NOTES
Chapter 1 The Program of Interaction Ritual Theory
1.  Even more misleading is the usage in a pejorative sense as an unthinking going through the motions, or meaningless fixation on mumbo-jumbo.
2.  There is yet another usage in the field of animal ethology, sometimes borrowed in child development psychology. Here "ritualization" means abbreviated communicative gestures, which operate as "signals" to another organism in a habitual process of action, in contrast to "symbols," which are conventions for referring to shared meanings (e.g., Tomasello 1999, 87). In this usage, ritualization is just a shorthand used in coordinating practical action, not a source of symbolic intersubjectivity. Despite the fact that the terminology is more or less reversed from that of sociological IR theory, we shall see in chapter 2 that work by Tomasello and others of this group of researchers does indeed corroborate important parts of IR theory.
3.  Freud's Totem und Tabu appeared in 1913, at the height of interest in these phenomena. Van Gennep's.Rzfes de Passage was published in 1909, Frazer's To-temism and Exogamy and Lévy-Bruhľs Les fonctions mentales dans les sociétés in-férieures in 1910, Durkheim's Les formes élémentaires de la vie religieuse in 1912, the same year as Harrison's Themis, Cornford's From Religion to Philosophy, and Murray's Four Stages of Greek Religion. Stravinsky's controversial ballet music about a primitive rite, Le sacre du printemps, was first performed in Paris in 1912.
4.  The line of criticism that the functionalist movement took against their predecessors made an exception for the Durkheim school, since Durkheim was strongly committed to a general science of society. Durkheim and Mauss paved the way for the functionalist program of studying rites and beliefs in their current context of social action rather than as isolated survivals of past history, although they also had more evolutionary concern with social change than did the movement of functionalist anthropologists. Durkheim did adopt an evolutionist stance that enabled him to view Australian aborigine society, because of its apparent simplicity, as an "elementary form" that would show both the evolutionary starting point of more complex societies, and reveal the analytically central processes of social solidarity and symbolism. Thus, although Malinowski was the organizational leader of what became known as the British school of social anthropologists, the members of that school tended to adopt Durkheimian theoretical formulations. This was notably the case with Radcliffe-Brown, who taught in South Africa and Australia independently of the group (Evans-Pritchard, Fortes, and others) that came from Malinowski's seminar at the London School of Economics. Radcliffe-Brown was in contact with the Durkheim school via Mauss since the mid-1920s, and explicitly developed its theory of rituals (Goody 1995).
376                                      NOTES TO PAGES 14-21
5.  There was a direct network transmission: Parsons was a member of the Malinowski seminar at LSE in the early 1930s, before beginning work on his systematic structural-functional theory (Goody 1995, 27); Merton was a student of Parsons at Harvard in the mid-1930s.
6.  Durkheim (1912/1965) also analyzed mourning rites, but what he was concerned to show was that mourning is not spontaneous but obligatory by the group. He notes that assembling the group at the funeral results in a type of collective effervescence, albeit one based on a negative emotion. This gives us the mechanism by which Radcliffe-Brown's (1922) functional integration is carried out: the collective emotion initiated by shared grief pulls individuals back into the group and gives them renewed strength.
7.  "The rules of conduct which bind the actor and the recipient together are the bindings of society. . . . Opportunities to affirm the moral order and the society could therefore be rare. It is here that ceremonial rules play their social function... . Through these observances, guided by ceremonial obligations and expectations, a constant flow of indulgences is spread through society, with others who are present constantly reminding the individual that he must keep himself together as a well demeaned person and affirm the sacred quality of these others. The gestures which we sometimes call empty are perhaps in fact the fullest things of all" (Goffman 1956/1967, 90).
8.  Old-fashioned usage best conveys the sense that "society" has in IR theory. Society is not a distant abstraction; it means what an upper class matron at the turn of the twentieth century would mean if she spoke of her daughter "going out into society"—i.e., going out from the domestic circle to take part in polite social gatherings. IR theory generalizes this usage from a restricted sense of "polite society" to all social interaction in its ritualistic aspect. The sense is similar to that in Henry David Thoreau's epigram: "I have three chairs in my house; one for solitude; two for company; three for society."
9.  According to newspaper reports, the controversial basketball coach Bobby Knight was fired by his university in the late 1990s after he responded angrily to a student who accosted him with "Hello, Knight."
10.   "The Meadian notion that the individual takes towards himself the attitude others take to him seems very much an oversimplification. Rather the individual must rely on others to complete the picture of him of which he himself is allowed to paint only certain parts. Each individual is responsible for the demeanor image of himself and the deference image of others, so that for a complete man to be expressed, individuals must hold hands in a chain of ceremony.. . . While it may be true that the individual has a unique self all his own, evidence of this possession is thoroughly a product of joint ceremonial labor" (Goffman 1956/1967: 84-85).
11.  It follows, although Goffman does not go into this, that overt conflict breaks out when the normally accepted level of ritual cooperation is broken, for instance, by a new manager becoming unusually assertive in invading the workers' backstage. For an empirical case, see Gouldner's Wildcat Strike (1954).
12.  In terms of the detailed model of IR to be presented later, this is typically done by keeping up the normal rhythm and focus of attention via the flow of topics, and through an emotional tone appropriate to a nonhostile interaction,
NOTES TO PAGES 23-32
377
while giving either a content or a nonverbal / paralinguistic gesture that challenges the other person's demeanor as a competent self.
13.  Goffman's critiques of these interpretations were not systematically stated but were carried on rather offhandedly in passing remarks and footnotes. On Goffman's intellectual stance vis-ä-vis other positions, see Winkin (1988); Burns (1992); Rawls (1987).
14.  Goffman broached this topic in early work, raising the question of whether there is an ultimate backstage, where no performing work in done. Yet even sexual intercourse, which is normally regarded as the realm of uttermost intimacy and privacy, Goffman avers, may be regarded as a kind of performance (1959, 193-94). I will extend this theme in chapter 6, where we see just how fully sexual intercourse fits the model of interaction ritual.
15.  Around the same time (the late 1950s and early 1960s), Chomsky developed his quasi-mathematical analysis of the deep structures of language, concentrating not on phonetics, as Saussure had done, but on syntax. The program was called generative grammar since it laid out a formal system by which the surface structure of language is generated from underlying elements and transformations. The Chomskyian method starts with existing sentences in a natural language and decomposes them into fundamental elements called "underlying strings." Turning back in the opposite direction, these elements are reassembled by applying a series of operations (phrase structure rules, transformational rules, morphophonemic rules) until we end up again with recognizable sentences. The set of operations may thus be said to have generated the particular sentence. Chomsky's explanatory strategy is parallel to Levi-Strauss's, although differing in substance. But Chomsky's generative grammar did not yield a code for different languages that lined up with Lévi-Strauss's types of kinship systems, or other elements of social structure, and thus provided no support for Lévi-Strauss's grand system.
16.  Saussure, writing at the turn of the twentieth century, was in sympathy with Durkheim's program. There was some theoretical resemblance insofar as the arbitrary differences that constitute phonetic significance are a collective product, not explainable by the psychology of individuals (Saussurel915/ 1966, 15-16; for detailed references on Durkheim's influence on Saussure, see Jameson 1972, 27). Lévi-Strauss, a protege of Mauss and grandpupil of Durkheim, in effect brought together several branches of a broad Durkheimian community.
17.  The reasons were to a considerable extent political, a reaction led by the existentialist generation of the 1930s and 40s against what was seen as the Durkheimian school's advocacy of solidaristic nationalism in French politics of the 1920s (see Heilbron 1985; Collins 2003).
18.  This view of ritual is often attributed to Durkheim. Catherine Bell (1992) straightens out this issue of interpretation, and develops some of the possibilities that open up when ritual is seen as social action producing and reproducing a symbolic code rather than vice versa. Other scholars occupy a halfway position, seeing the code or repertoire of codes as temporally prior and analytically primary but allowing considerable flexibility in the way in which codes are invoked in particular situations.
378                                      NOTES TO PAGES 33^0
19.  There is, however, at least one way in which the view that meanings are constructed by secular, worldly activity, can be reconciled with some aspect of religious transcendence. David Preston (1988), in analyzing the techniques of Zen Buddhism, argues that Buddhist meditation practices are socially organized not to construct a transcendental religious meaning, but to strip away the accretion of meanings already constructed; thus the aim of this meditation is liberation or transcendence from everything constructed, allowing whatever remains (a transcendental reality?) to come shining through.
20.  The most important of these is The Elementary Forms of the Religious Life, Book 2, chapter 7. There are several English translations, which differ mainly in arbitrary stylistic renditions, and in pagination.
21.  The importance of the body for sociology has been underlined in recent years, especially by Bryan Turner (1996).
22.  It may be worthwhile here to head off several misconceptions or prejudices about Durkheim's style of argument. Durkheim's analysis of the emotions of group assemblies in The Elementary Forms is sometimes regarded as a version of the "crowd psychology" of the turn of the twentieth century, as in Le Bon's The Crowd (1908). Thus it is thought that Durkheim is merely repeating conventionl arguments for the leveling and animalistic effects of the mob triumphing over the higher rationality of the individual. A good deal of modern sociology of collective behavior and social movements, such as McPhail's The Myth of the Madding Crowd (1991) uses this model as a foil, citing evidence that individuals typically take part in crowds not as isolates but in a small circle of friends who accompany each other. Against this critique, three points need to be made.
First, modern network research on crowds does not undercut the significance of group influences on individuals; it merely replaces one model of group behavior with another. A better way to interpret the data is that the primary groups that make up the crowd facilitate and amplify the effects of the larger crowd as a focus of attention and emotional entrainment. For members of these little groups, the larger crowd's cheering or other emotion becomes especially significant because they feed it back and forth among each other. Thus we can also say that these small groups entrain each other in a larger group.
Second, Durkheim does not view group assemblies as animalistic, lowering the individual to a subhuman level. On the contrary, he points to assemblies as the occasions where moral ideals are created and put into action. Heroic, self-sacrificing, highly moral individuals are created by experience in such occasions.
Third, we cannot assume, Robinson Crusoe-like, that the rational individual preexists all social experience, and thus that crowds are simply made up of individuals who might or might not be brought down from their natural level of rationality. Durkheim is attempting to show how individuals are formed or socialized by groups, and how the conceptions that make up their rationality are formed and inculcated in them.
Another, minor misconception can also be addressed: that Durkheim singled out aborigines in order to look down on them as primitives below the level of modern rationality. On the contrary, Durkheim focuses on gatherings of abo~
NOTES TO PAGES 42-53
379
rigines because they are displaying our common humanity. The processes that he singles out, the focus of attention and shared emotion that generate collective effervescence, are in basic outline the same as those that operate throughout history and continue to operate today.
23. In this respect, Bourdieu continues what I have called the "code-seeking" program of Lévi-Straussian structuralism; and this is so even though Bourdieu (1972/1977) established his theoretical reputation by taking Lévi-Strauss as a foil against which to emphasize that symbolic capital is always used in a practical way in the contingencies of ordinary life. Bourdieu avoids the term "code," substituting "habitus" for its component as internalized in individuals, and the "logic" or "principle" of "fields" for the overarching macro-pattern. In later works, Bourdieu emphasizes the logic of "practice," borrowing microsociological insights from Goffman and the ethnomethodolo-gists, and denying that the overall structure does anything apart human agents. But the outcomes are structurally preordained nevertheless. For Bourdieu, the enactment of culture by individuals (including the use of language: see Bourdieu 1991) is always effective in reproducing the same kind of stratified social order or "field of power"; hence he calls it "symbolic violence," stressing its character as the micro-instantiation of macro-domination. In another version of the terminology this is called the "homology among fields," a concept that betrays the Lévi-Straussian structuralism from which it was taken. For a typical application of this style of argument see Bourdieu (2001), where he claims that the deep structural logic of gender domination remains the same from the extreme masculine domination of ancient Mediterranean tribes through the liberalized Western societies of the late-twentieth century. For a critique of Bourdieu, see Lamont and Lareau (1988). See also the debate between Bourdieu's follower Wacquant (2002) and Anderson (2002), Dunier (2002), and Newman (2002).
Chapter 2 The Mutual-Focus / Emotional-Entrainment Model
1.  For another version of formally modeling ritual, see Marshall (2002).
2.  A combination of personal and collective rebellion against formal ritual, even down to the level of Goffmanian politeness, was characteristic of the "counterculture" of the 1960s. The consequences for the shifting style of interactional stratification will be examined in chapters 7 and 8.
3.  The president of a major American university during the 1990s, known for his extremely gregarious and affable style of greeting and interacting with faculty, students, visitors, potential donors, and indeed all comers, had a breakdown after a few years on the job and had to resign his position. As a child, I observed my mother, who as wife of the U.S. Consul General in an overseas post was the leading hostess of the local diplomatic corps, throwing herself with great emotional effervescence into the round of sociable rituals. But it was clearly a staged effort, as I could see from the change in her mood as soon as she closed the door on the last guest; and periodically, she would
380                                      NOTES TO PAGES 54-63
take a break, retiring to a hotel to read novels and see no one at all for a week. Turner (2002) argues that human beings are descendents of a rather unsociable lineage of primates, and hence that humans are in fact not naturally very sociable and must put considerable work into keeping up rituals. I believe that he exaggerates the evidence for humans' unsociable biological heritage; and his argument that humans have to work hard at putting on rituals appears to be drawn from observations of forced rituals.
4.  A plausible reconstruction of the evolutionary pathway is given by Turner (2002, chapters 3 and 4), using the evidence of paleontology and cladis-tics, primate behavior, and brain physiology. Turner emphasizes that the human animal became unique by developing elaborate emotional expressions that enabled much more refined social coordination than other animals, and that tied these emotions to the cognitive centers of the brain.
5.  The example of sports celebrations shows historical differences. American sports celebrations were more restricted before 1970, consisting of handshaking, and some shoulder hugging among close teammates. In the early part of the century, a typical ritual was to carry the coach or player-hero off the field on the shoulders of the team—a restricted form of bodily contact concentrated on one token representative. Late twentieth-century style (which has continued into the early twenty-first century) consists in full body hugs and piling on in a heap of bodies. Thus, even rather informal ritual patterns are influenced by an accumulation of tradition; we have little analysis of what conditions bring about shifts from one pattern of bodily celebration ritual to another. All of these patterns, nevertheless, are variants within a basic pattern: the sudden eruption of strong emotions arising from group experience leads to a desire for bodily contact, which in turn heightens and prolongs that peak emotion. The prolongation in fact may not be so very long: from a few dozen seconds or less of peak excitement if there is no bodily solidarity ritual, to something on the order of ten minutes in maximal celebrations. Some evidence is provided by film and video-recordings of the celebrations that followed Roger Maris setting a new home run record in 1961 (forty seconds of applause) and Mark McGwire breaking the record in 1998 (nine minutes of applause). During the earlier record moment, bodily contact consisted of handshaking; during the later, of a round of bodily hugs with teammates and others.
6.  In the ceremonial drinking traditions of Sweden, a toast to an esteemed acquaintance is carried out by looking each other in the eye at the moment of emptying one's glass.
7.  The ritual character of drinking together explains the mild taboo or sense of social shame that exists for drinking alone. Although the disapproval is rationalized as referring to a sign of alcoholism, the sense is more one of misuse of a ritual substance. There may well be heavier drinking in collective gatherings than solo, but drinking with others is generally seen positively through the veil of solidarity. A similar mechanism has contributed to the delegitima-tion of smoking in the late twentieth century, as we see in chapter 8.
8.  It can be argued that not just the student audience feels deprived by hearing lectures on remote hookup; the speaker to a remote audience feels especially acutely the lack of feedback from the audience, unless there is an audi-
NOTES TO PAGES 66-67
381
ence immediately present as well. It is generally harder to lecture to an extremely large classroom, because one cannot gauge the reactions of the students who are far away.
9.  This explains patterns found in psychological experiments that show there is more laughter at comic material when there are sounds of laughter, when subjects can see the laughing audience, and when the group is larger (Leventhal and Mace 1970; Provine 1992; Yong and Frye 1966; Bush et al. 1989).
10.  Katz (1999) demonstrates the importance of social participation and, even more importantly, of mutual focus of attention, for happy laughter to take place. Using recordings of visitors to the hall of fun-house mirrors in an amusement park, Katz shows that individuals do not automatically laugh at their distorted images. Instead, they call for others in their group (usually family members) to come and see the image, whereupon they encourage each other by bodily motions and vocal rhythms to build up laughter together. Bystanders, who are seeing the same images in the mirrors, do not join in the family group's laughter; it is not the funny stimulus that causes the laughter, but the social entrainment. These instances display very sharply the boundaries of inclusion and exclusion that are manifested and re-created in the collective experience of producing laughter.
11.  Here, too, Durkheim provides a precedent, in emphasizing the importance of shared rhythm in establishing a condition of collective effervescence:
And since a collective sentiment cannot express itself collectively except on the condition of observing a certain order permitting co-operation and movements in unison, these gestures and cries naturally tend to become rhythmic and regular; hence come songs and dances. . . . The human voice is not sufficient for the task; it is reinforced by means of artificial processes; boomerangs (in the Australian aborigine ritual) are beaten against each other; bull-roarers are whirled. It is probable that these instruments ... are used primarily to express in a more adequate fashion the agitation felt. But while they express it, they also strengthen it." (Durkheim 1912/1965, 247)
12.  This fits with the propensity of ethnomethodologists to refer to their subjects as "members," as if taking for granted that persons are already part of a culture. In this respect ethnomethodologists follow the assumptions of researchers in cognitive anthropology (D'Andrade 1995). The IR tradition, on the other hand, prefers to start with human physical bodies in interaction and to derive culture from they way in which they coordinate their attention. Thus Goffman has at times been tagged as a kind of human animal ethologist.
13.  Such "rules" are just an observer's way of characterizing these regularities. It would be a mistake to assume that there is a cultural blueprint that the actors are referring to for how to talk. I would argue, to the contrary, that the mechanism of rhythmic coordination is naturally given in all human beings (indeed, possibly in many animals), and that violations of it are universally felt as breaking solidarity. Sacks et al. argued like structuralists of the code-seeking school, perhaps because they oriented toward their most accessible scholarly audience, anthropological linguists.
382
NOTES TO PAGES 68-71
14.  This conversation has no socially recognized gaps. It does have a small amount of overlapping, indicated by the brackets [ ] which show when two speakers are talking at the same time. This is normal; overlap happens just at the points where one speaker might be ending her utterance, and the other is starting to say something so that there will be no empty space. But as soon as they both recognize what is happening, one of them stops and lets the other talk.
15.  This examples illustrates Simmel's point that conflict is also a form of sociality, in contrast to the thorough breaking of social ties that is obliviousness, or the withdrawing of attention. We can say that conflict is a disputed effort to dominate a situation of social coordination, to bend mutual-focus / emotional-entrainment to one's advantage over the resistance of the other. I
will draw out the implications of this line of argument in a future work on violent conflict (Collins, forthcoming).
16.  Conversation analysis, with its roots in ethnomethodology is concerned with contextual meaning of utterances as provided by the sequence of what has just been addresssed; and with enacting social structure from moment to moment as an ongoing achievement (Heritage 1984; Schegloff 1992). Like its intellectual parent, conversation analysis focused on the production of a sense of social structure in general, rather than on mechanisms of variability from one social situation to another.
17.  There is cultural variability within social classes as well. Educated upper-middle-class persons are more likely than working-class persons to have hesitation pauses in their speech (Labov 1972). These are gaps that occur, not between turns, but in the middle of a sentence; the inference (which can be confirmed by subjective experience) is that these are times when the person is searching through alternative words to say next. Thus persons who have more cultural capital, as well as a more reflexive style of thinking, will have more hesitation pauses than other persons. Bernstein (1971-75) describes a similar phenomenon as the "elaborated code" of the middle class and the "restricted code" of working-class speech: the latter flows off more straightforwardly, because it consists to a larger degree of formulaic utterances. The hypothesis of IR theory is that hesitation pauses of this sort are more disruptive to conversational solidarity when they occur between social classes, that is, when one side is used to an uninterrupted rhythm, which is not forthcoming. Conversely when two members of the upper-middle class engage in conversation, hesitation pauses are more easily accomodated—although it would remain true that utterances with a more continuous rhythm delivered within the elaborated code would generate more solidarity as well.
18.  A second cross-cultural objection is that there are cultures in which it is typical for several persons to speak at once; sociability in Italy is often described as many animated conversations going on across a dinner table, with the same individuals attempting to keep up with each one. This is a complicated case that awaits further analysis. It is not clear, for example, whether there are several different circles of conversation going on at the same time, in which particular individuals may try simultaneously to participate; this would not violate the no-overlap rule for any particular conversation. Alternatively,
NOTES TO PAGES 72-83
383
it may be that the speaker and the addressee are both speaking at the same time, which would imply disattention to the other's words and an aggressive effort to usurp the floor (see evidence in Corsaro and Rizzo 1990). This would need to be studied in careful micro-detail.
19.  Durkheim gives an explanation in discussing the elevated language of a public speaker. He refers to
the particular attitude of a man speaking to a crowd, at least if he has succeeded in entering into communication with it. His language has a grandiloquence that would be ridiculous in ordinary circumstances; his gestures show a certain domination; his very thought is impatient of all rules, and easily falls into all sorts of excesses. It is because he feels within him an abnormal over-supply of force which overflows and tries to burst out from him; sometimes he even has the feeling that he is dominated by a moral force which is greater than he and of which he is only the interpreter. It is by this trait that we are able to recognize what has often been called the demon of oratorical inspiration. Now this exceptional increase of force is something very real; it comes to him from the very group which he addresses. The sentiments provoked by his words come back to him, but enlarged and emplified and to this degreee they strengthen his own sentiment. The passionate energies he arouses re-echo within him and quicken his vital tone. It is no longer a simple individual who speaks, it is a group incarnate and personified. (Durkheim 1912/1965, 241)
20.  Another comparison helps bring out the mechanism: a crowd of human bodies on a street is mildly exciting; but a crowd of automobiles on a highway is just a traffic jam. Both are unfocused crowds, but the crowd of cars lacks even the minimal mutual interchange among human bodies that people passing on the sidewalk have. Katz (1999) shows that the frustrations of driving come precisely from those moments when the lack of mutual feedback becomes most palpable.
21.  The different components are brought out in the experience of being in a wave going around a stadium: you have first a sense that the crowd action is bearing down on you and that you are being pushed to rise up with others nearby at just the moment when the wave arrives, and then the sensation that you are pushing the people beyond you to join in.
22.  In between the murderous violence of ethnic riots and the cheering or jeering of the sports crowd is the destructive victory celebration, or protest, that sometimes breaks out at sports events. The organization of English soccer hooligans illustrates how the intense collective experience of violent participation becomes the main attraction, to be deliberately scheduled and enacted (Buford 1992). In effect, such activity becomes an addiction, not so much to violence but to the excitement and collective identity produced by violence (King 2001).
23.  Crowds are generally made up of small subgroups of friends and acquaintances, but these subgroups are anonymous to each other.
384
NOTES TO PAGES 84-94
24.  I owe this information to liana Redstone and Kirsten Smith, reporting on their experience as interviewers and observers in Togo and Malawi in the late 1990s. For wide-ranging cross-societal comparisons, see Mauss (1938/1985).
25.  This is a respect in which Burt's (1992) emphasis on bridging ties, across holes in networks, is overstated. Redundant networks make an important complementarity to bridge ties, because the former enhance reputation, which may be an even more important resource than scarce information when the political task at hand is to put together a coalition.
26.  There is a third way in which symbols circulate and prolong the sense of membership: symbols recirculate in the inner conversations that make up thinking in an individual's mind. These symbols are offshoots of the first two kinds; they begin as internalizations of them, although they can be modified and developed in internal conversations. These complexities will be considered in chapter 5.
27.  I leave aside here the ways in which symbols may cross over from one circuit to the other. Principally, audience-shared symbols may also be used in personal converational networks. But because these symbols are so widely available, they carry no great significance for personal relationships, and thus their exchange in sociable conversation does not bring about very close ties, which is to say strong membership solidarity. Pretty much everyone can talk about the local sports team, so that kind of conversation does not differentiate good friends or close professional or business allies. Differentiation in sociable ties, however, may occur not so much via the topics as the length to which conversationalists will go in talking about them. Particularly among youths (who have little stock of symbolic memberships from their work experience), the strength of a personal friendship tie comes not so much from the unique content but from their mutual willingness to talk about their entertainment-heroes / sacred objects at inordinate length.
Another possible crossover among these circuits happens among the professionals who stage political, religious, or entertainment events. For them, the public symbols are not generalized but particularistic; they are part of their backstage and sociable talk, not from the point of view of the adulating (or otherwise) audiences, but from the point of view of those personally knowledgeable about the everyday narratives of being performers.
28.  This will be further discussed in chapter 4.
29.  There are other group perspectives from which the World Trade Center towers were a symbol. For the attackers, the towers were doubtless a symbol of the New York skyline, and along with the Pentagon, emblems of American financial and military power around the world. From an outside, enemy perspective, symbols of a group's identity may thus be more sharply defined than they are to members of that group itself. It is notable, too, that the damaged Pentagon never became a widespread symbol of post-9/11 American solidarity, nor did the ostensibly heroic airline passengers who fought the hijackers and prevented yet another ground attack. In none of these instances was there a process of construction of symbolic status such as that which made the firefighters into emblems of American solidarity and courage.
NOTES TO PAGES 95-100
385
30.  The most personalized of these symbols was the mayor. Here the transformative power of an intense IR is striking. Prior to 9/11, Mayor Giuliani was intensely disliked by a considerable portion of the NYC population because of his policies on militant police tactics, and his political career was generally regarded as at a nadir.
31.  In other words, we are no longer a first-order actor, the social patterns of which action now become consciously visible because that is what we consciously observe. At the same time, as second-order observer, we are necessarily defocusing from ourselves as actors on this second level of intellectual action. We could, of course, go on to take a different observer's standpoint and do a sociology of thinking, an analysis of the social activity of the person in the intellectual observer mode. Which is to say, we can become intellectually and reflexively conscious of anything human beings do; but we can't make everything conscious at the same moment. For an exposition of levels of observers as locations in social networks, see Fuchs (2001).
32.  It is indeed the case that many gun owners use them for sports and hunting; yet many of the weapons possessed, such as automatic weapons and machine guns, are too powerful to be suitable for hunting. See the various lines of argument and evidence in Wright and Rossi (1994), and Cook and Ludwig (2000).
33.  Talk about guns is easiest to observe in the way gun salesmen talk to customers, bringing up such topics as what kind of weapon you would need in a dangerous situation, what weapon would be adequate to take out a threatening challenger, or an intruder in your house. The talk that typically takes place in the gun shop invokes imaginary uses of guns in dramatic situations, which are rather far from the routines of the gun cult itself. This dramatic content is a form of sales talk, although taken rather seriously by customers and perhaps salemen themselves; in effect, it is the content of the fantasy they are buying. Like buying pornography, buying a gun is chiefly buying an opportunity to fantasize.
34.  Guns in films and television shows may be regarded as focal points of vicarious rituals. The use of guns is typically a high point of the drama, whether the emotions are build up through the plot format of action-adventure or mystery / suspense, giving a strong focus of attention, and usually an implicit membership marker between those who have weapons and those who are mere bystanders. There is considerable research on the extent of exposure of weapons on TV and its effect or noneffect on violence. IR theory leads us to question whether people's ritual experience of the entertainment media leads directly to violent behavior; it may largely remain part of the round of secondary circulation of symbols, part of what people talk about, or what children play-enact in make-believe games. The question for investigation is whether and how this secondary circulation of symbols becomes articulated with the first-order gun cult; and yet further, whether participating in the gun cult leads persons to fire their guns in ordinary life, against other human beings, outside the routines of the gun cult: against or by criminals, accidentally or intentionally against family members and acqaintances, in angry disputes and in escalations of other conflicts. These "real-life" uses of guns are doubtless much more
386
NOTES TO PAGES 100-107
chaotic than the regularized rituals of the gun cult. It may be the case that the several different realms have little to do with each other.
35. In the same way, the movement for alcohol prohibition in the late-nineteenth and early-twentieth-century United States enhanced social identities and boundaries on both sides. According to Gusfield's (1963) analysis, the prohibitionist movement was an attack on the saloon as ritual gathering place of urban immigrant males, especially by native rural WASPs and upper- and middle-class females. The enactment of prohibition then fostered a wider countermovement in that drinking parties became an emblem for self-consciously modern, youthful, sexually liberated people. Symbolic markers thus go through a historical development shaped not only by their participants but by their conflicts.
Chapter 3 Emotional Energy and the Transient Emotions
1.  This has been developed most explicitly in social movement theory that has recently reached out to incorporate emotions (Jasper 1997; Goodwin, Jasper, and Polletta 2001).
2.  See, however, Lawler and Thye (1999), who propose a model of how emotions may be brought into rational exchange theory. Emotion is central to Affect Control theory, although this has usually been taken as a theory within the sociology of emotions rather than a general theory of microsociological action (but see MacKinnon 1994). See also note 4.
3.  This is now changing, as emotion research has been promoted for its application to a range of sociological questions (e.g., Barbalet 1998). For research programs in sociology of emotions, see Kemper (1990). Among the classic traditions of social science, Freudian theory is most directly concerned with emotions. But it hasn't helped us much in advancing sociology. In part this is because Freud makes emotions derivative of drives, whereas I will argue in chapter 6 that the reverse is a more plausible research program, indeed on Freud's home territory of sexuality. Insofar as Freud is a microsociologist, he is a microsociologist of early childhood family situations. My position is that we can learn more about such situations by viewing them through the lens of interaction rituals that we see in adult life, than by viewing adulthood through the lens of early childhood.
4.  In this respect IR chain theory is paralleled by Affect Control theory put forward by Heise (1979, 1987) and Smith-Lovin (1990). IR theory gives a more elaborate model of the situational process itself. Affect Control theory builds on its distinctive form of data: questionnaires that rate actors and actions on the dimension of goodness, power, and activeness, and that predict changes in these ratings when already existing (and hence already rated) actors and actions occur in new combinations. The model has been implemented by computer simulations.
5.  I might add, to reassure those sociologists who are wary of the intrusion of physiology into meaningful,  interpretive human action—of Naturwis-
NOTES TO PAGES 109-119
387
senschaft into Geisteswissenschaft—that the prime mover of human action remains on the level of social communications. Social emotions are not being reduced to physiology; to the contrary, human brain physiology is activated, and takes on the condition that it is in at any particular moment, by the flow of interaction in the IR chain. Physiology is the substratum, while the causality flows from the social interaction. The human brain is largely programmed from the outside in.
6.  A point stressed by Rodney Stark (2002) in a comparative analysis of religious rituals.
7.  Order-giving can occur in a number of different contexts, and hence individuals may have mixed experiences across their lives. Such mixtures are most likely to happen in complex modern societies (although not to everyone), and are least likely in traditional societies organized around patrimonial households that concentrate all spheres of activity in one location. As Lamont (2000) shows, modern working-class order-takers shift their evaluation criteria as they reflect on their overall position in the class structure, and thereby build up their subjective status. These complications concern long-run patterns of IR chains over time. Here we are considering the dynamics of each micro-situation of the enactment of power taken in itself, examined for its immediate effects upon situational emotions. In chapter 7,1 distinguish between deference-power (D-power)—the power to give orders in the immediate situation—and efficacy-power (E-power)—the power to make consequences happen outside the immediate present. The present discussion concerns the emotional consequences of D-power.
8.  This occurs most palpably in torture, such as practiced by prison guards, slave overseers, soldiers dealing with guerilla fighters, police in dealing with intractable arrestees, and also child bullies (Collins 1974; Montagner et al. 1988). Torture is a highly and inescapably focused ritual designed both to gain emotional dominance over the subordinated individual in the immediate situation, and to broadcast the symbolic message of group domination and subordination.
9.  See evidence summarized in Gans (1962, 229-62), as well as the descriptions of working-class ethos in Rubin (1976) and Halle (1984), and most extensively in Lamont's (2000) investigation of the way in which working-class men view the class above them.
10.  An obvious case is America at the turn of the twenty-first century, where the socially preferred tone is casualness, and sanctions are given for those who are overly formal and overly moralistic. These complexities are discussed in chapter 7.
11.  Frijda (1986, 13, 71) describes emotion as a felt but latent action tendency: a readiness for contact with the environment at the high end; at the low end, disinterest and apathy.
12.  I am leaving aside the complexities on the physiological level, where several different components of hormonal and neural systems are apparently involved. On this level in general, specific states of emotional arousal are due more to the balance between various systems rather than to the activation of some system by itself. See also Frijda (1986, 39) on both simple and complex varieties of depression. There can also be specific chemical processes associ-
388                                         NOTES TO PAGE 120
ated with depression, and these may have some genetic component, and can be treated by medication; but IR theory holds that physiological processes are not solely determined by chemistry and genes, and that some significant proportion of them occur because of the flow of successful and unsuccessful IRs in everyday life.
13. Kemper's theory has the additional complication in that he postulates anger (as well as shame) as resulting from situations in which an actor feels he / she is short-changed in status, vis-ä-vis someone else. That is, Kemper deals with the more complicated situation of comparisons between the status one thinks oneself ought to get compared to someone else, and what they actually get. I prefer to begin the explanation from a simpler and, I believe, more fundamental process: the emotions that derive from dominating or being dominated, and of being a member or a nonmember. The Kemper theory adds not only expectations from past experience, but also a moral judgment as to the propriety of the outcome compared to some valued ideal. The two theories may be congruent, in the following respects. I propose that experiences in power situations, and in status-membership situations, result in increases or decreases in emotional energy. EE itself involves expectations for future situations; but the IR mechanisms that produce EE in the first place are, so to speak, first-order mechanisms of emotional production. Emotional energy becomes an ingredient in allowing future situations to occur, and in determining their emotional outcomes. The expectations that are important in Kemper's model may be regarded as situationally specific arousals of EE. Kemper's theory seems to me to explain a second-order quality of emotions, those that arise from violation or confirmation of expectations. Both types of mechanisms may be operating in the same situation: for instance, there can be depression from nonacceptance in a status group (my hypothesis of first-order effects), and anger from one's assessment of this nonacceptance as unjust (Kemper's second-order effects).
Kemper adds further complexities, including the attribution as to the agent responsible for the experience (one's self, other persons, impersonal forces). I would suggest that these cognitions themselves are explainable by the Durk-heimian social density. Blaming oneself occurs only when there is a differentiated group structure producing categories of individual agency and responsibility; blaming impersonal forces (e.g., magic) or the violation of a taboo occurs where there is a tightly bounded and internally undifferented group. Mary Douglas (1973) refers to the former situation as high "grid" and to the latter as high "group," and provides data from anthropological comparisons for their correlation with different modes of attributing danger and responsibility (Douglas 1966). Black (1998) systematizes data to support the general pattern that conflicts within tight undifferentiated groups are quickly smothered and offenses are left unavenged; individual responsibility and punishment occur in structures of social inequality, relational distance, and heterogeneity. Thus an individual's prior experience in living within particular kinds of network structures should affect what agency he or she perceives as operative in his/ her immediate situations, and will shape specific emotions along the lines that
NOTES TO PAGES 120-127
389
Kemper proposes. As stated, Kemper's model is too closely tied to modem social conditions.
14.  Theorizing on the basis of the four primary emotions of anger, fear, happiness, and sadness, Turner (2002, 72-78) analyzes shame as a second-order emotion, blended from several primary emotions. The strongest component in shame is disappointment-sadness, combined with a lesser degree of anger at oneself, and fear about consequences to oneself. Turner suggests that pride is a blend of happiness directed at oneself, with an undertone of anger directed against others.
15.  This is why in races, running behind the pace-setter is often the strategy that ends up winning. The second-place runner feels psychologically pulled along by the leader's effort. Then at a key moment when the finish-line is in sight, she or he breaks the leader's rhythm and moves to the front, leaving the former leader locked into his/her former rhythm—a rhythm that after all must feel right since it had been shared by the followers up to that point. It is hard for the former leader suddenly to shift from a leading rhythm to following and matching the new leader's rhythm, and then to shift yet again to break the rival's rhythm. The same dynamics apply to horse races, i.e., among non-human mammals.
16.  For more detail see the appendix to this chapter. The weakness of Erick-son and Schultz's technically impressive study, and other sociolinguistic studies of this kind, is in the larger theoretical apparatus. The authors interpret their findings in terms of cultural differences, as if counselors and students misunderstand each other because they are using different paralinguistic codes of different ethnic groups. The implication is that misunderstandings can be overcome by learning the multiple cultures of tacit communication codes. This might be so in some instances, but it misses the key source of variations in solidarity in micro-interactional situations: the process of the interaction ritual itself. In IR chains, individuals build up different amounts of emotional energy, differing symbolic repertoires, and hence differing attractions and repulsions toward various kinds of conversations; and the micro-situation itself has its own dynamic principles that determine what level of rhythmic coordination it reaches. It is not to be expected that every dyad, even from the same ethnic group, would automatically produce solidarity. In short, the authors of these studies limit themselves to macro-variables on the input side; their contribution is to descriptive measures on the output side.
17.  There are further complexities in conflictual situations that I will not pursue here. Short-term dynamics of conflict initially raise EE within mutually hostile groups, sucking them more deeply into the emotions of conflict; further emotional ups and downs occur in victory, defeat, and long-term stalemate. These patterns are the subject of my forthcoming work on violent conflict.
18.  Individuals who dominate groups may deliberately provoke weaker persons on the margins of the group to become angry: an example is the game of trading insults found among youth gangs (the game at one time called "the dirty dozens"). This is a game to humiliate weak persons, who are goaded into expressing anger, but are unable to back it up by a show of physical dominance. This is playing on the underlying principle that strong persons keep
390                                     NOTES TO PAGES 127-140
their cool; when they do rise to anger, they express it in such a powerful form as to drastically penalize anyone who is its victim. There are, of course, some situations in which this kind of provocation is played mildly as a form of friendly teasing; it generates solidarity precisely insofar as it raises the level of collective effervescence, but does not push the teasing to the level of provoking anger. In this respect teasing differs from bullying, although there is a continuum where they shade into one another.
19.  See Black (1998) for evidence that in societies organized as loose social networks individuals react to affronts by avoidance.
20.  Thus crying, like anger, tends to occur in a relatively "realistic" manner: it is most often expressed in situations in which it has a chance of accomplishing its end. This analysis is confined to the kind of crying that is related to fear. Crying at moments of ceremonial triumph, or in response to a scene of personal reconciliation in a sentimental film, is a different emotional dynamic, related to intense feelings of solidarity. See Katz (1999) for a detailed analysis of the bodily rhythms and vocal inflections in situations involving crying and whining. In cases like those which Katz describes, whining is not an expression of fear but is a mode of exerting interactional control as well as manipulation of self.
21.  At least, this is the pattern where depression is socially induced and not merely a genetic / chemical condition.
22.  Legally speaking, the family car may belong to the father, but if the teenage daughter always gets to use it when she wants to, it is situationally her material resource, not his. Analogous relationships are important in the financial world, as in the case of high-EE investors using other (lower-EE) persons' money. This model of economic stratification is pursued in chapter 7. Legal property relations are situationally challenged, of course, in a range of interactions from aggressive borrowing to robbery. Black (1998) presents evidence that much property crime is viewed by the perpetrators as a form of self-help, managing their own personal obligations and grievances in an ongoing chain of struggles over possessions.
23.  I have sometimes referred to membership symbols by Bourdieu's term "cultural capital"—in part because it makes a nice symmetry between the shorthand "CC" and "EE." Under either term, these are symbolic possessions that may be invested in further interactions, and are subject to constraints of a market, including deflation in the value of the currency as it becomes more abundant (see also Lamont and Lareau 1988). The difference between the two theoretical schemes is in the emphasis given to micro-situational process or to abstract macro-structure. My use of the concept of "cultural capital" or "membership symbols" refers to all items of culture charged up by interaction rituals, which thus shift in local significance with situational processes over time.
24.  One of the problems in survey research about happiness, a concept that may be loosely related to EE, is that respondents tend not to say that they are unhappy; hence questionnaires give a series of refinements on the positive end, ranging from "very happy" to "not too happy." (Bradburn 1969). From the point of view of micro-situational analysis, the situation of being interviewed may be a positive IR, which raises the EE level of the respondent at
NOTES TO PAGES 143-150
391
that moment. Thus we would want to be able to trace self-observations of EE from one situation to the next, to capture the situational source of variation, apart from the features of making a report about oneself.
Chapter 4 Interaction Markets and Material Markets
1.  See Waller 1937; Homans 1950, 1961; Blau 1960. It was reading the latter paper by Blau that motivated me, as a young graduate student in psychology, to switch to sociology.
2.  Figure 4.1 draws attention to the detail that the initiating ingredients must rise above thresholds to set off collective processes, otherwise the IR fails to get off the ground. An additional time-dynamic underlined here is that ritual outcomes last only for a limited period of time. This fact is indicated by the dissipation sinks (drawn in the convention of flowchart modeling) on the far right side of figure 4.1 (and also drawn just below failed thresholds); these mean that the feeling of group solidarity, EE, and membership significance of symbols, fades away after the ritual is over, and eventually will disappear unless another IR is carried out. In a computer simulation of this model, we can enter a rate for the dissipation of this level over time, and observe how the level of solidarity builds up and declines depending on the strength of the flows into it and how often the whole process is repeated. For an example of a simulation, see Hanneman and Collins 1998.
3.  See Frijda 1986. This satiation is not shown in Figure 4.2 but could be indicated by further rows and columns showing diminished payoffs for extremely long periods of sustained attention. In figure 4.1, the feedback loops as shown would result in a continuous escalation of all variables; a more complicated flowchart would indicate the points at which emotional satiation occurs.
4.  One can see this in the long-term feedback loops in figure 4.1 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 assembly and focus of the group.
5.  That is to say, the IR mechanism shapes all situations of interaction. Whenever people come together, there is always some degree of mutual focus and emotional entrainment, ranging from zero to intense, and thus there will always be some effects on ritual outcomes. This process is inevitable even if there are other inputs into individuals' bodies. If an individual's EE is low because they are starving or diseased, their bodily interaction with others will still be shaped by the amount of mutual focus and emotional entrainment, albeit in this case constrained by the low physiological condition of one of the participants—which thereby may become propagated through the chains of emotions and symbolic significances among others in the network who are not physiologically affected. The IR mechanism never turns off, even as its inputs vary. Even if there are genetic influences on behavior, they must flow through IRs, hence social interaction always shapes how genetic influences are experi-
392
NOTES TO PAGES 155-163
enced in social situations. It is untenable to suppose that gene therapy or some other kind of medical intervention would automatically change people's social behavior irrespective of situational dynamics.
6.  The famous inability of General de Gaulle to get along with Churchill during the former's residence in England during the World War is not to be attributed to unusually egotistical personalities, since it is just one instance of many. Hemingway (1964, 28) reports that when attending Gertrude Stein's salon in Paris of the 1920s, one never spoke of James Joyce, who was lionized elsewhere; "it was like speaking of one famous general in the presence of another general."
7.  It is unclear whether the highest-EE individuals will interact with "upper-middle-EE" or with "lower-middle-EE" persons. For example, one pattern could be the group of ebullient "party animals," "leading spirits," etc., clustering togther with the highest energy star in their midst; another pattern could be that the "upper-middle EE" crowd makes up one cluster, while the energy star collects a stable of "lower-middle EE" persons who slavishly follow and applaud.
8.  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.
9.  More broadly, it is worth emphasizing that work situations generate their own stocks of membership symbols. These constitute the local culture of the stock broker, the financial manipulator, the industrial manager, the professional politician, and every other occupational milieu. Membership symbols are generated locally within the various realms of work. Thus Bourdieu greatly overstates the importance of "cultural capital" created and transmitted within the formal culture-producing institutions such as schools and museums, as well as that which is passed along in the family as class "habitus." As empirical evidence suggests (Lamont 1992; Kanter 1977; Dalton 1951,1959), business executives and other high-ranking persons do not owe their ability to negotiate membership with other such persons to any great extent to their "cultivation" in the formally produced cultural symbols, but rather to their use of the symbols of their immediate milieu. Financiers assemble financial coalitions not because of their knowledge of literature and opera, but above all because they talk the language of finance in a convincing way. In contrast to Bourdieu's emphasis on what might be called generalized cultural capital, individuals in elite occupations are successful because of the particularistic cultural capital or stock of symbols that circulate in their immediate network.
NOTES TO PAGES 164-170
393
10.  The relative subjective value of money earned should decline at high levels of income, according to conventional economic theory, as there is diminishing trade-off between money and the effort put into working. But where work consists in high intensity IRs that generate energy, there is no increasing taste for leisure. On the distribution of leisure and work hours, see Jacobs and Gerson 2001.
11.  I have argued above that high-EE individuals tend to shun each other. Does this contradict the evidence I am citing here that high-EE intellectuals cluster with other high-EE intellectuals? Closer examination of the temporal patterns shows no contradiction. Those intellectuals who become highly productive (manifesting high EE in their work) typically start their careers both as pupils of previously high-productive intellectuals, and often as members of a group all of whom move up into creative activity together. Once an individual makes their intellectual breakthrough into independent reputation in the social attention space, however, he or she generally breaks both with his / her teacher and with early compatriots who have now become rival successes. These patterns are documented in detail in my study of networks of philosophers (Collins 1998).
12.  From this point of view, there are two different aspects of what Grano-vetter (1973) famously called "the strength of weak ties." One is the shape of connections of that tie in the larger network: here a "weak tie" is one that ties to other people who are remote, and thus conveys information that is not locally available. Burt (1992) reformulated this kind of weak tie as a bridge tie across a structural hole in the network, in contrast to ties that are redundantly interconnected among the same group of people. The second way in which a tie can be "weak" or "strong" is in terms of the kind of IR that takes place when these persons meet: a weak tie would be a perfunctory ritual, generating little solidarity and emotional energy; whereas a strong tie exists among those persons whose encounters generate these outcomes strongly, and thus makes them friends, confidantes, valued colleagues. The two kinds of strength or weakness of ties can combine in different ways. It seems likely that the advantageous "weak ties" in Granovetter and Burt's sense (bridge ties) must be at least minimally successful as IRs, otherwise nothing would get transmitted through them; and it may be that having a strong (interactionally intense) tie to a bridge across a structural hole is what makes those ties effective. Conversely, one may have ties that are clustered in redundant, multiply interrelated groups, but the group itself may be emotionally flat and perfunctory in the symbols they pass around.
13.  For a striking example of the creation of markets as collective enterprises allying competitive organizations whose identities are inseparable from their competition, see Leifer 1995.
14.  The nearest modern equivalent to the ascetic saint would be athletes, who sometimes undergo considerable bodily pain and receive in return the loud emotional support of an admiring crowd. We are less inclined to see them as altruists, since successful modern athletes are commonly quite highly paid (either immediately or in the long run), and in addition are quite egotistical and spoiled in their off-field behavior. Monks were honored as beings apart
394                                     NOTES TO PAGES 170-177
from ordinary life because they made a lifetime commitment to asceticism; athletes make only a situationally specific commitment to undergoing bodily pain in transient situations. For an explanation of the changes in social structure that made monastic life an honored focus of attention in some traditional societies, but not in modern ones, see Collins 1998, 206-8.
15.  IR theory implies that leaders of altruistic organizations can be expected to become quite egotistical. If several leaders are energized by the adulation that they receive, there are struggles over power positions inside altruistic organizations. One way in which these are typically resolved is that ambitious, especially younger followers, once they have served their apprenticeship and learned the techniques of mobilizing a movement of this kind, split off to form their own organization. This is a typical pattern in the formation of religous cults (Stark and Bainbridge 1986). For an instance of a power struggle in a classic altruistic organization, carried out by the techniques of demonstrating one's own altruism and questioning the motives of one's opponents' demonstrations of altruism, see The Life of St. Teresa of Avila (1565/1957).
16.  Cf. Miller 1998. Contemporary shopping malls and entertainment complexes deliberately attempt to counteract this by ritualizing the shopping experience, as documented by Ritzer (1999).
17.  It is questionable that individuals handle this kind of decision by comparing ratios of numerators over denominators, since this is an abstract conception that is more cognitive than emotional. I suggest instead that the comparison is simply among differences: EE benefits minus EE losses. The decision is made by comparing the immediate situation fresh in the mind against remembered and prospective situations, in terms of the emotional intensity of the symbols by which they are brought to mind. This may indeed take place in conscious verbal thinking, or even in a conversational discussion, as well as in less-articulated emotional attractions and repulsions. In every form of decision making, the symbols representing the choices are surrounded by a halo of varying degrees of dazzle.
18.  Here the experimental nature of research on choice anomalies may make such behavior appear more irrational than it really is. In real-life situations, the costs of seeking information may be high; and unlike the neatly framed experimental alternatives, there may be potentially endless problems in arriving at a full range of relevant information. Under such conditions, it is reasonable to satisfice, in the sense of March and Simon (1958), rather than engage in extensive informational search. A similar argument is made by Esser (1993).
19.  IR theory predicts that persons avoid free-riding in proportion to the extent of their emotional ties with other persons in the situation. Hence most persons do not free-ride in experiments as much as would be predicted by purely materially interested calculation (Marwell and Ames 1979, 1980).
20.  See Blood and Wolfe 1960, 241. The situations and manners in which persons talk about money has not been much investigated by naturalistic research on conversation; it is a subject well worth further investigation. Zelizer (1994) depicts the many different sorts of moneys operating in distinctive social circuits of exchange; from the viewpoint of IR chains, these currencies are given their value by the conversational rituals in which people use them as
NOTES TO PAGES 178-186
395
topics; and these conversational networks are what constitute the solidarity and identity of those communities as economic actors.
21.  In a different theoretical context, this is what Garfinkel (1967) found in his famous experiments breaching ordinary expectations.
22.  The source both of math-aversion in the general populace, and of mathematics-identification in particular academic and professional communities, can be traced to the ritualized experiences of both groups during schooling. Mathematics training focuses heavily upon the rite de passage of solving mathematical problem sets, an activity that takes many hours of the daily lives of math and science students, and sets them apart from the social activities of most other students. Mathematical problems, formulated by teachers who identify emotionally with the elite standing of their profession, are typically designed so that students must internalize the symbolism and the problem-solving procedures of the field; mathematical school problems are set up in a graded series of hurdles of progressive difficulty, which keep up a level of emotional tension in the student attempting to pass them. Thus the activity of working on mathematical problem-sets becomes a fairly intense ritual, creating group-specific emotions, symbolic and social barriers between insiders and outsiders.
23.  It is doubtful that in many real-life social situations individuals know quantitative probabilities of outcomes, hence we may neglect risk, in the strict sense of the term.
24.  Garfinkel's ethonomethology is congruent with Simon's (1957) analysis on this point. Garfinkel (1967) enhances 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.
25.  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 satisficing model has no way in which to maximize across situations because it lacks a common denominator.
Chapter 5 Internalized Symbols and the Social Process of Thinking
1.  See Collins (1998, 858-62) for an argument as to why both solipsism and the philosophical tradition of arguing, in Descartes' manner, from "cogito ergo sum," are unrealistic simultaneously on sociological and philosophical grounds.
2.  At the turn of the twentieth century there were several schools of psychological research that amassed considerable data on introspection (see summary in Kusch 1999). Most of this research is not relevant to the sociological model, since it was concerned not with the natural flow of thoughts, but with isolated associations between words and sometimes images, often using artificial words constructed for laboratory purposes to avoid contamination by the ordi-
396                                     NOTES TO PAGES 189-190
nary flow of talk. Introspectionist psychology engineered its methods to eliminate the very social interaction context that we are concerned with here. There is also a large body of recent research in cognitive science that overlaps to a degree with the sociological issue being addressed, but that I will not attempt to review here.
3.  This sketch does not exhaust the number of methods for studying thought, even if we confined ourselves to a model that includes the social context of ongoing interaction. Other methods include computer simulations, classic psychoanalytic methods of free association, and analysis of dreams. I omit these, in part to avoid overweighing this discussion with a huge literature review; in part to keep the theoretical focus here upon the IR perspective and how we could advance this research program, rather than muddying it with other theoretical perspectives that cut in quite different directions.
4.  Needless to say, we need as large a collection of empirical instances as possible. Readers are encouraged to collect and analyze their own thought-situations as well as observations of semi-internalized and semi-externalized talk and to improve the theoretical scheme. Margaret Archer has launched a similar research program in England.
5.  Rankings as major, secondary, and minor are determined by the amount of reference to them in subsequent writings. The study includes 2,670 philosophers in China, India, Japan, ancient Greece, the Islamic world, and medieval and modern Europe.
6.  The criticism may be raised that everyone is linked to everyone else in a network. There is some evidence suggesting that every person in the USA can be linked to every other person in six links (Travers and Milgram 1969); thus it is not surprising that famous intellectuals would be linked to each other. The criticism gives an opportunity to attend to what precisely is being shown here. We are studying links among intellectuals, not among lay persons who do not transmit intellectual ideas and reputations; the fact that one philosopher in ancient Athens may have had a landlord who was two links away from the butcher who served another philosopher does not constitute a link among those philosophers. Network analysis tends to be too glib about what constitutes the content of a tie, usually taking for granted that there is some kind of homogeneity in what flows through those ties. Research which shows that arbitrarily chosen individuals in modern America can get a series of postcards through six links to someone they do not know does not indicate the existence of an effective social network; in a sense, it is merely an artifact of the research.
My other point of emphasis here is that important intellectuals are closely connected to other important intellectuals. These are for the most part one-link chains, which also tend to concatenate into longer chains so that, for example, the most important ancient Greek philosophers accumulated an average of 5.9 major and secondary thinkers within 2 links, and 12.1 within four links; for secondary philosophers the corresponding figures are 2.2 and 4.5. Studies of the transmission of rumors show that by the time a message has gone through several links, it tends to become badly distorted (Bartlett 1932); what gets transmitted through six-link chains (as in the "small world" research) are likely to be the merest banalities. Intellectual networks do not operate like this, be-
NOTES TO PAGES 193-197
397
cause they are extremely intense interactions, with great emphasis placed upon the significance of membership or nonmembership through the way that idea-symbols are used. The length of chains among differents kinds of intellectuals differs systematically; an argument that everyone is networked to everyone else is manifestly not the case here.
7.  This pertains up to the last generation that I examine in my study of networks. I cut off the analysis with the near past, since it takes several generations before the historical reputation of intellectuals stabilizes as eminent, minor, or forgotten. We do not know how to assess the creativity of thinkers who are our contemporaries or our teachers, because not enough time has passed to see what the generations downstream will do with their ideas.
8.  This can be illustrated by my experience as an undergraduate of listening to Talcott Parsons lecture at Harvard in the early 1960s. What one got from him that was useful later on was not the details of his own theoretical system, but his emphasis that the forefront of contemporary theory comes from links to the classics, especially Weber and Durkheim, taken together as an alliance. Parsons (and even more so his assistants and circle of followers) stressed the contrast between this sophisticated tradition and what was referred to as "American dust bowl empiricism." Implicitly, too, it was contrasted with symbolic interactionism, against which Parsons promoted Freud as an alternative theoretical ingredient on the micro-level, but fully coordinated with the macro theory rather than being micro-reductionist. Parsons set the starting point for my own career as a sociological theorist; this developed by adding other ingredients during my graduate studies at Berkeley, turning Parsons's version of Weber in directions promoted by contact with Marxists and historical sociologists. An additional ingredient came from contact with Goffman and even more so with the network of his students, among whom Goffman was a prime topic of conversation, including much discussion of how the rituals of deference and demeanor and the presentation of self were visible in the interactions right around us. Most of these other Goffman followers took his ideas in the direction of symbolic interaction, since Herbert Blumer was another imposing presence in the Berkeley department. Blumer constantly reminded us that he personally transmitted ideas from George Herbert Mead, and polemically contrasted the symbolic interactionist camp against the other schools of sociology. One circle of Goffman's students, regarded as rather exotic and iconoclastic, formed an opposing subsect, combining Goffman's sociology of everyday life with the ideas, still percolating in an underground, of Harold Garfinkel. It was this movement of followers that created ethnomethodology, publicizing and helping to get published Garfinkel's works (which although formulated earlier did not come out as a book until 1967), as the movement broke out into the open. I found my own niche by giving Weber and Durkheim the high theoretical importance Parsons assigned to them, but interpreting them in an alliance with Goffman and Blumer. I offer this account both as a concrete example and to show that IR chains can be tested by self-reflection.
9.  Herbert Blumer, who had been Mead's teaching assistant at Chicago in the 1930s, used to explain Mead's model of thinking during his lectures at Berkeley in 1964 in the following way: The "I" carries out a rehearsal of the
398                                     NOTES TO PAGES 204-209
action, by sending out the "me" into an imagined image of the world; when the "me" becomes depicted as encountering an obstacle, the "I" reroutes it by imagining a different way of getting to the goal. The capacity of the adult self to visualize an objective world, as well as to view oneself in it, is the viewpoint of the "Generalized Other." It is this division of the self into interacting parts that frees the adult human from the immediate pressures of the situation, allowing reflexive distancing, planning, and redefinition of the situation. It is the key to being human. See also Blumer 1969.
10.  See Borkenau's (1981) historical comparisons in "The Rise of the I-form of Speech." Latin, for example, rarely separates out "I" but includes it in the verb inflection. Japanese tends to use impersonal forms: one says "concerning this, wanted is" (kore wa hoshii desu), where an English speaker would say "I want this."
11.  As we have seen in chapter 2, the "I" as independent actor's viewpoint is the last component of the self to form, when internalized self-talk gives the child the capacity for internal self-direction and autonomy from immediate situational pressures.
12.  It appears that all subjects in Katz's research had an experience to tell; there was no driver who had not gotten angry at the bad behavior of other drivers. And that means that the drivers committing the bad behavior had to be part of the same sample; these are the same actions, seen from opposing sides.
13.  Another type of "magical" gesture documented by Katz is the insult of "giving the finger" to the offending driver, usually accompanied by a curse. Here the "black magic" is especially contagious, since the recipient who actually notices this gesture typically retaliates, with the same gesture or by a further escalation. This is ritual entrainment, being pulled not only into a common (foul) mood and common rhythm and focus of attention, but often into mirroring the very same gesture and the same formulaic curses.
14.  These taboo words were ostentatiously spoken by the younger generation at the time of the "counter-culture" movement of the 1960s and '70s, as part of a general repudiation of traditional patterns of deference and demeanor. Part of the impetus to their spread was a sense of breaking the social barrier between men and women, in which women had been conversationally segregated from "rough talk." Spoken obscenities are now widespread in those sections of society that consider themselves hip, youthful, sophisticated, and up to date. Nevertheless, the use of obscenities is situational: they are rarely written, especially in official documents, and are generally censored in newspapers; the same persons who use them in casual conversation avoid them in public speeches; schools generally prohibit and penalize their use.
15.  Katz quotes an Asian-American driver who says he is annoyed at the stereotype of Asian women as slow drivers. Caught behind a slow driver, he became enraged to find it was an Asian woman, and uttered a curse at her, using that categorical stereotype.
16.  That is, there might be another passenger in the car, but the cursing is not a communication toward that person. The presence of passengers provides a test of Katz's model that cursing arises from the driver's sense of interrupted flow and failure to be recognized as a conscious agent. Passengers are objec-
NOTES TO PAGES 210-213
399
tively as much in a situation of danger or of being impeded as the driver, but passengers rarely engage in cursing at other drivers, and instead tend to regard their own driver's behavior as irrational. It is the driver who experiences the sense of flow with the car in the rhythm of traffic, and hence the driver who has the experience of frustration that needs to be rectified by magical action.
17.  For example, when Baruch Spinoza was expelled from the Amsterdam synagogue in 1656, he was formally cursed by the congregation, who ritually stepped over him outside the doorway.
18.  For instance, watching a video of Ninotcha and disliking the performance of Melvyn Douglas, who seems quite miscast, I say aloud, "Well thank God it wasn't Gary Cooper." Subvocal thinking feels inadequate for making a strong statement; in the absence of someone to tell it to, one speaks it aloud to oneself (or writes it—whence the impulse to write letters to newspaper editors).
19.  There are some specialized exceptions here, such as the process which musicians go through in composing. But although this process seems mysterious to outsiders, who are likely to invoke folk notions of "genius" or "inspiration," it appears to be similar to the kind of internalization of techniques from professional networks, reshaping of ingredients, and reexternalization toward known audiences, that operate in the case of intellectual creativity. See Denora 1995.
20.  At least this seems to be approximaly so. The image does not last noticeably longer than the internal speaking turn; nor does it flash by so quickly that visual imagery shifts much more rapidly than verbal topics do.
21.  A marginal case here is dreaming. Dreams that are most vivid, and that comprise the main connotation of the term, take place largely in imagery, with only intermittent talking; and these voices may or may not be identified with oneself. There is, however, another, very large class of dreams, or mental activity, that goes on during sleep, which consists entirely of verbal thinking; this has been studied by waking up subjects during times of rapid eye movement (Kryger, Roth and Dement 2000). Thus even sleep-thinking consists quantitatively in a large portion of self-talk, often obsessively repetitive and disjointed. This dreaming self-talk appears generally to consist of elaborations on reverberated talk from the previous day, or anticipatory talk for topics coming up in the immediate future, and thus gives the sleeper a not very restful night, since it is too close to the contents of daytime consciousness. Visual-dominant dreams go further afield. Here, too, there seems to be a sociological component. Visual dreams are a form of thinking in concrete images: since verbal thinking is minimized, every thought is spelled out in a picture, not presented as an isolated image as if it were a picture in a book, or a sign in an alphabet, but taking up the entire visual field as if it were a world in which one is bodily present. Dream-thinking proceeds from image to image, but via the imaginary world-with-yourself-bodily-in-it, thus bringing about some strange incongruities by the standards of waking reality. This implies that Freudian efforts to make sense out of dreams as a language are misguided: dreams are clumsy and rather unsuccessful forms of thinking; they reveal some of the ingredients out of which thoughts are composed, but they do not usually translate into coherent verbal thoughts or even thoughts expressing desires.
400
NOTES TO PAGES 217-228
Dream-thinking is thus very slow, compared to the speed of verbal thought. It is at the opposite pole from Turner's model of visual thinking as lightning-fast, a way to size up a situation and take rapid action. In sleep, when the body is immobilized (and the imagery-dream usually takes place in a period of deep trance, far from readiness for motor activity), a chain of thought by means of a succession of visual images is perhaps the slowest form of thinking of all. This can be taken as indirect evidence that verbal thought, free from the exigencies of concrete imagery, is the most effective medium for wide-ranging thought chains.
22.  The process is obviously different with intellectuals who create poetry or distinctively patterned literary prose.
23.  One can illustrate this point with the vocabulary that most rapidly conveys the worldview of a theoretical position: "legitimation," "world-system," "identity politics," "textuality," or even "interaction ritual."
Chapter 6 A Theory of Sexual Interaction
1.  That is, anal penetration by the fist. This was proudly announced by a gay movement writer as "the only sexual practice invented in this century" (Rubin 1994, 95).
2.  This interpretation is congruent with evidence that persons in monogamous relations tend to have higher physical pleasure as well as emotional satisfaction than those with multiple partners (Laumann et al. 1994, 375).
3.  Theories that interpret heterosexual erotics as male dominance are accurate enough in many historical settings. Sex has indeed been a form of property; but this needs to be analyzed in relation to historical changes in negotiation of kinship alliances, marriage markets, and individual prestige relations, rather than treated as a constant. Ideologically formulated theories of male dominance lack a plausible micro-theory of interaction, substituting rather fanciful Freudian speculations, and miss the central features of erotic interactions.
4.  Zelizer (2000) notes that the borderline of prostitution is not clear-cut, and that there are a variety of sexual relationships that differ in how immediate or specific the material payment, with both greater social respectability and more diffuse exchanges on credit in more long-term relationships.
5.  Sources on prostitution: Sanchez 1997; Hoigard and Finstad 1992; Chapkis 1997; Stinchcombe 1994; Monto 2001. Customers reporting their experiences with prostitutes on worldsex.com frequently complain about being cheated by sexual come-ons. It would be illuminating to have materials on female experiences with gigolos. This is a somewhat more long-term relationship than an isolated commercial transaction, but has the reputation for callousness and exploitation on the side of the male provider of sexual services. Data on homosexual prostitutes (Kulick 1998) show that hard bargaining and exploitation of customers is a function not of gender but of the buyer-seller relationship.
NOTES TO PAGES 229-236
401
6.  In other words, the customer pays a definite sum, whereas the prostitute contributes pleasure, which is less easily measured and more subject to interpretation. Accordingly, prostitutes give no guarantee of satisfaction. Some feminist theorists (e.g., Barry 1995) stress the point that males exploit prostitutes, but this is a macro-structural argument, i.e., the claim that the very existence of prostitution is the result of a sexist society. Sticking to the micro-level, if the deal is money for sexual pleasure, then prostitutes generally exploit their customers more than vice versa. This would be the case no matter how legitimated and publicly regulated prostitution may be, such as in the case of legalized and semi-legalized prostitution in contemporary Netherlands and Germany.
7.  This is a predominant theme in reports that male customers of prostitutes post on web sites like www.worldsexguide.com.
8.  Historically, male-female couple dancing became popular at just the time, the nineteenth century, when an individualized marriage market came into existence, but with strong restrictions on nonmarital sex, and a considerable role for public opinion in choice of marriage partners, conveyed by a widely shared sense of social prestige. Ballroom dancing flourished as the micro-interaction appropriate for structural conditions favoring openly inspectable rankings of sexual popularity. Earlier forms of group dancing were not part of sexual negotiations, and often were carried out by all-male or all-female groups. Following this line of argument, the change from couple dancing with touching to non-touching dancing after the 1950s must indicate some change in the way in which sexual relations are negotiated; courtship on the dance floor apparently became no longer very important.
9.  Biological researchers have shown that the amount of sperm a man releases varies with the amount of time spent apart from his pair-bonded partner. The interpretation given is that this is an evolutionary mechanism to make his sperm win out over competition with other potential males (Baker and Bellis 1995; discussed in Thornhill and Palmer 2000, 44-45, 74). IR intensity is an alternative explanation: the greater the symbolic focus and the more intense the buildup of sexual IR, the more intense the physiological climax. This may happen by gazing at pictures of, or fantasizing about, the absent lover; similarly we may expect that the more attention to pornography, the more sperm released when excitement is finally climaxed either in copulation or masturbation. Possibly an innate biological mechanism for increasing sperm release may be activated in this way, but the process is disconnected from reproduction, and determined by the intensity of the IR mechanism.
10.  Although not necessarily immediately, since the physiology of intercourse follows climax with a falling off of sexual excitement and a refractory period. Something equivalent to a refractory period after satiation occurs in all kinds of IRs: the climax of sociable entrainment in shared laughter, or the eventually falling off of motivation to continue an entralling conversation. Without this, no IR could ever come to an end, and individuals could not detach themselves to get on with the utilitarian part of their lives. In Durkheim-ian theory, rituals are repetitive, not everlasting. This satiation point had to be diagrammed into the flow chart in chapter 4 (figure 4.1), to keep the simulation model from escalating to infinity.
402                                     NOTES TO PAGES 237-245
11.  Jewish religious tradition, which formulated the sabbath or seven-day ritual cycle, also enjoined weekly intercourse.
12.  The high level of sexual intercourse among those just establishing a partnership further supports the interpretation offered in note 9 for the high amounts of sperm released in pair-bonded intercourse. It is the excitement level that determines all aspects of the intensity of sexual arousal, and the excitement level is built up highest by the dramatic emotions that go into the early period of sexual negotiations, and that die down later with routinization of the relationship.
13.  Historically the existence of such groups has waxed and waned. Very likely the erotic prestige-setting influence of all-male groups grew to its height in the early twentieth century in Western societies, with the trend toward decline of patrimonial households and mobilization of age-cohorts in autonomous sociable settings. Their influence may well have waned in the late twentieth century, although perhaps only in the upper-middle class where males have been socialized into feminist culture. Like many other features of the sociology of sex, this awaits systematic historical ethnography.
14.  The Kama Sutra says, "The love of a woman who sees the marks of nails on the private parts of her body, even though they are old and almost worn out, becomes again fresh and new. If there be no marks of nails to remind a person of the pasages of love, then love is lessened in the same way as when no union takes place for a long time. Even when a stranger sees at a distance a young woman with the marks of nails on her breast, he is filled with love and respect for her. A man, also, who carries the marks of nails and teeth on some parts of his body, influences the mind of a woman. In short, nothing tends to increase love so much as the effects of marking with the nails, and biting" (Vatsyayana 1964,106-7). In mid-twentieth-century America, less elaborate bite marks were used for a somewhat similar purpose among young teen-agers. Malinowski (1929/1987, 281) describes the prestigiousness of such marks, and counters the notion that they are signs of dominance: "On the whole, I think that in the rough usage of passion the woman is the more active. I have seen far larger scratches and marks on men then on women.... It is a great jest in the Trobriands to look at the back of a man or a girl for the hallmarks of success in amorous life ... the kitnali marks are a favorite subject for jokes; but there is also much secret pride in their possession."
15.  I am rejecting the evolutionary biology argument that breasts are indicators of a woman's breeding and child-rearing capacity. In most cultures, historically, breasts have been used primarily to symbolize exactly that; but these are the same cultures in which breasts were not erotic. Furthermore, although in the twentieth century large breasts tended to be regarded as more erotically attractive than small breasts, very large breasts (which would be the most obvious representations of mothering) become less attractive; and lactating breasts, the best indicator of all, are not erotic at all (opinion survey evidence presented in Patzer 1985, 144—45). A related explanation attributes the erotic allure of breasts to a displacement of infantile sucking. But in that case, women should be as strongly attracted to breasts as men; indeed, more so, if Chodorow's (1978) theory of female under-separation were true. But it appears that most women seem not to be erotically much attracted to breasts, even lesbians,
NOTES TO PAGES 247-266
403
whose most frequent sexual activity is cunnilingus, and for whom erotic symbolism is predominantly genital.
16.  It appears that males have a stronger motivation to lick a woman's genitals, than women have to fellate men: 35.5 percent of males but only 16.5 percent of females regard performing oral sex on their partner as very appealing. There is also a good deal of female reflation of males but this seems to be largely at the male's initiative: 45 percent of males and 29 percent of females say they regard receiving oral sex as very appealing. The actual incidence is that 67.7 percent of females have performed oral sex during their lifetime, and 18.8 percent in their last sexual event, both of which figures are higher than their preference (Laumann et al. 1994,98-99,152). The difference, as I elaborate later can be explained by the existence of circuits of erotic conversation principally among males, which generate prestige-seeking through erotic activities.
17.  Masturbation is thus a form of self-interaction with symbols, structurally analogous to the relationship between public religious ritual and private prayer. And both are analogous to the internalized social process of thinking.
18.  On the historical emergence of this scene, see Chauncey 1994; Weeks 1977; D'Emilio 1983. In the nineteenth century, "gay" was used to refer to the heterosexual carousing arenas of prostitutes' quarters, and especially to the expensive entertainment of high-class courtesans in Paris (Griffin 2000). Only later did it acquire its present-day connotation as homosexual.
19.  On these types generally, see Collins 1986, chapters 10 and 11; 1999, chapter 6; and references therein.
20.  In the mid-1990s 35.2 percent of males and 34 percent of females reported having sex more than once a week; 8 percent of males and 7 percent of females four or more times a week. At the opposite extreme, 27.4 percent of males and 29.4 percent of females had sex a few times or not at all during the year (Laumann et al. 1994, 88).
21.  Scheff (1990) puts it more benignly: intact social bonds produce pride.
22.  To be sure, class and ethnic differences and network boundaries among youth have not disappeared, and sexual markets tend to go on within class and ethnic pools. Nevertheless, the ideal display image of the sexual elite has a strongly class and race-transcending character.
23.  In chapter 8, we will examine an ancillary ritual connected with this action "scene," the cult of cigarette smoking.
Chapter 7 Situational Stratification
1.  When sociologists incorporate these traditional concepts into their model of class hierarchy, they are being taken in by the ideology of the leisure upper-class status group, perhaps because this group is more talkative and easier to interview than the upper class that is actively making money. Thus Baltzell (1958) is much more informative about the cultural and leisure activities of the upper class than about their business activities.
2.  There are also anonymous aspects of labor and goods markets, which are the topics of classical and neoclassical economic theory. Nevertheless, as
404                                     NOTES TO PAGES 269-282
emphasized in recent economic sociology, the social structuring of markets by networks makes particular personal connections the most important aspect of entrepreneurs' lives (Smelser and Swedberg 1994). The relationship between anonymous and particularistic aspects of exchange is just beginning to be formulated.
3.  The main exception among religious groups appears to be evangelical Christians, for whom there is evidence of having a large percentage of personal friends within their congregation. Sociability is often confined to the group, and rival settings for social encounters may be avoided, such as by home schooling their children. The "New Christian Right" is one part of society that is trying to reconstitute a moral hierarchy of status groups. For this reason they are viewed with suspicion by many other Americans, who resist anything but purely situational stratification.
4.  Even at the turn of the twenty-first century, differences along this continuum are still palpable in the contrast between the compulsory casualness of American academic life (and similar upper-middle class sociable rituals) and the pockets of British ceremoniousness found in Oxford and Cambridge colleges.
5.  Subdimension (a) is an extra twist on the social density dimension described in chapter 3; here we are concerned not merely with the density of bodily copresence over time, but with the density of ritual performances in time. Subdimension (b) is referred to in chapter 3 as ritual intensity.
6.  This principle is corroborated by comparing the other end of the spectrum: Oxford and Cambridge college social rituals downplay ostentacion and personal bragging, as these are situations where status is quietly but unmistakably conveyed by the very fact of being admitted to formally-organized sociable occasions (e.g., dinner at High Table; chats in the Senior Common Room), monitored quite explicitly by gate-keeping personnel.
7.  The latter historically would show a shift from bowing and honorific address to persons who held certain categorical statuses, toward more subtle deference in the form of who gets speaking rights and control over turn-taking. For micro-situational data on the latter, see Gibson 1999; on the long-term trend, Annett and Collins 1975.
8.  News stories reported that the U.S. Congress, as well as the president, stopped their official proceedings to hear the outcome of the O. J. Simpson trial in 1995.
9.  There is precedence in cases of persons treated as religious sacred objects: for example, a medieval saint whose trances drew spectators who would poke her with knives and burning objects to marvel at her imperviousness to pain (Kleinberg 1992).
10.  This is a move along the continuum from relatively unfocused toward highly focused public interaction. At the upward extreme historically were the Chinese mandarins carried down the street and surrounded by armed guards, while members of the ordinary populace were required to avert their eyes from them by prostrating themselves to the ground.
11.  Historically, this happened in situations where bands of men made longdistance voyages or raids, often capturing women. In all of these cases, there was much emphasis on establishing fictive kinship. We see this both in Anderson's (1999) data on fictive fathers, mothers, and brothers among alliances of
NOTES TO PAGES 284-298
405
protection and support; and it was common where tribal order was broken into fluid bands of marauders (Finley 1977; Borkenau 1981; Njal's Saga; 1280/ 1960; Searle 1988).
12. Michael Mann (1986) referred to this as "off with their heads" power, and suggested that in traditional despotisms the actual reach of such power might be very limited; he termed this the difference between "intensive" and "extensive" power.
13.. For example, Francis Bacon, son and nephew of high-ranking officials in the Tudor monarchy, himself the holder of high offices and a member of the aristocracy, addressed himself with great ceremonial deference to his own patrons. The pattern of deference in patrimonial households is illustrated throughout Shakespeare's plays, Chinese novels of the Ch'ing dynasty and earlier, and indeed in virtually all of the narrative literature of the world prior to the twentieth century.
14.  The theme comes through strongly in Shakespeare's King Lear, the plot of which concerns how many armed personal retainers a lord could have around him. Stone (1967) documents that this was a struggle going on at the time Shakespeare was writing, around 1600, as the state attempted to limit the scope of private armaments to a few household guards, thereby monopolizing control for an increasingly centralized state as part of the opening phases of the "military revolution."
15.  Entertainment stars are outside the circuits of economic class and organizational power, and even the networks of categorical status group. They have large amounts of money but do not participate in the activities that constitute upper-class financial circuits. They have neither E-power nor, in the strict sense, much D-power.
16.  Situational dominance by means of noise may occur by virtue of a loud voice and raucous language; or through equipment such as boom boxes, cell phones, and car alarms. The latter two, although commericially promoted on utilitarian grounds, have their largest effects in the struggle for ephemeral situational dominance.
17.  The author has observed this, from both sides, over some 500,000 miles of highway driving. For analysis of interviews with drivers, stressing their frustration as autonomous agents unable to communicate with the others who impede them, see Katz (1999).
Chapter 8
Tobacco Ritual and Anti-Ritual: Substance Ingestion
as a History of Social Boundaries
1. For convenience in what follows I sometimes use the terms "smoking rituals" and "anti-smoking movement." In some instances this is inexact, since there are other forms of tobacco use (snuff and chewing) and opposition to them, and the anti-smoking movement is mainly focused on cigarettes, not cigars or pipes. Comparison among these various activities will figure in the argument. The broader or narrower usage should be apparent from context.
406
NOTES TO PAGES 298-304
2.  Compare the prediction that I published in 1975, in the context of discussing previous historical prohibitions of alcohol, drugs, and gambling:
The prohibition of smoking is a good candidate for manufacturing a huge deviance culture in the future. The politics of drugs in general seems likely to be central, with constant technological innovation (which has already produced, during the twentieth century, strong narcotics and psychedelics, as well as tranquilizers, amphetimines, and barbiturates). Categories of drug deviance will be the product of interactions among a number of interest groups: pharmacists and physicians with economic and status motives for monopolization; career interests of enforcement agency officials; various occupational and community groups with status interests in maintaining particular standards of demeanor; and politicians who play upon mixtures of such interests and act as brokers of pluralistic ignorance by which wide-spread consumer interests may be kept suppressed. (Collins 1975, 469).
3.  The anti-smoking movement during the 1980s began to acquire funding for state-imposed contributions from tobacco companies. At this time billboards appeared bearing such messages as "Smokers are addicts. Tobacco companies are pushers"—the implication being that the drug-enforcement campaign and its dire penalties (including life imprisonment for third-offense users, or first-offense sellers) should be carried out for tobacco-users and sellers. Others depicted smokers as killers: in one widespread advertisement, a man says "Mind if I smoke?" to which a woman replies "Mind if I die?"
4.  Sources for the historical materials that follow: Brooks 1952; Glantz 1996; Goodman 1993; Kiernan 1991; Klein 1993; Kluger 1996; Sobel 1978; Troyer and Markle 1983; Wagner 1971; Walton 2000.
5.  It is possible, however, that the drop in smoking that did occur in the years when tobacco advertising was cut back may be attributed to the lack of the stimulus of advertisements; but a careful study on this point would also have to take into account not merely the lack of advertisements but the growth in an agressive anti-smoking campaign. Even here one might doubt whether these messages would have much more effect—that is, whether negative advertising comes across any better than positive advertising. The anti-smoking campaign during this period was centrally in the news, in pronouncements of politicians, as well as on a personal level of individuals directly confronting smokers in public and personal spaces. Assuming that the face-to-face encounters have the most powerful effects, one would conclude that the drop in advertising had little effect on cutting smoking. Advertising is highly visible and thus gives the anti-smoking movement an easy target and a sense of clear victories when cigarette advertising is legally prohibited, which is just what a social movement needs to keep up its morale. But these are largely symbolic rather than substantive victories.
6.  Here I am following an argument in the sociology of addiction outlined by Darrin Weinberg in a presentation at University of Cambridge, 2000.
7.  Inserted into the IR model in chapter 2 (figure 2.1), the ingestion of nicotine, caffein, etc. becomes one of the ingredients on the left side of the model;
NOTES TO PAGES 307-319
407
that is, it is part of the transient emotional stimulus that feeds into a common mood. But this common mood has other components coming from the character of the social interaction itself—the orientation toward tranquil relaxation, carousing, sexuality, etc. Through the process of feedback intensification by rhythmic entrainment in the group, the physical feelings of nicotine, etc. take on the emotional tone of the surrounding situation. Moreover, a successful IR that progresses to higher levels of mutual arousal generates collective effervescence and thus energizes the individuals taking part. In that way, tobacco etc. become what the participants regard as genuine sources of motivational energy, although in fact the energy is tacked on from the outside by the social experience.
8.  The substitution points up the ritualism of ordinary eating. Much of the craving for food in hardship situations may be for the normalcy of regular meals, including their social character as group assembly in a break from working and other harsh duties. To be deprived of food is also to be deprived of the implied social membership in a normal society. If one ritual can be substituted for another, insofar as it brings solidarity and on that basis, construction of shared meanings, this explains how tobacco can be a substitute for food. We see the same kind of substitution in the case of drug "addicts," but also of "workaholics," especially those in high-culture activities that bring a strong subjective sense of participation in elite symbolic action. For example, both Beethoven and Newton were known for neglecting their meals while absorbed in their creative work.
9.  Smoking was also associated with prostitution in Japan, in the entertainment culture of the geisha quarters of the Tokugawa period, as we see in particularly raffish Ukiyo-e prints. From the evidence of contemporary paintings, however, it appears that at least some women in nineteenth century China smoked pipes in respectable domestic scenes. In the eighteenth and nineteenth centuries, women in the Islamic world (Turkey, Persia, north Africa) were very frequently depicted as smoking hookah water pipes, either in the company of men or by themselves (Lemaires 2001). Perhaps the greater enclosure of harem-like womens' quarters in China and the Islamic world allowed smoking in respectable privacy, whereas the exposure of upper-class European women to public sociability led to greater concern over maintaining marks of respectability.
10.                    "Coffee, which makes the politician wise, And see thro' all things with half-shut eyes."
—Alexander Pope, "The Rape of the Lock"
First published 1714, this poem contrasts the various scenes of ritual sociability of London high society in the 1710s.
11.  In Britain in 1900,4/5 of tobacco use was in pipes, only 1/8 in cigarettes, the rest in cigars (Walton 2000, 75).
12.  There are idiosyncratic exceptions: for example, the circle of "ordinary language" philosophers meeting for discussions around John Austin at Oxford in the 1950s all smoked pipes, a practice that was jocularly taken as emblematic of their intellectual movement.
13.  "Do but notice what grimaces snuff-takers make, how their whole features are convulsed, how they dip into their snuff-boxes in mea-
408                                     NOTES TO PAGES 321-333
sured rhythm, cock up their noses, compose their mouths, eyes, and all their features to a pompous dignity, and, as they perform the solemn rite of snuff-taking, they look as if they scorned the whole world, or were bent on some enterprise on which they might say, like Bouflet, "I will make the whole world tremble!"
"I have found by certain experiments that such men have the idea that, in the moment when they sniff the snuff up their noses, they are as men inspired, transformed into mighty kings and princes, or at least made royal and princely at heart" (German orig. 1720; quoted in Walton 2000, 51).
14.  Winston Churchill was known by his omnipresent, oversized (and thus ultra-expensive cigar). Much of the time he kept it in his mouth unlit (Gilbert 1988). The fact that he also did so when working alone suggests that the cigar carried a subjective sense for him, of his place in society (manly, upper-class), apart from any physiological effects.
15.  Women reporting on their smoking habits often comment that socializing with close friends are the most tempting occasions to smoke, and thus pose the greatest difficulty in giving up smoking. This is parallel to ex-drug users' tendency to relapse when exposed to social interactions that remind them of their early drug highs. These experiences represent not so much the feelings of physical ingestion but the emotional tone of the IR that was symbolized by the drug (Darrin Weinberg, personal communication).
16.  One might interpret Shakespeare's portrayal of Falstaff and Prince Hal in Henry IV, parts I and II, first staged in 1597, as an expression of this conflict. Here carousing aristocrats take part in a realm that crosses status lines into the world of low life, in pursuit of momentary fun and excitement. A real-life version of this milieu is depicted in the poems of the rakish Lord Rochester for the 1670s.
17.  These figures apparently reflect the atmosphere of all-out mobilization during World War II, when smoking was central to military solidarity rituals. After this peak, in 1973, 65 percent of British men smoked, 42% of women for the United States in 1965, 52% of men smoked, 34% of women (Walton 2000, 94, 106; Los Angeles Times March 29, 2001).
18.  Horowitz (2001) shows that the social process of assessing the threat posed by an enemy is a central dynamic in the growth or decline of hostile ethnic movements leading to deadly riots. On risk attribution to cancer generally, compare Stirling et al. 1993. For some cross-national variations in approaches to disease risk, see Nathanson 1996.
19.  There were dramatic declines in influenza and pneumonia (which accounted for 202 deaths per 100,000 population in the year 1900), and in tuberculosis (198 per 100,000). By 1956 these figures had dropped to 8 and 28 respectively. Still larger were rates of infant mortality: about 100 per 1000 live births at the turn of the century, or about 10 percent. The most common non-infant cause of death, in 1900 as in 1990, remained cardiovascular diseases; these actually rose from 345 per 100,000 in 1900, to a peak of about 510 in the 1950s, dropping to about 365 in 1990. Cancers of all kinds accounted for 64 deaths
NOTES TO PAGES 333-337
409
per 100,000 population in 1900, rising to 140 in 1950, 184 in 1980, and 202 in 1990. (Historical Statistics of the US: Series B-107, B-114-128; Statistical Abstracts, no. 114,1992). To keep this in perspective, we may translate these figures into percentages: for instance, the last figure, cancer deaths of all kinds in 1990, is 2 per 1000 people, or 0.2 percent (one-fifth of one percent of the population dies of cancer every year). For lung cancer (the kind of cancer that is tobacco related), about 0.057 percent of the population dies every year, or one twentieth of one percent.
20.  Life expectancy at birth went from 46.3 years for males and 48.3 years for females in 1900, to 72.0 years for males and 78.8 for females in 1990 (Historical Statistics, series B- 93-94; Statistical Abstracts, no. 103,1992).
21.  Death rates for 1990 from cancer (all kinds) per 100,000 population at various ages are in the first column. In the second column, these are translated into percentages of the age group who die of cancer. For comparison, the third column gives the percentages of the age group who die of anything.
	cancel	- deaths	all deaths
25 to 34 years old	12.1	0.012%	0.138%
35 to 44 years old	43.1	0.043%	0.221%
45 to 54 years old	157.2	0.157%	0.475%
55 to 64 years old	445.1	0.445%	1.204%
65 to 74 years old	852.6	0.853%	2.647%
75 to 84 years old	1338.1	1.338%	6.139%
over 85 years old	1662.3	1.662%	15.035%
Source: Statistical Abstracts,		no. 117,1992.	
After about age 55, chances of dying from cancer start becoming noticeable, although the actual percentage chances of dying from it in any particular year are still rather small (a little more than 1 percent chance for those over 75, and still under 2 percent for those over 85). But by these ages the chances of dying from something are becoming substantial: 6 percent of those of us who reach age 75 drop off every year, as will 15 percent of those who reach age 85. In other words, cancer generally kills you when something or another kills you.
22.  Walton 2000, 107. In a population of 280 million, this gives 1/5000 annual chance of dying of second-hand smoke, or 0.02 percent. Even over fifty years of adult lifetime, this adds up to a 1 percent chance. Statistically effects on this scale are not very strong. Kluger (1996; quoted in Walton 2000, 107) concluded, in regard to the studies available in the 1980s, that when the furor over second-hand smoking took off, "the data were neither abundant nor coherent—and certainly not conclusive." More recent evidence is summarized in Taylor et al. 2001; Nelson 2001.
23.  During these years, the one country in which the authorities paid attention to data on the connection between smoking and cancer was Germany (Proctor 1999). And the head of government, Hitler, was a fanatical member of healthy-lifestyle movements, and was strongly opposed to smoking. Nevertheless, even Hitler with his dictatorial powers was unable, given the widespread popularity of smoking rituals, to impose a prohibition, even in government
410
NOTES TO PAGES 338-348
offices and the armed forces. At best, officials avoided smoking in his presence. This led to some bizarre scenes: when Hitler committed suicide in his bunker in April 1945, the first sign that he was dead was that the remaining staff lit up cigarettes (Walton 2000, 93-94).
24.  American and British studies made in the 1950s give evidence that heavy smokers were "of restless, ardent, energetic personality, non-smokers steadier, more dependable, quieter." Cigarette smokers participated in sports more often and changed jobs and domiciles more frequently than non-smokers. "Cigarette smokers were more extraverted than non-smokers, while pipe smokers were the most introverted group" (Walton 2000,169-170).
25.  The Women's Christian Temperance Union and the Anti-Saloon League withheld official endorsement for the Anti-Smoking Movement to avoid multiplying political opposition, but there was considerable overlap among participants in these movements (Wagner 1997, 20).
26.  It is revealing of the social processes involved that pipe-smoking has made no significant comeback in the decades when cigarette-smoking has become widely banned; technically, it could evade the charges of the health statistics, but in the war of rituals it was on the sidelines. Cigars have made some comeback, probably because of their association with eliteness and sophistication. As a form of sociable ritual, cigars made a substitute for at least some of the ritualism of cigarettes, although their connotation as a masculine sanctuary—precisely what had made them give way to cigarettes—was now a liability in the era of gender integration.
27.  Putnam (2000) presents evidence on the decline of formally organized sociable groups, but inteprets it in terms of a general decline of community, the "bowling alone" phenomenon. Against this stands evidence for the continued presence of social networks rather than isolated individuals (Fischer 1982, 2001). I suggest that what has happened is the decline of the formal aspects of rituals that connect them to structural stratification and thus to larger communities; the rituals that remain are more private, purely situational stratification.
28.  Dunhill (1954) wrote, "[T]he world-wide practice of smoking is rapidly becoming, except for a small minority, a lost art and a limited pleasure. ... [C]hoice Havana cigars, hand-made cigarettes, and lustrous meerschaum pipes, which graced the smoking-rooms of fifty years ago, must seem almost as remote as the elaborate smoking paraphernalia which brought such excitement to Elizabethan England" (251).
Chapter 9 Individualism and Inwardness as Social Products
1. Network analysis, as we have seen, is not a micro-situational analysis, but operates on the meso-level, counting repetitive IRs among individuals at a relatively high degree of intensity. Network analysis does not, strictly speaking, deal with situational assemblies of more than two persons; its equivalent concept on the meso-level is the measure of network density of connections or redundancy of connections; this describes a tightly interconnected group of
NOTES TO PAGES 348-359
411
individuals where all or most of them have ties to each other. (This could happen even if all of them are never in the same place together at the same time, although that situation would also be a way of producing high-redundancy of ties.) Such high-redundancy networks are similar to Durkheimian mechanical solidarity in producing a high degree of group conformity.
2.  In Fuchs's (2001) terms, group symbols are treated as essences.
3.  That is, any of the other network shapes besides those with highly interconnected ties; these can include networks with sparse ties throughout, as well as networks with bridging positions over the structural holes between relatively densely connected regions.
4.  Some light is cast on the type by Chambliss's (1989) study of star athletes; those who normally win (and are confident about winning) their races enjoy practicing alone, focusing on their technique. The technique itself feels deeply pleasurable to them, no doubt because it connotes their significant place in the social world of their competitors.
5.  In this sense, socially excluded persons differ from some other kinds of introverts who are not low-EE but may derive high amounts of EE from their solitary circulation of symbols.
6.  The pattern is notable in English literature, but the same type shows up in Chinese writings: the gentleman living in the country, who spends much of his time alone in his study or garden, with his books and his painting, but who becomes lively and gregarious when friends of the same social class come to visit. See for example Wu (ca. 1750/1972).
7.  I have given a description of an upper-class male here because this personality type is virtually always described in literary sources as male, even in writings by female authors. It appears that upper-class females in these settings were usually surrounded by other women, and thus were less situation-ally introverted than some of the top-ranking men (see Girouard 1978).
8.  For an example, see Adams 1907/1931, chapter 13.
9.  When changing historical conditions made backstage privacy widely available, sociable individuals, whom we would otherwise consider to be extroverts, spend part of their time preparing for or thinking over their frontstage performances. Thus they may spend a fair amount of time in solitude, quite possibly highly focused and emotionally intense moments precisely because of their high degree of sociability. But we would not usually call them introverts in the sense of being inward-oriented, withdrawn from society. Persons in this kind of situation shade over into neurotic introverts, whose life consists in second-guessing themselves about their social relationships.
10.  Well into the nineteenth century, it was common for texts to be read aloud; thus the association of the "bookworm" with solitude is a relatively recent phenomenon.
11.  This personality style should not be confused with the overt contents of their intellectual work. The typical American postmodernist in a university literature department uses the language of reflexivity, alienation, and multi-per-spectivity, but does so in a highly standardized way, befitting his or her distance from the center of Parisian intellectual life where these concepts originated.
412
NOTES TO PAGES 359-362
12.  The "Unabomber" of the 1980s and '90s was a former mathematics student, a withdrawn technical "nerd" introvert, at the University of California, Berkeley, at the time of the radical student movement. He combined the two cultural styles into his own solitary political cult. He sent explosive devices through the mail, usually to scientists in mainstream industry or government: which is to say, he circulated his own cult expressions in the same technical network that he himself was most familiar with. It should be noted that solitary individual "terrorists" of this sort are extremely atypical of most political activists or social movement members, even in the most radical movements; as researchers on religious movements have shown, individuals who are extremely withdrawn or mentally ill make very poor converts, since they lack the network ties to aid in further growth of the movement, and they are not effective organizers (Stark and Bainbridge 1985).
13.  I use the masculine pronoun here deliberately; virtually all the cases of embittered intellectuals of this sort that I know of are men. An exception is Gertrude Stein, but she did not withdraw, but instead was the center of a thriving salon. Social conditions of gender must enter the causal pattern.
14.  We cannot lean too much on evidence of fictional characters in highly dramatic plots, but consider the complexity of Hamlet's networks: He is in the center of public view at court as the Prince. According to some kinship conventions and political supporters, he is heir to the throne, while according to other kinship conventions and political alliances he is a minor ward of his uncle, who exercizes the kinship through leviratic marriage to Hamlet's mother. As a political conspirator he has friends whom he meets convertly and enemies whom he spies upon and believes are spying upon him. From previous sojourning at German universities, he belongs to a network of students, simultaneously carousers and wits. He is having a sexual affair with a woman related to his political enemies and who is too low in social rank for him to marry. He is a patron of theatrical performances and has some experience in writing for the stage. If we take Hamlet as a real person rather than (which seems to me more likely) as a stage character constructed for the sake of the plot, he is not a full-time introvert but a situational introvert, quite capable of gregarious and high-spirited repartee when the occasion arises. Nevertheless, there are plenty of opportunities for frontstage / backstage shifts and manipulations; and his various alternative networks pull him in different directions. These network structures and Goffmanian situational encounters are sociologically adequate to motivate both Hamlet's backstage soliloquies and his indecisiveness in action.
15.  The oldest cult of individuality was the ritual focus of attention on the political chief. Such "great" individuals, however, were usually embedded in a family succession, and they received categorical rather than individual deference. In Chinese history, the emperor was usually swallowed up in a round of rituals that left him rooted to the spot, and his individual name was obliterated by his reign-name. The few outstandingly famous Chinese emperors were the usurpers who founded a new dynasty, or notorious philanderers who brought one down, thus versions of mobility in or out of high ritual position. Hegel, who was an early, groping comparative sociologist of world history, formulated the pattern that in early states, only one is free (the ruler); in modern
NOTES TO PAGES 362-365
413
societies, everyone is free. But it is in the twentieth century where political personality cults in a highly individualized sense are most prominent. These, however, are staged by using the techiques of modern mass media for the reproduction and wide dissemination of symbols: ubiquitous pictures of Lenin, Stalin, or Mao, and similarly in promoting the cult of other political dictators and leaders. Individuality does indeed spread in modern societies; at the same time, the means for broadcasting a superficial image of hyper-significant individuals also grows.
16.  Richard Burton, an Oxford fellow and vicar who wrote The Anatomy of Melancholy between 1610 and 1640, described the scholar's life as prone to miseries and strange wanderings of fancy. But this is not the modern concept of the alienated intellectual or even an introverted personality type, since Burton's discussion of melancholy is devoted chiefly to the woes and fancies of love and jealousy (mainly culled from literary sources), to which the scholar's life was related insofar as it was practiced by celibate clerics. Among the causes of melancholy, Burton considers various misfortunes of life such as poverty, imprisonment, and thwarted ambition, along with an encyclopedic list covering supernatural and astrological forces, food, climate, diseases, and the theory of physical humors. Burton's view of melancholy emphasizes its strange fantasies, among which he includes all forms of "excess," including religious heresy, magic, superstition, and even (showing his local political bias) Catholic ritual. The category of melancholy was embedded in late medieval scholasticism and humanism, at considerable distance from the modern conception of the introverted individual.
17.  On the shift to the musical market and the concomitant construction of the cult of musical genius, see Denora 1995. On relatively more commercial and more autonomously self-oriented sectors of fields of cultural production, see Bourdieu 1993.
18.  Another cause of modern cult of the individual was the growth of bureaucratic organization, displacing the familistic connections and personal subservience of the patrimonial household. Bureaucracies are organizations of positions defined by formal rules and regulations. Individuals occupy those positions only temporarily, and move through them by accumulating a dossier or resumé of formal records; and these records are kept on them as individuals, not as members of families or other groups. Thus individuality as a category-system is built into the procedures of modern organizations. The process of education, within which modern people are all caught as the result of a lengthening process of credential inflation, can best be viewed as the accumulation of individual records, which constitutes the official presentation of the modern self as career, whatever the backstage realities are that went into making those records. Similarly with another large structural source of modern individualism, the legal conception of political rights. Movements strugging for modern democracy have pushed toward fuller participation in the state. The slogan "one man, one vote" itself had to be expanded, through a redifinition of political individuality, to include not just "man" as head of household (the early-nineteenth-century liberal conception of the independent property-owner as unit of society) but every one regardless of gender or condition of dependence;
414                                     NOTES TO PAGES 365-367
this redefinition extended to lowering the voting age to accommodate some slices of the population that were previously considered dependent children. There is a ritualistic aspect of the "one person, one vote" slogan; societies with the widest democratic ethos and the most emphasis on individuality also tend to be ones in which considerable portions of the population do not bother to vote in most elections. The concept of voting is a political symbol for the democratic era, more than a political reality.
19.  See the Oxford English Dictionary. "Extroversion" and "extrovert" have a similar history running from religion to psychology: in 1656 we find, "Extroversion ... in Mystical Divinity ... a scattering or distracting one's thoughts upon exterior objects"; in 1788, "The turning of the eye of the Mind from Him [Christ] to outward things the Mystics call Extroverson." There was an overlapping period of usage in early modern science in which the terms had a purely physical sense: in chemical texts from 1670 to 1750 "extraversion" meant the outward manifestation of a chemical reaction, while in physiology as late as the 1880s "introverted" meant an organ turned in on itself, as "introverted toes."
20.  This would apply even to forms of meditation that aimed to concentrate consciousness on emptying out the contents of the mind to experience what Buddhists and Hindus called enlightenment, and what Christians and Muslims regard as a vision of God. Buddhist doctrine was explicitly aware that the contents of thought, so-called "name and form," are part of human social discourse, and meditative practices were regarded as devices for getting beyond such attachments to the world of ordinary experience. But the religious condition aimed at Nirvana, or Shunyata ("Emptiness"), are collective symbols too, sacred objects of the Buddhist community. This is one more illustration that not just things and images but any object of collectively directed attention, including actions and experiences, can become a Durkheimian sacred object. For details on varieties of mystical religious practices and their social organization, see Collins 1998, 195-208, 290-98, 964-65, and references therein.
21.  This is poignantly illustrated by the anthropologist Victor Turner (1967). He describes walking on an isolated pathway in the last days before leaving his tribal field site; the local witch doctor fell into step with him, and without expressing as much overtly, gave Turner the feeing that he, the witch doctor, was saying goodbye to his counterpart, the nearest thing to a lonely intellectual that a tribal society had.
22.  I say "perhaps," because there is little systematic historical or contemporary data on the situations in which people pray.
23.  Weber famously explained the influence of Protestantism on modern individualism; his anti-Catholic bias kept him from appreciating the extent to which the innovations of Counter-Reformation Catholicism contributed to the modern psychological orientation (cf. O'Malley 1993).
24.  We have seen that the differentiation of religious specialists in ancient civilizations created separate enclaves that were in effect additional regions of mechanical solidarity, although they included more moments of concentration on inward experience.
NOTES TO PAGES 367-371
415
25.  Thus arise two opposing kinds of boredom: being bored of being alone, and bored with other people. Kierkegaard regarded boredom as a distinctively modern emotion.
26.  A premodern analogy for the latter would have been the mass production of crucifixes or holy relics; analogies for the former would be a massive expansion in the means of taking part in ceremonies which gave crucifixes their emotional meaning, or a vastly increased capacity for people to go on pilgrimages to the sites where holy relics were displayed.
27.  Garfinkel (1967), who shared Goffman's emphasis upon the analytical nature of microsociological observations, was emphatic in rejecting any iron-icizing intentions in his ethnomethodology.
28.  This sketch of the "Goffmanian revolution" is an example of what I mean by "micro-history." The realm of micro-interaction has a history; not merely in the concrete sense of a descriptive history of changes in manners, but an analytical history of the conditions for micro-situational interaction and their consequences.