Emotional Energy as the Common Denominator of Rational Action

1. A full statement of the theory is complicated by the fact that EE is affected not only by the intensity of interaction rituals but by differences in power. Power is asymmetrical in a way that IRs are not, and it adds a conflictual element to the dynamics of emotional energy. Here, I present an analytically simplified version of the theory of emotional energy in order to bring out its congruence with the theory of rational action. For the more elaborate version including power relations, see Collins (1990).

2. There is evidence of rhythmic coordination in intensely focused groups of many different kinds. Gestures become coordinated in political crowds (Wohlstein and McPhail 1979); speaker and audience can anticipate each other's actions in emotional speeches (Atkinson 1984); in two-person conversations, acoustical frequency levels of voices become attuned as the conversation becomes more engrossing (Gregory, Webster, and Huang 1990; see further evidence cited in Collins 1988, 261-62).

3. Sociophysiology (Barchas and Mendoza 1984) holds that an individual's physiological condition is affected by current and recent social experience. Face-to-face social interaction takes place among physiological systems, not merely among individuals as cognitive systems or bodily actors. From an evolutionary perspective, it is not surprising that human beings, like other animals, are neurologically wired to respond to each other and that social situations that call forth these responses are experienced as highly rewarding.

4. It is for this reason that occasions of political crisis or revolution produce intense emotions and an upsurge of idealistic beliefs; these effects are concentrated among those persons who take part in the highly focused crowds in the major gathering places and are much weaker among persons who remain unmobilized in the routine of their everyday lives. These differences are apparent, for example, in media reports on the attitudes of citizens of the disintegrating Soviet Union in the summer of 1989, in the contrast between the highly mobilized urban centers and the more apathetic residential or rural areas.

5. Emotions of love among sexual partners are affected by other factors besides achievement of high rhythmic coordination in love making; Heimer and Stinchcombe (1980) pointed to shared play as well as to conversational intimacy. One may regard this is a combination of several types of IRs: those that produce friendship ties as well as erotic ties. The combination of these is what is typically called “love” in the marketplace of individual sexual and marital ties that characterizes modern societies.

6. See Frijda (1986). This satiation is not shown in Figure 2 but could be indicated by further rows and columns showing diminished payoffs for extremely long periods of sustained attention. In Figure 1, the feedback loops as shown would result in a continuous escalation of all variables; a more complicated flow chart would indicate the points at which emotional satiation occurs.

7. One can see this by drawing the long-term feedback loops in Figure 1 from the outcomes at the right—solidarity, emotional energy and symbols of group membership—to the facilitating conditions for interaction given at the left. If the cycle is to be broken, it must happen because of change in exogenous conditions affecting the assembling and focus of the group.

8. The term “emotional energy” clashes in part with the common usage of “emotion,” which is usually confined to the dramatic and disruptive emotions such as anger and excitement. But such dramatic emotions are best explained as departures from a baseline of long-term emotional energy (see Collins 1990). My focus on EE as the central emotion in social interactions is an attempt to specify what Durkheim meant by pointing to the “moral force” or “energy” which comes from social rituals. Durkheim ([1912] 1954) spoke of the result of ritual solidarity in making participants feel strong. EE has a moral dimension; one feels that one's actions are right, that they are supported and approved by the group. EE is related to the continuum from pride to shame, which Scheff (1990) argued is the main result of the continuum from intact to broken social ties. The sociology of emotions is a relatively young field of research, and work continues in establishing the relationships among fundamental and derived emotions and their sources in particular types of social interactions; for an overview see Kemper (1990).

9. In Figure 1, the sinks at the far right-hand side of the diagram indicate that not only EE but group solidarity and group symbols lose their force over time, in the absence of processes that replenish them.

10. Collective symbols might also be referred to by Bourdieu's (1984) term “cultural capital.” They are symbolic possessions that may be invested in further interactions and are subject to constraints of a market, including deflation of the value of the currency as it becomes more abundant. My use of the concept of “cultural capital” or “collective symbols” is somewhat broader than Bourdieu's and includes all items of culture charged up by interaction rituals.

11. Empirical evidence relevant to this point shows that individuals in higher-ranking occupations and in occupations exercising autonomous power over others, are more committed to their work, work longer hours, and are more likely to allow work to spill over into their private lives (Kanter 1977; Rubin 1976; Kohn and Schooler 1983; Gans 1962). Studies of work situations across occupations have not focused explicitly on the IR density of such situations, but available evidence is in keeping with the proposed pattern. A related line of investigation suggests a relationship between the tightly focused interpersonal groups within modern Japanese business organizations and their tendency to work long hours with few vacations (Nakane 1970); one might describe this as a high interaction ritual density within Japanese organizations.

12. The marginal substitutability of money earned for effort put into working should decline at high levels of income. But high-intensity IRs generate energy, so there is no increasing taste for leisure in such situations.

13. As described elsewhere (Collins 1990), exercising power over other persons increases one's emotional energy; being subjected to someone else's power decreases one's EE. Individuals who give away considerable power to others, such as joiners of charismatic movements, are those who had little prospects of acquiring power for themselves in their own IR marketplace. They have little power-generated EE to sacrifice and will receive much higher EE payoffs from total commitment to ritual participation.

14. Garfinkel's ethnomethodology is congruent with Simon's (1957) analysis on this point; Garfinkel (1967) enhanced the depth of the problems faced by the human cognitive agent by pointing to irremediable sources of ambiguity in defining collective reality. Hence Garfinkel's actor is even more conservative than Simon's, preferring to take most things for granted rather than to have to consider their justification and their alternatives.

15. In the IR model, the actor maximizes EE overall. Satisficing is a procedure for dealing with a large number of different arenas of action simultaneously; the purpose of satisficing, however, is to maximize overall EE. Simon's (1957) satisficing model has no way to maximize across situations because it lacks a common denominator.