reason I pose these obvious examples is that when people talk about “acting emotionally,” it is often not these examples they cite. That is, we tend to associate the idea of emotion more with irrational or unwise actions than with rational or wise actions. This tendency results more from our cultural policy toward emotional life (“watch out for it, manage it”) than it results from observing the relation between feeling and action in all the common but inconspicuous instances in which they are related.