For several decades social scientists have explored the role of emotions in society. The relationship between the social and affective processes, it is now recognized, is a two-way street: society shapes its members’ affective life which, in turn, affects society’s processes and structures (Collins 1990; Davitz 1969; Elias 1982; Franks 19911; Franks and Smith 1999; Geertz 1959; Goffman 1972; Harre 1986; Hochschield 1975, 1983, 1998; Kemper 1978; Lutz 1988; Scheff 1990, 1994; Scheff and Retzinger 1991). Democracy pays a heavy price when its members are emotionally illiterate and civic virtue is in short supply. Societies where emotional littering is rampant show high levels of conflict and violence.
This paper discusses the Emotion Template Matrix Analysis (ETMA) – a theory and methodology designed for studying the affective dimensions of human agency in its socio-historical setting. The paper begins with the existing approaches to emotional intelligence and its impact on democratic polity, after which it introduces the ETMA theory and the ETM survey – a self-assessment tool that generates an Agency scan (A-scan) illuminating the emotional wellness of individuals and groups. Preliminary findings of the ETM survey are presented next. The concluding section discusses prospects for future research based on ETMA theory and possible applications of the A-scan.
Emotional Intelligence and Emotionally Intelligent Democracy
Cognitive abilities are known to predict success in life. As measured by Intelligence Quotient (IQ), strong analytical skills improve our chances to excel in school, do well at work, and achieve recognition in society. According to Richard Herrnstein and Charles Murray (1994), this connection is so strong that the entire social structure can be explained by differences in IQ levels found in the economic classes and social strata comprising our society. And since cognitive abilities are largely inherited and immune to change, the argument goes, social inequality is inevitable, for it merely reflects the cognitive inequality revealed by IQ scores.
This argument has been challenged on several fronts. As critics point out, even if innate factors constrain individual achievements, environment plays a key role in determining whether the individual reaches his or her full intellectual potential (Turkheimer, 2003; Kilp, 2006). Research also suggests that IQ-based aptitude is not the only form of intelligence that counts – other abilities play a significant and independent role in predicting success in life (Gardner, 1993). Why else would people with an IQ of 160 work for people with an IQ of 100, asks Howard Gardner, author of the theory of multiple intelligences? In addition to logical-mathematical intelligence captured by IQ tests, Gardner identifies several other intelligences, including linguistic intelligence, spatial intelligence, musical intelligence, bodily-kinesthetic intelligence, intra-personal intelligence, and inter-personal intelligence. Different-type intelligences appear to be associated with separate parts of the brain, and they may continue to operate adequately even when portions of the brain responsible for other aptitudes are damaged (Gardner 1993).
Since Gardner proposed his list, scholars and business leaders have given close attention to the last two intelligences. Reuven Bar-On (1997) calls the phenomenon under study Emotional Quotient (EQ), Peter Salovey and Jack Mayer (1990) define it as Emotional Intelligence (EI), and Daniel Goleman (1995) and his associates refer to the affective skills in question as Emotional Intelligence Competencies (EIC). While all these authors associate emotional intelligence with the ability to understand and manage emotions in oneself and others, they disagree on what exactly the concept means and how it should be measured.
Emotional intelligence can be defined as (a) aptitude, (b) competence, or (c) wellness. Those who see EI as a generalized ability to label and analyze emotions, investigate this form of intelligence through a series of objective tests. An example of this approach is the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT), http://www.eiconsortium.org/measures/msceit.htm, featuring pictures, music tunes, and narrative stories that a person is asked to analyze or complete, with the results interpreted as a demonstrated aptitude for handling emotions. Those working in this tradition treat emotional intelligence as a cognitive skill or a species of general intelligence that has affective life as its primary object.
Another approach focuses on EI as a competence or practical skill that reveals not just an aptitude which might be undeveloped or misapplied (think about a con artist or a cult leader), but a demonstrated proficiency in handling emotions intelligently and constructively in real life. Daniel Goleman and Richard Boyatzis (Goleman, Boyatzis, and McKee, 2002; Goleman, 1998) work from this perspective. Their Multifactor Emotional Intelligence Scale (MEIS), http://www.eiconsortium.org/research/ei_issues_and_common_misunderstandings_caruso_comment.htm, gauges a person’s EI level through a 360 degree assessment technique that calls on a person to judge one’s ability to understand and manage emotions, and then invites his or her supervisors, subordinates, and colleagues to evaluate this person’s skills. The individual is assessed on several competencies, including the ability to recognize feelings, judge one’s strengths and weaknesses, trust one’s ability to succeed, manage disruptive impulses, speak one’s mind, act on convictions, seize the opportunity, grasp other people’s moods and agendas, understand one’s place in the hierarchy, boost other people’s performances, handle difficult people with grace, inspire a group and lead team efforts.
A third approach treats EI as emotional wellness that manifests itself in our capacity to manage stress, in the more or less balanced way we integrate our abilities to cope effectively with life’s problems. Reuven Bar-On takes this path, postulating that EQ abilities and skills are rooted in the individual’s personality and reflect relatively stable patterns of affective life (Bar-On, 1997; Bar-On and Parker, 2000; Bar-On and Handley, 2003). The Emotional Quotient Inventory (EQ-i), http://www.eiconsortium.org/measures/eqi.htm, is designed to capture distinct emotional traits and personality types. Using a self-assessment method, EQ-i invites respondents to rate their ability to maintain a relationship, sustain intimacy, show affection, control a disruptive impulse, work in a group, brave challenges, manage stress, enjoy company, and sustain an optimistic outlook on life.
The relative contribution that IQ and EI make to professional success is still in dispute, as is the nature of emotional intelligence (Allen and Murphy, 2006; Barchard 2003; Barchard and Hakstian 2004; Barchard and Russell 2004), but the consensus is growing that EI is a real phenomenon, that cognitive skills alone do not explain success and failure in life, that both cognitive and affective skills contribute to one’s performance, a company’s success, or society’s wellbeing. IQ scores tell us something important about a person’s ability to secure a degree in rocket science or survive medical board exams, but once you enter a profession, cognitive scores loose a good deal of their power to predict how much you will excel on the job over the long run. Studies conducted by private companies and scholarly institutions suggest that a successful workplace is emotionally intelligent, that star performers demonstrate remarkable people skills, that employees scoring high on EI tests are the ones who are likely to show superior results and find themselves on the fast track to the top (Bar-On and Handley, 2003; Bocchino, 1999; Elias and Arnold, 2006; Goleman, 1998, 2006; Urch Druskat, Mount, and Sala 2005; Weisinger, 2000).
Emotional Intelligence plays an even greater role outside work. Success in marital life, achievement in a community, an overall sense of wellbeing are inseparable from the ability to identify, label, analyze, control, and deploy creatively our emotional resources. An intellect unleavened by noble affect is as likely to go astray in our world of immense complexity as emotions bereft of reason and unenlightened by analytical intelligence. An emotionally intelligent person is someone who is aware of one’s conflicting impulses, understands how our moods affect other people, possesses a vocabulary sophisticated enough to articulate complex feelings, shows some stoic resolve under adverse conditions, seeks to ameliorate the community’s emotional climate, and communicates ambivalent sentiments in a manner that acknowledges their critical edge while protecting others from their abrasive effect and destructive consequences.
Emotional intelligence is important not only for personal growth and wellbeing. It is central to building an emotionally intelligent democracy, a civic society grounded in the premise that “intelligence is emotional just as emotions are intelligent” (Shalin, 1992:256; 1993). Free speech, multi-party politics, constitutional checks and balances are necessary but not sufficient conditions for a viable democracy. For democracy is also an emotion or “experience,” as John Dewey (1954) used to say. It thrives in an emotional culture which promotes trust, tolerance, prudence, compassion, humor, and it wilts when overexposed to suspicion, hatred, vanity, cruelty, and sarcasm (see the mission statement of the UNLV Center for Democratic Culture: http://www.unlv.edu/centers/cdclv/mission/mission.html). No quantum of hatred we impart to the world disappears without a trace, nor does the quantum of loving kindness. Violence is contagious, its outbursts compound until they result in explosions like the one the world witnessed in New York City on Sept. 11, 2001. Emotional energy can also stock an affective synergy that facilitates political breakthroughs, as we saw in the Civil Rights era. Emotional sanity is as central to democracy as discursive political rationality (Shalin, 2004, 2005). Democracies whose citizens are emotionally illiterate are given to emotional littering, emotional littering breeds emotional violence, and emotional violence lurks behind every Columbine High massacre. To guard public discourse against political as well as emotional distortions, we need to understand the affective dimensions of human agency.
E-motion Template Matrix Analysis
Emotion Template Matrix Analysis builds on ideas about the key role emotions play in our life, yet it derives its inspiration from a philosophical line of inquiry that goes back to American pragmatism and explores emotions from a sociological rather than a psychological perspective. Psychologists start with the premise that our feelings, actions, and thoughts reflect relatively stable, predictable personality patterns which persist over time and manifest themselves across situations. Psychological testing tends to privilege tools that reveal enduring traits and discriminate against personal qualities which attest to the volatility of our actions and sentiments. ETMA, by contrast, considers such indeterminacy to be a normal reflection of conflicting social pressures. It treats human beings as nonclassically propertied objects akin to particles in micro physics: When we don’t look at a particle, it is everywhere at once, it is a bundle of probabilities that requires a measurement event to materialize as a particle with a definite mass, position, momentum, and other properties. In a similar fashion, our affect continuously and subconsciously scans the world for saliency; it generates conflicting feelings, it is pulled in different directions at once, and it takes a specific occasion – a personal encounter, a request to fill a survey, a need to take a public stance – for a human agent to adopt a specific emotional attitude called for in a given culture. Predictable though such an attitude might be, it is only a matter of probability that a person will show this or that affective stance in any given situation.
The ETMA methodology does not contest the value of psychological approaches as much as it seeks to complement the paradigm by performing a figure-background reversal. Whereas psychologists start with the constancy observed in overt conduct and treat out-of-character acting as deviation from a personality pattern, pragmatist sociologists start with the individual’s contradictory performances and cast the apparent cross-situational consistency as an emergent phenomenon occasioned by the situation as well as by the agent’s multiple agendas. This is how Michell Montaigne expressed it some four hundred years ago:
. . . [A]ny one who turns his prime attention on to himself will hardly ever find himself in the same state twice. . . . Every sort of contradiction can be found in me, depending on some twist or attribute: timid, insolent; chaste, lecherous; talkative, taciturn; tough, sickly; clever, dull; brooding, affable; lying, truthful; learned, ignorant; generous, miserly and then prodigal – I can see something of all that in myself, depending on how I gyrate; and anyone who studies himself attentively finds in himself and in his very judgment this whirring about and this discordancy. There is nothing I can say about myself as a whole simply and completely, without intermingling and admixture. The most universal article of my own Logic is DISTINGUO (Montaigne 1987:377).
We can find a kindred sentiment in this profile of George Steinbrenner, owner of the Yankees baseball team, recently published in the New York Times: “As always, there are at least two distinct sides to him: brash bully and charitable gentleman. . . . He is a loyal friend who turns distrustful. He’s the calm and the storm. Kindness turns to cruelty. He is just as apt to tear apart as blow up” (Macur, 2004). Rather than seeing such inconsistency as an aberration or abnormal volatility, ETMA considers such paradoxical conduct to be a norm and focuses on the patterns of uncertainty and structures of indeterminacy endemic to human existence in the social world.
ETMA offers a broad gauge of human agency as it finds itself in the world with a particular body to inhabit, social status to enjoy, opportunities to take advantage of, and disabilities to contend with. Affect is understood here as a body index of arousal – affects are the neuro-somatic indicia of arousal an agent experiences in a situation that implicates the agent’s well-being, calls for an appraisal, and moves one to take a stance. Emotion is an affect filtered through the symbolic grid supplied by culture – emotions are affects aroused by a situation, processed through a rhetorical frame, and disposing the agent to react in a certain fashion. E-motions (emos) designate affectively charged motions, actions, and situations that signal how the agent is likely to feel/act under the circumstances (e.g., alienated, playful, stoic, subservient). Human agency is a somatically-grounded, emotionally-laden, self-referentially guided, culturally informed, and structurally-constrained capacity for action. Given that affect is an unlabeled emotion and emotion is a recognized affect, we can define the unconscious as a mislabeled affect and a misrecognized emotion.
ETMA operates with a chart that functions as a kind of a periodic table of emotions (see Table 16 in the appendix or visit CDC web site, http://www.unlv.edu/centers/cdclv/programs/etm_chart.pdf). The ETM chart is divided into four quadrants representing primary color emotions – Joy, Anger, Fear, and Serenity, which happen to mirror traditional temperaments – sanguinic, choleric, melancholic, and phlegmatic. The term “Serenity” is used here as a better known equivalent of “Ataraxy” – a notion favored by stoic philosophers who used it to stress the importance of tranquility, self-mastery, and ethical conduct. The ETM chart is segmented in several ways. Its coordinates point to the intensity/valence of affect and the strength and the type of agency. The upper half of the chart harbors positive emos that gain in affect strength as we move upward. The lower half is the region of negative emos that intensify as we move downward. The left side denotes the outwardly directed agency, which becomes steadily more proactive as we move leftward, and the right side points to the inwardly directed agency, which gains momentum and becomes more self-absorbed as we move to the right. The intersection of the coordinates signifies a zero point where affect is neutralized, the outwardly directed agency turns inward, and vice versa.
ETM quadrant contains 5 columns, each one comprising emos of roughly the same affect strength. There are 10 templates in a column and 50 templates in a quadrant, combining for a total of 200 emo labels in the chart. A template is identified by three digits, the first one indicating its quadrant or a family of emotions – 1 (Joy), 2 (Anger), 3 (Fear), and 4 (Serenity). The second digit points to the emo’s intensity – 1 (weakest emo) through 5 (the strongest emo). The third digit identifies one of the 10 agency dimensions to which a given template belongs, (see next section for explanations). Every template has counterparts in other quadrants mirroring its affective qualities. Thus, the 1.5.1 joyful emo exhilarated is echoed in the 2.5.1 choleric emo infuriated, the 3.5.1 fearful emo terrified, and the 4.5.1 serene emo template nirvanic. There is a total of 50 matched e-motion sets, with four templates in each set.
The 200 emo templates assembled for this study and organized in the ETM chart seek to render explicit the affective structure of human agency in Occidental culture. We should stress that tying an ETM chart quadrant to a particular emotion family (Joy, Serenity, Anger, Fear) or valence (positivity/negativity) is a matter of interpretation. The emos cynical, cruel, vindictive and devious have a negative connotation in our culture, but it is not obvious that they can be profitably subsumed under the heading “Anger.” The emos magnanimous, confident, humorous, or sincere may imply a positive affect, but does it make sense to lump them together as joyful affective states? The same questions arise with regard to the emos guilt-ridden, impotent, solicitous, and oppressed which are assigned in the ETM chart to the Fear quadrant, and the emos stoic, vigilant, didactic, and paternal assigned to the Serenity family of e-motions. To answer these questions, we must look at the affective halo surrounding the emos in each quadrant, check the survey wording corresponding to a given template, then confer with the agents residing in a given population/culture to ascertain whether the proposed grid mapping the affective structure of agency has any validity.
It might be prudent to identify ETM chart quadrants solely by numbers, and for certain purposes that might be a prudent tack, but such a procedure is inconsistent with the constructed nature of emotions. There are no emotions, as the term is defined above, outside a pre-interpreted symbolic grid through which affect is processed. Affect is a thing-in-itself that turns up as a thing-for-us – an intersubjectively valid emotion – only after it has been subsumed under a category supplied by a given culture. Different cultures are known to use disparate terms designating similar affective phenomena and apply the same term to diverse affective situations. For instance, Japanese treated depression as a sign of sensitivity or artistic melancholy before Prozac was introduced to local consumers, and such labels are certain to feed back to the population where such verbal emotion constructs have currency (Blazer 2005; Schulz 2004). Social scientists studying emotions must begin with the words/labels pre-existing in a given language community rather than impose on the affective datum a neutral scientific label. But researchers need not, and probably should not, stop there. The empirical consensus among agents inhabiting a given cultural niche is not a precondition for adopting a particular conceptual framework. This is especially true about the multicultural societies where disparate groups, sub-cultures, and ethno-racial communities may differ in their sensitivity to certain segments of the full emo range. The high intensity emos gathered in the Serenity quadrant of the ETM chart are likely to be better known to the respondents with roots in, or a significant exposure to, Eastern societies, even though a more recent “New Age” interest in meditation has rendered Serenity emo templates more readily discernable throughout the West. By the same token, individuals steeped in religious lore will be more sensitive to the spirituality-tinged emos than respondents unschooled and unversed in the sacred. Disparate verbal skills may also limit the respondent’s ability to understand a particular emo label and the survey wording that goes with it. The conceptual structure articulated in the ETM chart is based on certain realities observed in our culture, but it also transcends the empirical reality it purports to describe. Validating such a construct requires an expert evaluation by competent agents representing a given culture. Comparing the results of A-scan with the findings obtained through alterative methods could also lend credence to our efforts to reconstruct the affective dimensions of human agency.
Agency Dimensions and Agency Scan
E-motion terms in the ETM chart are calibrated to express a particular dimension in which human agency is appraised, with the number alongside a template indicating the domain in question. The following 10 dimensions map human agency as it is apprehended by the agent or an outside observer:
1. Vitality/Mood – felt quality of life, emotional tone, the sense of well-being
2. Energy/Arousal – agency mobilization, affect strength, proactive stance
3. Self/Confidence – confidence, self-esteem, self-mastery, temporal orientation
4. Charity/Others – generosity, gratefulness, fair play, willingness to cooperate
5. Recreation/Romance – playfulness, romance, sublimation, competitiveness
6. Wit/Discursivity – discursive prowess, wordplay, edifying discourse, dissimulation
7. Polity/Society – civic virtue, liberality/conservatism, communitarian sentiments
8. Community/Civility – manners, ethical bearings, community spirit
9. Sacred/Infernal – blessedness, spirituality, superstition, ominous feelings
10. Power/Status – authority, leadership, public esteem, subordinate-superordinate status
The proposed conceptual blueprint for the affective agency structure is admittedly ad hoc. It grew out of the efforts to find an order in affectively-laden terms gleaned from the fourth edition of Roget’s International Thesaurus, and it is influenced by “aesthetic” considerations as well as by a certain advantage of doing calculations based on decimal metrics. There is no particular reason why there should be 10 agency dimensions rather than 6, 8, 12, or 16. Indeed, an earlier version of the Emotion Template Matrix listed somewhat different dimensions and emo sequences (Shalin 2001), and it is possible that the present conceptual framework will require revisions in the future. Whatever its final shape, the Emotion Template Matrix should retain the scalability features built into the present model.
The ETMA theory postulates that the affective dynamics encoded in each agency dimension follows specific logic. You can see that when you scan a particular agency dimension along the identically numbered templates in the ETM chart. Moving from one quadrant to the next, you will discern how the sense of agency changes from one column to another.
Take agency dimension #1, Vitality/Mood. If you move diagonally along this dimension from the 1 st quadrant (Joy) to the 3rd one (Fear), you will encounter this sequence of emo templates: “exhilarated” (1.5.1), enthusiastic (1.4.1), buoyant (1.3.1), affable (1.2.1), content (1.1.1) – these are positive affect emos assembled in descending order of intensity and connoting a joyful, positive, contented mood. As we slide into the Fear quadrant, we encounter the following emo sequence: blue (3.1.1), sad (3.2.1), anxious (3.3.1), anguished (3.4.1), terrified (3.5.1) – these are negative affect templates capturing an increasingly fearful, moribund agency. Now, scanning the agency in the same Vitality/Mood dimension along the Anger-Serenity diagonal, we spot the following progression: infuriated (2.5.1), enraged (2.4.1), frustrated (2.3.1), annoyed (2.4.1), and bothered (2.1.1). Continuing into the Serenity quadrant, we spot templates calm (4.1.1), cool (4.2.1), serene (4.3.1), high (4.4.1), and nirvanic (4.5.1). This progression tracks the mood swing from the extreme anger through irritable affective states to reserved watchfulness and, ultimately, to meditative and trance-like tranquility.
The series of templates for the Romance/Recreation agency dimension (the 5th position in the column) looks this way: chivalrous (1.5.5), romantic (1.4.5), playful (1.3.5), game (1.2.5) in the mood (1.1.5); and now crossing into the Fear quadrant: blasé (3.1.5), alarmed (3.2.5), guilt-ridden (3.3.5), assaulted (3.4.5), and masochistic (3.5.5). This progression describes the agency movement from deep play and a jovial engagement and toward the feeling of being put upon, violated, coming under assault. Moving diagonally from Anger to Serenity, we have the following emo set: sadistic (2.5.5), cruel (2.4.5), callous (2.3.5), rough (2.2.5), crossed (2.5.5); and then poised (4.1.5), vigilant (4.2.5), observant (4.3.5), enraptured (4.4.5), and sublimated (4.5.5). As we become increasingly serene and stoic, the feelings associated with the aggressive and rough play are gradually sublimated into discursively-mediated, ritualized states marked by restrained, controlled desires.
Agency dimension #4, Wit/Discursivity, describes the affect dynamics associated with the agent’s use of verbal media. The following progression comes to the fore as we trace the discursive agency movement from Joy to Fear: brilliant (1.5.6), dazzling (1.4.6), witty (1.3.6), humorous (1.2.6), sincere (1.1.6), and then, insincere (3.1.6), solicitous (3.2.6), obsequious (3.3.6), travestied (3.4.6), and unauthentic (3.5.6). We can see how a good-natured, assertive verbal display terminates in a minimally joyful but still positive emo sincere before it is replaced with a defensive linguistic posture characteristic of agents anxious to please, ready to deceive, susceptible to dissimulation. The Anger-Serenity diagonal features the ten emo sequence related to affectively charged language production: sardonic (2.5.6), sarcastic (2.4.6), acerbic (2.3.6), cynical (2.2.6), testy (2.1.6), and now sliding into the Serenity quadrant – earnest (4.1.6), didactic (4.2.6), exhortatory (4.3.6), dazzled (4.4.6), and luminous (4.5.6). The offensive, negatively-tinged discursivity gradually gives here way to the didactic, moralistic, and spiritually tinged stance attributed to ataraxic agents.
Other agency dimensions in the ETM chart reflect the analogously framed affect dynamics discernable in the Occidental culture.
While the ETMA model is not derived from a proven theory and requires empirical validation, the proposed A-scan is not entirely arbitrary. The periodic structure of the ETM chart links symmetrically situated and semantically intertwined emo pairs and quadruplets which exhibit patterns found in everyday life. This can be seen in such matching emo pairs as manic-depressive, sadistic-masochistic, optimistic-pessimistic, proud-self-loathing, infuriated-terrified, dominant-subordinate – these are affective polarities known to social scientists studying the human psyche. Other emo pairs and quadruplets are less obviously linked in the public mind, but the patterns they form deserve further investigation.
The ETM survey is a self-assessment tool registering perceived moods, and self-assessment does not necessarily predict behavior. Yet, the relationship between emotion and action is a two way street. The persistent gap between the way we feel and the way we act tends to lower our emotional wellness, especially when the emotions in questions grow to be persistently negative. ETMA is based on the premise that emotional wellness is directly related to positive affect, but emotional intelligence also presupposes the ability to recognize, properly label, and own up to negative affect emotions. We can expect a strong attribution effect in the way emo templates are applied in real life situations: We tend to favor positive and shun negative affect labels when describing our own dispositions while using a full range of emos, including the negative ones, appraising other people’s affective states. Medieval crusaders spurred by the church to recapture the Holy Grail were likely to experience their agency as valiant, audacious, blissful, and virtuous, while their victims were likely to label their attackers as savage, sadistic, vicious, and devilish. It is one thing to “lecture, punish, or arrest” and quite another to be “lectured, punished, or arrested.” In the Occidental tradition, the self-activated and self-propelled agency is a condition of joy and happiness, while an action originating outside and forcing the agent to carry someone else’s will is likely to be experienced as suffering and misfortune.
One should bear in mind, also, that cultures may be biased against certain emos and exercise caution using the term “positive/negative affect” in an evaluative sense. A joyful disposition is welcome, but the excess of cheerfulness may signal an emotional obtuseness, a failure to recognize suffering. Someone who is never bothered or upset in the post 9/11 world is either oblivious to suffering or self-reppressed. Serene attitudes and spiritual yearnings are ennobling, except when they morph into sanctimony and conceal repressed desires. A priest abusing his charges will have his piety suspect. For all its destructive potential, anger may be instrumental in mobilizing agency for a righteous struggle. A teacher unable to show indignation to misbehaving students might be unfit for the job. And melancholy, though poisonous in its extreme form, is known to correlate with compassion and creativity. Shouldering the weight of the world may crush the artist but it can also produce poignant stories or beautiful tunes in the process. What matters ultimately is how perceptive we are about our moods, subtle in labeling our affects, honest in communicating our feelings, inventive in meeting other people’s affective needs, and creative in bringing about the world that is emotionally sane.
ETMA postulates that no agent falls squarely into one quadrant or maintains the same affect strength, even though some individuals and groups may lean toward one or another ideal-typical family of e-motions in their self-construction and overt behavior. We can hypothesize that agents with the pronounced joyful disposition tend to be future-oriented, confident, proactive, exhibit higher energy levels, and see themselves as having a positive attitude. Melancholic agents dwell in the past, have problems with self-esteem, show lower energy levels, hesitate to act, somatize bad experiences, and are more apt to conceal their feelings. Agents with choleric proclivities tend to focus on the present, be over-confident, impatient, aggressive, sarcastic, and reveal a good deal of negativity in their affective life. And the agents with a pronounced serene disposition aspire to transcend time, reach for the eternal, seek enlightenment, strive for self-mastery, favor edifying discourse, and sublimate their desires.
It would be wrong to use A-scan to pigeon-hole individuals or groups, however. It is the movement across the affective boundaries that reveals the most interesting and enduring patterns in our affective life. Nonclassically-propertied objects that we are, we are ever ready to take the role of the other, leap from one semantic emo field to another, and appropriate discrete labels from the culturally certified terminological frames in terms of which we can terminate affective indeterminacy in any given situation. A-scan is best suited for studying patterns of emotional ambivalence, revealing the structures of indeterminacy, and compiling multi-dimensional e-profiles for individuals, groups, and cultures.
The ETM survey is a self-assessment tool that scans human agency for clues to how an individual or group agent experiences its affective being in key cultural domains. The survey elicits responses to affectively-laden statements corresponding to specific emos in the ETM chart. Emo templates cannot be presented to respondents directly, for they may have multiple meanings, carry off-putting connotations, or elude respondents with limited vocabulary. The ETM survey translates emo templates into definitions which help compensate for the vagueness of the emo terms plucked from natural language. Narrative sequences in the survey also permits a better calibration of each emo’s intensity relative to the kindred terms appearing in the same ETM quadrant.
Many respondents would be perplexed if asked whether they are “travestied” (3.4.6), but once the survey spells out the term’s meaning as “pretending to feel what you actually do not feel,” you are likely to get a response, along with the frequency with which the respondent finds oneself in such situations. Some respondents may be less than certain what being “tolerant” (1.3.8) means, but when the concept is unpacked as “accepting of people whose views or ways you disagree with,” the response is likely to point in the right direction. It might be awkward to invite the individual to identify with a template “impotent” (3.4.2) but not with its narrative equivalent – “failing to rise up to the occasion when you are expected to perform.” And survey takers are certain to find off-putting the question of whether they have “masochistic” (3.5.5) or “sadistic” (2.5.5) tendencies, yet we can dig relevant information by asking respondents how frequently they feel “being humiliated or violated yet doing noting to avoid future assaults” or find themselves “deriving pleasure from hurting someone morally, physically, or sexually.” For each of the 200 templates, the survey has a moniker that spells out in more or less common terms what a particular emo means. An added advantage of converting emo templates into narrative definitions and asking respondents to indicate the frequency with which they find themselves in such situations is that no one is tagged here outright with a label masochistic, vindictive, courteous, or serene – we are dealing only with tendencies, with each emo checked against kindred and alternative templates. This operation yields an affective complexity while highlighting certain tendencies that can be gleaned from the agent’s choice of affectively-tinged emos and accompanying frequency indicators.
Some survey entrees may still pose a problem for respondents. For instance, the template nirvanic (4.5.1) is explicated in the survey as “being in a state of magnificent trance as if transported to another world or dimension,” and the template crucified (3.5.10) is defined as “feeling God-forsaken as if nailed to the cross” – these narrative explications may elude respondents with limited vocabulary or life experience. Those uncertain about such survey entrees can tag them “unsure.” Further analysis of aggregate data will show if a particular wording elicits a disproportionate number of such responses and needs to be clarified or replaced. Polling experts familiar with a given cultural milieu help validate proposed emo definitions.
The ETM survey, also known as MoodCounts, is presently available on the web: http://184.108.40.206/. The survey is free, anonymous, and confidential; it usually takes about 15-20 minutes to complete; and it generates a three page computer report. The instructions to those taking the survey on the web read as follows:
Place yourself in a situation that makes you feel or act in a certain way, think how often you find yourself in such situations, then click on the dot indicating how frequently this happens. The list of feelings and emotionally charged situations appears below. With respect to each item you should be able to say “I feel” or “I act” this way “always,” “very often,” “often,” “sometimes,” “rarely,” “very rarely,” or “never.” Take a longer-term perspective, think how you have been feeling in recent weeks or months rather than in the last day or two. You may come across an unfamiliar emotional experience or situation. If you are unable to figure out what it means, you can indicate so by clicking the “Unsure” button. If you miss a particular item, you will be prompted to return to it and answer the question.
The first page of the e-profile printout lists individual emos and the frequency with which they surfaced in the respondent’s answers to the survey questions. The second page of the printout shows the survey taker’s emotional indicators alongside the database mean scores. The third page is a graph visualizing 51 ETM indexes relative to the database means, with the respondent’s z-score readings for each indicator. The survey printout reproduces the E-motion Template Matrix (see table 1 or visit the CDC web site at http://www.unlv.edu/centers/cdclv/programs/etm_table.html) which displays 200 emos that are grouped together according to their valence (positive/negative affect), place in the family of e-motions (Joy, Serenity, Anger, Fear), intensity (#1 designating the weakest and #5 – strongest emos), and agency dimension (from 1 through 10). At the top of the printout that the user obtains after completing the survey one finds R# (Respondent number) assigned to each new respondent in the order in which the survey was taken, and a password known only to the respondent who chose to have one. Each emo in the A-scan printout is visually coded, depending on how frequently it occurs in the person’s life. Emos marked in the survey as “always” appear in the printout bolded, “very often” – bolded and italicized, “often” – italicized, “sometimes” – printed in regular case, “rarely” – earmarked by a star, and “very rarely” – carry a circle. The emos marked “never” are distinguished by a black dot. Respondents who left a particular question unanswered or marked “unsure” will see a question mark – “?” – against such item. The key to frequency coding is located at the bottom of the printout’s first page.
The first to be noticed on page one of A-scan are the bolded items. These are the most characteristic e-motions in the person’s or group’s profile, the ones that were marked in the survey as “always” or “very often.” Next, one can check the emos marked by a black dot – these templates are lacking in a given emotional profile, i.e., the respondent disclaimed such feelings altogether by tagging them with the frequency “never.” E-motions earmarked “sometimes” indicate that they may or may not appear on any given occasion. Items marked by a star or a circle are low frequency e-motions – emo templates uncharacteristic for a given profile, although discerned by the respondent. By scanning the entire E-motion Template Matrix the interpreter will instantly see which emos stand out in a given profile (bolded items) and identify regions in the A-scan where the respondent’s affect peters out. You can also spot the positive affect emos in the Joy and Serenity columns that have direct counterparts on the negative affect side in the Anger and Fear columns.
The last bit of information is particularly important, for unlike traditional personality tests, ETMA aims to elicit affective contradictions that attest to the kind of ambivalence that Montaigne was alluding to (see the quote above).
Someone who owned up to feeling “cheerful, sunny” (buoyant – 1.3.1) may also mark the monikers “put off by someone’s actions” (frustrated – 2.3.1), “worrying, upset by problems” (anxious – 3.3.1), or “maintaining peace of mind whatever happens” (serene – a 4.3.1). When these mid-intensity emos are checked with the frequency “sometimes” or higher, we can interpret the response as a sign of emotional ambivalence. Paired emos of high intensity (4th and 5th column position emos) and great frequency (“very often,” “always”) will attest to even stronger affective volatility and ambivalence in the respondent’s e-profile. This would be the case with respondents who tend to feel “disgusted with the world to the point of lashing out at those who defy you” (infuriated – 2.5.1) and who at the same time acknowledge that they are often “experiencing horror in the face of developments that threaten your well-being” (terrified – 3.5.1).
To ascertain what a particular template means, we should go to its survey definition. The situation is further clarified once we move to the indexes based on specific emo sets. The five emos comprising each cell in the Emotion Template Matrix form a basis of an index that furnishes insight into a particular e-profile, or if you will, the agent’s “emotional blood work.” Each of the 40 cell indexes is given a name, reflecting the character and affective halo of emos clustered in the cell. The ETM index table shows how each individual index is grounded in a particular ETM cell (see table 2). Cell indexes will appear on the second page of the ETM printout, each one identified by its name and a brief description, e.g., “Civility – treating people with courtesy,” “Stoicism – staying calm under pressure,” “Subordination – feeling dependent on powerful people,” “Verbal dexterity – having command of language,” “Anxiety – feeling persistently alarmed,” and so on. Individual cell indexes are aggregated into cumulative indicators, revealing the individual or collective agent’s Social affect (interpersonal skills), Political affect (community involvement), Spiritual affect (spiritual yearnings), Dynamic affect (energy resources), Agonistic affect (anger control), Existential affect (depression management), and Reflexive affect (self-regard). “Agentic Affect” (overall positive attitude) is a meta-index that averages scores in seven cumulative indicators and hints at an overall emotional wellbeing.
ETMA calculates several additional cumulative measures that cut across four families of emotions
and the divide separating positive and negative
affect emos. Such meta indexes include the “Positive/Negative Affect Ratio” that measures the ratio of Joy-Serenity to Anger-Fear emos
in the agent’s e-profile; “Negative Affect Recognition” and “Positive Affect Recognition,” indicating how many emos out of all possible in a given category show up in one’s emotional palette with some frequency; “Ambivalence” that reveals how many positive affect emos have a match on the negative affect side; “Negative Volatility” which reveals the tendency to flip from fear to anger and anger to fear, attested by the number of matched anger-fear emo pairs; “Overall Affect Strength and Persistency” that averages the respondent’s total affect across four families of emos; and “Outwardly/Inwardly Directed Affect Ratio” that compares the
strength and persistency
of Joy/Anger emos with the strength and persistency
of Serenity/Fear emos.
Page two of the A-scan printout displays the scores of a person or a group alongside the mean scores for the entire database. Individual scores appear on the right side of the second page and database mean scores appear on the left. Except for the Negative/Positive Affect Ratio and Outwardly/Inwardly Directed Affect Ratio, ETM indexes range from 0 to 10. To compare one’s index scores with the database averages, the respondent can check the rank or percentile column (e.g., if your rank is 90% on a particular index, you scored in the top 10 percentile and 90% of all respondents scored lower than you on this index; if your rank is 10%, you are at the bottom 10% and 90% of all respondents scored higher than you on this item). The Z-score that appears in the second column from right shows how much your individual score deviates from the database mean (the plus or minus sign indicates whether you scored below or above the database average). Page three of the A-scan printout is a graph visualizing how each of the respondent’s ETM indexes is related to the database mean scores in terms of standard deviation.
While emotional wellness is directly related to positive affect, this relationship is not linear: too much Joy and Serenity in the agent’s e-profile spells trouble when combined with the low negative affect recognition – a percentage of Anger/Fear emos present in the agent’s e-profile out of the total 100 negative emo templates in the ETM matrix. A high overall positive attitude (Agentic Affect) that goes with low negative affect recognition may signal the tendency to repress negative emotions and reveal a strong self-enhancement bias. It is postulated that emotional wellness – E-motional Versatility – correlates with the highest possible Agentic Affect (an overall positive attitude), consistent with the highest possible negative affect recognition. Interpreting A-scan results, one should also take under consideration that each person has a somewhat different internal scale on which he or she assesses the intensity and frequency of one’s e-motional states (what is “often” for one person might be “very often” for another), that individuals with a lower overall affect strength tend to have lower individual index scores, and that, consequently, the agent’s scores must be compared not only to the database averages but also to the agent’s baseline affect (the Overall Affect Strength).
One should avoid reading too much into any individual e-motion in a given profile. Strong emos (bolded items in the printout) clustered in a given ETM cell may signal a persistent mood, but it can be misleading outside the larger context. A cumulative index based on several cell indexes, especially if it differs substantially from the database average, suggests a notable affective pattern. More important, still, are the mood swings
captured by the cumulative indexes of Ambivalence and Volatility, as well as by the Positive/Negative Affect Ratio which encompass the full range of the agent’s emos. In the end, A-scan achieves its aim
to the extent that
it encourages the agent to count its moods, recognize its ambivalent attitudes, and take seriously one’s affective life.
We should note that measuring the ETM survey’s reliability raises uncommon theoretical and methodological questions. The standard test-retest series used to show how well a measuring tool reproduces the initial results when applied to the same object on similar occasions reflects the assumptions that the object under study is stable, enduring, reproducible. By contrast, ETMA is rooted in the premise that human agency is a stochastic phenomenon whose manifestations are a matter of probability, which means that the failure to elicit an identical response under similar circumstances may reflect the object’s inconstancy rather than the instrument’s deficiency (Shalin 1986:20). The respondent filling the ETM survey on two separate occasions is likely to make different selections, reflecting the respondent’s situational needs and changing life
circumstances. What we should be looking for in our case are the patterns of uncertainty that show up not on the level of individual emos or cell indexes but on the aggregate level of cumulative and meta indexes like Agentic Affect, Ambivalence, and the P/N Ratio. The arrangements are currently made to compare the ETM survey results from repeat users who took the survey once as a hard copy and a second time on the internet. (The first-time users taking the ETM survey on the internet have an instant access to the results, which may affect their answers in subsequent trials).
The validity of the ETM survey has been vetted through expert assessment by scholars, primarily sociologists, who reviewed the wording of the survey entrees and suggested how these could be rephrased to make them more transparent and reflective of the template’s meaning. Comparing the ETM findings about the relationship between emotions and demographic variables with the relevant data obtained by researchers studying the subject by alternative techniques will shed more light on the survey’s validity. The ETM survey contains information about the respondent’s gender, marital status, involvement in a steady relationship, the number of children, age, race/ethnicity, education, occupation, income, religiosity, and residence. Supplementary questions give respondents a chance to rate their health, work, financial status, family situation, and future prospects. Such data comes especially handy when researchers conduct A-scan and prepare an e-motional culture profile for a group or an organization.
Close to 1750 individuals have taken the ETM survey as of August 1, 2006. Students and faculty at the University of Nevada Las Vegas comprise the largest segment of the ETM sample (about 70% of the total). Surveys were also collected from Clark County School District teachers and administrators, senior citizens organizations, professional associations, and private companies in Southern Nevada. While this nonrandom opportunity sample is skewed toward younger people and respondents with a college education, preliminary findings are noteworthy insofar as they illuminate demographic differences and general affect dynamics.
Unless otherwise indicated, all correlations reported in the narrative summary below have a significance level of p < .001, Sig. 2-tailed (more detailed data and additional ETM findings are summed up in tables 2-15 of the appendix). For the purpose of the present analysis which operates with demographic and attitudinal variables, correlations with a confidence level not exceeding plus/minus .199 will be identified as weak, those in the .200-.399 range as moderate, .400-.599 as strong, and .600 and up will be labeled as very strong. The discussion starts with the descriptive statistics, after which it moves to the socio-demographic differences and leading social indicators and their association with affective life. The final section outlines several key insights ETMA has to offer regarding the general affect dynamics.
Individual Emo Distributions reveal the most popular e-motions, i.e., emos with the highest frequency in the population under study (the frequency “always” has been assigned the value of 7, “very often” – 6, “often” – 5, “sometimes” – 4, rarely – 3, very rarely – 2, and never – 1). The emo with the highest mean of 5.96 in the present sample is grateful (4.2.4), which means that an average ETM respondent marked the survey entry corresponding to this emo – “Very thankful for what you’ve got” – just short of “very often.” The lowest frequency emo in the ETM database is sadistic (“Deriving pleasure from hurting someone morally, physically, or sexually”) – its mean frequency is 1.45. That is to say, respondents who spot such a trait in themselves tend to see it as highly atypical, averaging for the entire sample as less frequent than “very rarely” (which has the value of 2).
The top 10 emos in the ETM database are grateful (5.96), hospitable (5.62), conscientious (5.61), benign (5.59), positive (5.58), confident (5.55), optimistic (5.45), generous (5.45), charitable (5.43), courteous (5.33), and proud ”(5.32) – all but one templates here are from the Joy family of emos (templates "grateful" and “conscientious” belong to the Serenity quadrant). The 10 least frequently chosen emos are sadistic (1.48), ominous (1.66), deviant (1.68), devious (1.72), crucified (1.77), bedeviled (1.81), callous (1.86), devilish (1.88), rabid (1.88), and dictatorial (1.99) – this list contains only negative affect emos, with two items from the Fear quadrant (crucified and bedeviled) and the rest from the Anger emo family.
It is interesting that the top 73 items in this frequency list of 200 emos are all positive affect templates. The highest frequency negative affect emo is “solicitous” (“Trying to win other people’s approval”), which has 4.07 mean score (just over “sometimes”). Following this one is a cluster of negative affect e-motions: brash (4.04) egocentric (3.90), blasé (3.85), abashed (3.84), anxious (3.83), bothered (3.81), reserved (3.78), withdrawn (3.76), bossy (3.59), and estranged (3.55). The least frequently chosen positive affect emos, beginning with the high intensity serene template nirvanic (2.76), are transcendent (2.75), Delphic (2.59), luminous (2.48), revered (2.41), dazzled (2.58), and Buddha-like (1.98). The least popular negative affect emos are dictatorial (1.88), cruel (1.84), devilish (1.84), bedeviled (1.79), crucified (1.76), devious (1.68), deviant (1.65), ominous (1.65), and sadistic (1.46).
The Positive/Negative Affect Ratio is an important gauge of emotional life, with the higher P/N ratio generally reflecting a greater overall affective wellness. The database mean for this emo indicator is 1.90, suggesting that an average respondent has nearly twice as many positive emos in one’s e-profile as negative ones. The lowest P/N ratio for the present sample is 0.5, the highest one is 14.1, with the next highest mark being 9.7. Sixty-seven respondents have P/N ratio of 1 or lower; 47 survey takers have the ratio of 4.0 or higher.
Individual Indexes represent the affect strength and persistency in one of the 40 cells of the E-motion template index chart. There are several ways to calculate individual indexes, the most obvious one involves averaging scores for all 5 emos in a given cell. Another method takes into account the cell saturation measured by the number of high frequency emos in the cell (the more bolded items in the cell, the greater its affective strength). For the present analysis, a cell index is calculated as the highest multiple of the emo’s intensity (where 1 is minimum and 5 is maximum) times emo’s frequency coefficient (2 is a coefficient assigned to the frequency “always” and 0 to the frequency “never”). The highest possible index score for any given cell is 10
(the highest intensity emo found in the fifth column position times 2, representing the highest frequency coefficient of “always”) and the lowest one is 0 (any emo times 0, standing for the lowest frequency coefficient of “never).
The top five mean scores for individual indexes are Gratefulness (7.122), Generosity (6.698), Self-confidence (6.548), and Civility (6.447). The lowest mean scores for individual indexes are Deviousness (2.062), Callousness (2.339), Authoritarianism (2.561), Subordination (2.670), and Envy (2.933). Note that the ETM printout displays 36 out of 40 cell indexes, with their names changed in some cases to make indexes more user-friendly (e.g., Insincerity index is transformed in the printout into Sincerely index – “speaking your mind,” Intolerance index is converted into Tolerance index – "respecting those who disagree with you,” and so on).
Cumulative Indexes aggregate data from several kindred cells. Thus, cumulative index labeled “Social affect” averages scores the individual or a group receives on individual indexes of Civility, Tolerance, Sincerity, and Verbal dexterity. The agonistic affect is the inverted mean score (10 minus the mean) for Testiness, Irritability, Callousness, and Deviousness. The Existential affect index is the reversed mean score for indexes Depression, Anxiety, Isolation, and Beleagueredeness. And Agentic Affect is a meta index averaging scores for 7 cumulative indexes.
The highest mean for cumulative indicators is 7.23 – Agonistic affect (anger management); the lowest is 5.69, and it represents Reflexive affect. Agentic Affect
(the overall positive attitude)
has a mean score of 6.28. Here are database means for several other cumulative indexes: Positive Affect Recognition (ability to spot positive emos) – 9.10; Negative Affect Recognition (ability to spot negative emos) – 7.65; Ambivalence (tendency to switch from positive to negative emos) – 4.65; Negative Volatility (tendency to switch from anger to fear) – 3.19; and Overall Affect Strength and Persistency (total affect level across four families of emos) – 4.33.
Gender Differences that come through in the ETM findings (see table 5) tend to be weak, and they are most pronounced in the negative affect areas. The most noticeable gender effect is in the anger family of emos where women’s higher score on agonistic affect (.246) points to their better anger management skills. Women also appear to be more civil (.119) and tolerant (.160) than men. Women’s e-profiles register less authoritarianism (-.103), strictness (-.138), egocentrism (-.140), envy (-.137), and negative volatility (-.119). Men show higher levels of verbal aggressiveness that comes across in emos like “sarcastic,” “sardonic,” and “acerbic,” as well as in the cell index showing testiness, which is less tangible in women (-.201). Women see themselves as having lower than men public esteem (-.103), verbal dexterity (-.098), and self-confidence (-.097). These findings are consistent with what we know from the prior research on gender differences in our culture, which encourages men to be more aggressive and intemperate than women. There is no tangible gender impact on the P/N ratio for ETM respondents.
Age Differences are mostly moderate in strength, and they can be seen particularly in the negative affect emos. The overall anger level diminishes with age (-.270), and so does fear (-.227), negative volatility (-.242), overall strength of affect (-.233), egocentrism (-.212), strictness (-.219), deviousness (-.208), subordination (-.199), alienation (-.192), envy (-.203), and overall stress level (-.201). The older you are, the better you are at managing fear (.183) and the higher is your political affect (.151) and tolerance (.104). An overall stress level drops with age (-.201) while P/N ratio increases (.142).
Marital Status Differences are evident in the overall affect strength, which is lower among married respondents (.171), as well as in the familiar patterns of anger (-.205) and fear (-.188) that tend to decline with age. Unmarried respondents are more likely to be vicious (.219), disaffected (.179), subjugated (.173), lowly (.165), deviant (.155), anxious (.128), and annoyed (.107). They have lower agonistic (-.200) and existential (-.157) affect levels, and they are more apt to flip from anger to fear and back, as attested by their higher negative volatility (.177). Unmarried respondents have lower P/N ratio (-.123). We do not know whether marriage increase the P/N ratio or individuals with higher positive affect markers are more likely to marry.
Education Differences are apparent in the fact that more educated people tend to score higher on existential affect (.151), agonistic affect (.105), and verbal dexterity (.100). They feel less subordinate (-.171), alienated (-.100), and exhausted (-.113) than their less educated counterparts. The P/N affect ratio is marginally higher for educated respondents (.066, p < .01). The 62 respondents with Ph.D.’s found in the ETM database have above average verbal dexterity means (6.49 compared to DB mean of 5.93) and sincerity (6.64 vs. 6.02); their subordination index is lower than the database mean (1.83 vs. 2.70). Ph.D. holders’ existential affect (the ability to keep fear and anxiety under control) is higher than the average (7.05 vs. 6.44), and so is their agonistic affect (7.61 vs. 7.22). Ph.D. holders also have a higher P/N ratio (2.06 vs. 1.90).
Income Differences follow a similar pattern, with an overall anger level falling with income (-.196)and an overall fear level diminishing with the earning power (-.260). The agentic affect (overall positive affect) improves with income (.204), as does political affect (.225), while the depression level drops (-.205). When it comes to individual emos, respondents with higher incomes are more likely to see themselves as benign (.253), lawful (.201), and less likely to feel suspicious (-.265), cursed (-.201), lowly (-.236), and subservient (-.189).
Religious Commitment Differences are somewhat more pronounced in the sample under review, with several relationships crossing the .300 and .400 marks. Respondents identifying themselves as “religious” and “very religious” tend to have higher overall serenity (.338) and joy (.271), and a higher overall positive affect level (.328). Religious individuals sport higher energy levels (.141), empowerment (.168), and self-regard (.312), although their fear level is not appreciably lower than in their nonreligious counterparts. Predictably, religious respondents score higher on such cell indexes as observance (.315) and charisma (.439), with their spiritual affect cumulative indicator registering well above the database mean (.352).
Leading Indicators measure one’s satisfaction with health, finances, job/work, family situation, and future prospects. Respondents rate their own situation on a five-point scale, ranging from “excellent” to “very bad.” A cumulative quality of life index based on these components is called “Leading Indicators” (LI). There is also an overall stress (OS) index and overall life satisfaction (OLS) index, both measured on a five-pint scale, ranging from “very high” to “very low.” Focusing on LI for a moment, we can see a moderate to strong relationship between affective markers and perceived life situation (see table 6). Respondents with higher LI indicators show lower fear and anger and higher joy and serenity levels. As leading indicators and an overall life situation improve, so do the respondent’s affective markers. The OLS measure is inversely related to guilt (-.301), alienation (-.340), isolation (-.350), and depression (-.502), while showing a moderate-to-strong association with cumulative indexes: social affect (.230), reflexive affect (.315), spiritual affect (.346), political affect (.382), dynamic affect (.382), existential affect (.479), and agentic affect (.486).
Moving to specific leading indicators, we can spot several noteworthy relationships. The more respondents are satisfied with their health the less likely they are to feel alienated (-.201), guilty (-.213), exhausted (.-255), and depressed (-.295), and the more likely they are to be tolerant (.148), spiritually committed (.172), politically engaged (.173), self-confident (.210), and enthusiastic (.218). Health is positively associated with dynamic affect (.267) and P/N ratio (.229).
Similar dynamics holds with other quality of life indictors. Married respondents have a stronger existential (.323) and agentic (.304) affect, and they are more likely to be satisfied with their overall life situation (.513). Job satisfaction significantly lowers fear levels (.314) and improves self-confidence (.224). Respondents with better finances feel less isolated (-.225) and anxious (-.235), just as they show lower negative volatility (-.228). Correlations between leading indicators and affective markers tend to be moderate, but their direction, consistency, and high significance level attest to their potential theoretical and practical importance.
General Affect Dynamics
One finding that stands among the preliminary results and that is consistent with the ETMA premises is a pattern of ambivalence characterizing our affective life, i.e., the tendency to experience seemingly contradictory affective attitudes toward oneself and one’s social engagements. In a narrow sense, the term “ambivalence” refers to a tendency to feel incongruent, sometimes downright contradictory, sentiments like “sincere” and “insincere,” “confident” and “discouraged,” “authoritative” and “subservient,” “proud” and self-loathing,” “tolerant” and “intolerant,” and so on. The measure of e-motional ambivalence used in ETMA is the number of anger/fear templates that have a match in the joy/serenity domain alongside an agency dimension where the paired emos are nested.
Another ambivalence indicator is “negative volatility,” and it refers to the tendency to switch from fear to anger and from anger to fear. While ambivalence entails crossing the border separating negative from positive emos, negative volatility tracks movement between fear and anger, which comprise the negative affect domain. Negative volatility index is calculated as the number of fear emos that have a match on the anger side. Correlatively, “positive volatility” is measured as the number of paired emos belonging to joy and serenity families of emotions. One can also track the affective agency movement along the serenity/anger and joy/fear diagonals.
As articulated above, ETMA treats ambivalence not as aberration but as a fundamental quality of human agency. Affective and behavioral consistency is seen here as an emergent phenomenon, an ongoing accomplishment, rather than as a property inherent in the individual. It is not so much “affect” as “affectivity” that the ETM pragmatist methodology is after, not “conduct” but “conductivity” that situates the individual on the intersection of overlapping social currents and meaning fields from which the agent can borrow symbols, behavioral gambits, or emotional stratagems promising to satisfy the agent’s needs and situation demands.
The ambivalent, or if you will, the stochastic nature of human agency is most evident in the case of negative volatility that captures the tendency to flip from fear to anger and back. As you can gather from the relationships reported in table 7, paired emos in the fear section are correlated with their counterparts in the anger quadrant. What is particularly interesting is that the association grows in strength as we move from the weakest emos appearing in column 1 through the medium strong emos in column 3 and all the way to the highest intensity emos situated in column 5. Moving along the self-confidence continuum (the Vitality/Mood agency dimension), we can see that fearful emo “blue” is positively associated with the angry emo “brash” (.77, p < .01). But the highest affect emo represented by the template “depressive” (its survey moniker is “Feeling like a complete failure”) has a stronger association of .344 with its pair “manic” (“Frenzied, charged with unreasonable enthusiasm”). The medium strong emo “guilt-ridden” (“Feeling ashamed, rebuffed”) is tied to the emo “callous” (“Ignoring pain you cause other people”) by a moderate correlation of .251. The highest intensity fear emo in the same dimension – “masochistic” (“Being humiliated or violated yet doing nothing to avoid future assaults”) is strongly correlated with its choleric counterpart “sadistic” (“Deriving pleasure from hurting someone morally, physically, or sexually”) as evidenced by Pearson correlation of .413. The mid-range emo “subservient” (“Feeling powerless when dealing with your superior”) and its counterpart “authoritarian” (“High-handed, not very concerned about the options of others”) are correlated at the level of .238. But at the highest level of intensity we spot a strong correlation of .616 between e-motion “crucified” (Feeling God-forsaken, as if nailed to the cross”) and its counterpart “dictatorial” (“Exercising an arbitrary power over people’s life”).
We see a similar pattern with positive volatility that tracks paired emos in joy and serenity quadrants. Just as in the case of fear-anger, all the relationships between joy and serenity emos are positive, although the association is weaker than the one reported above, and it is less dependent on the emo strength. Thus, the emo “exhilarated” (“Extremely happy about the way things are going”) is correlated with the emo “nirvanic” (“Being in a sate of magnificent trance as if transported to another world or dimension”), the association is .183. The correlation of .231 ties the emo “chivalrous” (Worshiping the person you love while nobly pursuing your loved one”) and the emo “sublimated” (“Freeing yourself from base instincts by the power of your will or spiritual uplift”). And the emo “audacious” (“Supremely confident about your ability to achieve all your goals”) is moderately associated with the emo “entranced” (“Reaching a profound sense of inner peace after disengaging from the troublesome world”), at the level of .323.
A noteworthy association exists between anger and serenity, although it is less consistent than in the case of anger-fear and joy-serenity emos. Some of the paired emos in the anger-serenity quadrants are negatively correlated. The emos that show up simultaneously in the agent’s e-profile and have a positive relationship are “infuriated” (“Disgusted with the world to the point of lashing at those who defy you”) and “nirvanic” (“Being in a sate of magnificent trance as if transported to another world or dimension”), the two are positively correlated (.167). The emo “manic” (“Frenzied, charged with unreasonable enthusiasm”) and the emo “entranced” (“Reaching a profound sense of inner peace after disengaging from the troublesome world”) are positively related (.135). And the emo “ominous” (“Being dark-minded, sinister” is associated with the emo “awesome” (“Overcoming the dark forces with the cheer power of your will”), .100.
The association is mostly negative between the Joy and Anger families of emos. Notable exceptions are emos “authoritative” and “authoritarian” (.050, p < .01), “optimistic” and “egocentric” (.069, p < .01),), “witty” and “acerbic” (.096), “heroic” and “dictatorial” (.171), and “competent” and “bossy” (.186). The strongest relationship between anger and serenity is between the emos “blissful” (“Charging ahead to fulfill your spiritual or divine destiny”) and “devilish” (“Trying to harness awesome powers to mastermind ominous events” – (.244).
The reported relationships can be explained in more than one way. It might be a semantic artifact reflecting poorly worded questions in the survey that do not tie in with the designated labels and do not justify assignment to a particular emo family. A case by case analysis of emo pairs and their likely meanings should answer the question as to whether the reported relationship is a semantic glitch. In some cases, particularly in the case of emos assembled in the serenity quadrant, this may prove to be the most plausible explanation. Such an interpretation would seem to be far-fetched in other cases, particularly when we deal with the more readily identifiable fear and anger emos.
Another explanation is that some people might be attracted to extremes, choose high intensity emos, and favor high frequency marks, such as “always” and “never.” There may be a contingent of respondents with such a response set, although this would be an interesting phenomenon in its own right.
The ambivalence phenomenon reported above may have another explanation, the one consistent with the ETMA premise that human agency is stochastic, indeterminate, and marked by ambivalence. This explanation is particularly plausible when we deal with negative volatility involving anger and fear. This reading of the ETM findings is consistent with the thesis advanced by Tom Scheff and other researchers studying emotions, who report the tendency to move from shame to rage and point to a close relationship between fear and anger.
This paper has outlined the ETMA theory and methodology and presented preliminary results of the ETM survey. The ETM Survey is a sociological instrument based on the premise that our sentiments are situational, that we often harbor contradictory feelings about ourselves, and that our actions reflect conflicting social pressures and the ambivalent way in which we experience the world. Demographic groups were shown to have distinct e-motional profiles, and affective life was tied to the leading indictors, such as perceived health, family situation, financial situation, job situation, and future prospects. Among the most intriguing findings reported in this paper was the pattern of ambivalence observed among ET respondents who identify with conflicting e-motions and show a moderate-to-strong tendency to switch from fear to anger. A similar, though less, pronounced tendency ties joy-serenity emos, and anger-serenity e-motions.
Future ETM research should track separate impact that different demographic factors have on affective life, as well as explore e-motional profiles of different social groups. Further research is needed to document the role that emotions play in a democratic polity. ETMA should also help articulate a social pedagogy that can facilitate civic society whose members are emotionally intelligent, affectively literate, and agentically creative.
Allen, J. and K. R. Murphy (2006). Critique of Emotional Intelligence: What Are the Problems and How Can They Be Fixed? Lawrence Erlbaum Associates.
BarOn, R. (1997). The BarOn Emotional Quotient Inventory (EQ-i): A Test of Emotional Intelligence. Toronto, Canada: Multi-Health Systems.
Bar-On, R. and James D. A. Parker, eds., (2000). The Handbook of Emotional Intelligence. Theory, Development, Assessment, and Application at Home, School, and in the Workplace. San Francisco: Jossey Bass.
BarOn, R., & Handley, R. (2003). The BarOn EQ-360. Toronto, Canada: Multi-HealthSystems.
Bocchino, R. (1999). Emotional Literacy. To be a different Kind of Smart. Thousand Oaks, CA: Corwin Press.
Barchard, K. A. (2003). “Does Emotional Intelligence Assist in the Prediction of Academic Success? Educational and Psychological Measurement 63:840-858.
Barchard, K. A. and A. Hakstian (2004). “The Nature and Measurement of Emotional Intelligence Abilities: Basic Dimensions and Their Relationships with Other Cognitive Ability and Personality Variables.” Educational and Psychological Measurement 64:437-462.
Barchard, K. & Russell, J. (2004). Psychometric issues in the measurement of emotional intelligence. In G. Geher (Ed.), Measuring Emotional Intelligence: Common Ground and Controversy. New York. Nova Science Publishers.
Blazer, D. G. (2005). The Age of Melancholy: Major Depression and Its Social Origins. Taylor & Francis.
Collins, Randall. 1990. “Stratification, Emotional Energy, and the Transient Emotions.” Pp. 27-57 in Research Agendas in the Sociology of Emotions, edited by T. Kemper. New York: Sate University of New York Press.
Dewey, J. 1954. The Public and Its Problems. New York: Henry Holt & Co.
Davitz, J. 1969. The Language of Emotions. New York: New York: Academic Press.
Elias, Norbert. 1978. The Civilizing Process. Vol.1. The History of Manners. New York: Pantheon Books.
Elias, M. J. and Arnold, H. eds. (2006). The Educator's Guide to Emotional Intelligence and Academic Achievement: Social-Emotional Learning in the Classroom. Corwin Press.
Franks, David. 1991. “Mead’s and Dewey’s Theory of Emotion and Contemporary Constructionism.” Journal of Mental Imagery 15:119-137.
Franks, David D. and Thomas S. Smith, eds. 1999. Mind, Brain, and Society: Toward a Neurosociology of Emotion. Social Perspectives on Emotions. Vol. 5. Stamford, Conn.: JAI Press.
Geertz, H. 1959. “The Vocabulary of Emotions: A Study of Javanese Socialization Processes.” Psychiatry 22:225-236.
Goffman, Erving. 1972. Relations in Public. Harmondsworth: Penguin.
Gardner, H. (1993). Multiple Intelligences. New York: Basic Books.
Goleman, D. (1995). Emotional Intelligence. New York: Bantam Books.
Goleman, D., R. E. Boyatzis, and A. McKee. 2002. Primal Leadership: Realizing the Power of Emotional Intelligence. Harvard Business School Pr.
Goleman, D. (1998). Working with Emotional Intelligence. New York: Bantam Books.
Goleman, D. (2006). Social Intelligence: The New Science of Human Relationships. New York: Bantam Books.
Harre, Rom, ed. 1986. The Social Construction of Emotions. New York: Basil Blackwell.
Herrnstein, R., and Murray, C. (1994). The Bell Curve: Intelligence and Class Structure in American Life. New York: The Free Press.
Hochschield, A. R. (1975). “The Sociology of Feelings and Emotions.” Pp. 280-307 in Another Voice, edited by Marcia Milman and Rosabeth Moss Kanter. Garden City, N.Y.: Doubleday.
Hochschield, A. R. (1983). The Managed Heart. The Commercialization of Human Feeling. Berkeley, CA: University of California Press.
Hochschield, A. R. 1998. “The Sociology of Emotions as a Way of Seeing.” Pp. 3-15 in Emotions in Social Life. Critical Themes and Contemporary Issues, edited by Gillian Bendelow and Simon J. Williams. London: Routledge.
Kemper, Theodore D. 1978. A Social Interaction Theory of Emotions. New York: Wiley.
Kemper, Theodore D., ed. 1990. Research Agenda in the Sociology of Emotions. New York: State University of New York Press.
Lutz, C. 1988. Unnatural Emotions: Everyday Sentiments on a Macronesian Atoll and Challenge to Western Theory. Chicago: University of Chicago Press.
Kirp, David L. “After the Bell Curve.” The New York Times, July 23.
Macur, Juliet. 2004. “Emotional Steinbrenner aims to put all his houses in order.” The New York Times, May 2.
Mayer, J. D., & Salovey, P. (1993). “The intelligence of emotional intelligence.” Intelligence 17(4):433-442.
Rosenthal, N. E. The Emotional Revolution. How the New Science of Feelings Can Transform Your Life. New York: Citadel Press.
Salovey, P. & Mayer, J. D. (1990). “Emotional intelligence.” Imagination, Cognition, and Personality 9:185-211.
Scheff, Thomas J. 1990. Microsociology: Discourse, Emotion, and Social Structure. Chicago: University of Chicago Press.
Scheff, Thomas J. 1994. Bloody Revenge: Emotions, Nationalism, and War. Boulder: Westview Press.
Scheff, Thomas J. and Suzanne M. Retzinger. 1991. Emotions and Violence: Shame and Rage in Destructive Conflicts. Lexington, Mass.: Lexington books.
Schulz, K. (2004). “Did Antidepressants Depress Japan?” The New York Times, October 23.
Shalin, D. N. “Emotional Barriers to Democracy Are Daunting.” The Los Angeles Times, October 27, 1993.
Shalin, D. N. (2001). “Emotion, Agency, and the Social Production of Affect. A Research Note.” Newsletter of the Sociology of Emotion Section of the American Sociological Association. 4:2-5.
Shalin, D. N. (2004). “Liberalism, Affect Control, and Emotionally Intelligent Democracy.” Journal of Human Rights 4:407-428.
Shalin, D. N. (2005). “Legal Pragmatism, an Ideal Speech Situation, and the Fully Embodied Democratic Process.” Nevada Law Journal 5:433-478.
Turkheimer, Eric (2003). “Socioeconomic Status Modifies Heritability of IQ in Young Children,” Psychological Science 14, no. 6.
Urch Druskat, V., G. Mount, and F. Sala. (2005). Linking Emotional Intelligence and Performance At Work: Current Research Evidence With Individuals and Groups. Lawrence Erlbaum Associates
Weisinger, H. (2000). Emotional Intelligence at Work. John Wiley & Sons.