Chapter
Two: A Guide to the Solar System
Does Agonistic Structure Exist, and if So, Is It Weak
or Strong?
For
the moment, let us assume that our gamble paid off and that we found robust
correlations among the three elements in our design—the attributes of
characters, the emotional responses of readers, and agonistic role assignments.
So what? Could we conclude that humanistic dualism and postmodern relativism
are both misconceived, that literary meaning, at least in these novels, is
determinate, that we have actually succeeded in reducing human nature to a few
specific, measurable dimensions, and that those dimensions are adequate to
account for broad structures of meaning in these novels? Well, yes. But there would
still be a question of causality, and a question of degree. Statistically
significant differences point to underlying regularities and thus suggest, tantalizingly,
causal mechanisms rooted in interactions between genetically transmitted
dispositions and environmental conditions. Such mechanisms invite adaptive
explanations and thus open out into the prospect of full and complete
explanatory hypotheses. Still, a statistically significant difference is in
itself not an explanation. It only points toward an underlying causal mechanism. Moreover, statistically
significant differences can be, and often are, very slight in the amount of
variance for which they account. An effect can be real but so trivial that it
offers little real hope of getting to the heart of a subject. Assuming that we
can demonstrate some degree of correlation among the elements in our design, we
still have to ask, just how important is this pattern? Is it a sun that governs
the motions of planets, or merely an asteroid wending its negligible way
through the vastness of space?
We
think agonistic structure is a central sun, but astronomers are naturally eager
to discover major celestial bodies. Cautious colleagues might well be justified
in suspecting exaggerations in the putative magnitude of any given discovery.
How to convince the sceptics, or even to reassure one’s self? The only way is
to go back over the ground, mapping out the coordinates, reducing bodies to
numbers so that relative magnitudes can be objectively evaluated. Adopting the
model of the scientific paper, divided conventionally into four
parts—introduction, methods, results, and conclusions—the material in this
chapter would be part three, the results for which methods were but
preparation, and without which conclusions would be mere speculation. In an
early draft of this book, the findings mapped out here occupied over half the
manuscript. The results now condensed into a single chapter can be regarded as
something like the model of a solar system, reduced to manageable proportions.
The level of resolution resulting from compression leaves only the main bodies
visible, filtering out the asteroids, bringing into high relief the basic
structure of the system.
The main bodies in this system are the
attributes of characters and the emotional responses of readers. There are three main groups of attributes:
motives, the criteria for selecting mates, and the five factors of personality.
Motives are the basis for action in human life. The basic motives used in this
study are the joints of human life history, articulating its main segments.
Selecting a sexual or marital partner drives reproductive success and evokes,
accordingly, exceptionally strong feelings. In the majority of the novels in
this study, selecting a marital partner is the central concern of the plot.
Personality traits are dispositions to act on motives. The dispositions are
themselves human universals, but individuals vary considerably in their scores
on specific traits—on pleasure-seeking, for instance, or sensitivity to pain. Personality
traits are primary constituents of individual identity, more basic and more
comprehensive, we argue, than the factors of social identity that shape
“identity politics.” Emotions are the proximal mechanisms that activate motives
and guide our social judgments, including our judgments of imaginary people.
In this chapter, we take up each of these four
categories in turn, describing the dimensions in each and displaying scores for
protagonists and antagonists. The scores for protagonists and antagonists are more
sharply polarized than the scores for all good versus all bad characters. Consequently,
scores for the major character sets most clearly illuminate the forces at work
in the system as a whole. Differences between major and minor characters are
noted along the way.
If scores on motives, mating, personality, and
emotions cannot demonstrate the significance of agonistic structure, it is hard
to imagine what could. If the agonistic patterns produced by these categories were
dim, feeble, and muddled, vague in outline and inconsistent in their relations
to one another, we would fairly have to admit defeat, acknowledging that agonistic
structure does not actually exist, or if it exists at all, does not account for
much in these novels. We would have to concede, however ungraciously, that other
categories, literary, ideological, epistemological, semiotic, perhaps, might
have more success in revealing deep structures of meaning in the novels. If, to
the contrary, motives, mating, personality, and emotions can in fact
demonstrate that agonistic structure is a central organizing principle in the
novels, it is hard to imagine what other categories more basic and
comprehensive could possibly be invoked. Agonistic structure in this design is
made up out of the elements of human life history. Critics challenging the
force of the findings described here would need to identify concepts and
patterns large enough so that by comparison human life history counts for
relatively little in the meaning produced by the novels. We are confident that challengers
will not be wanting, and that is a good thing. Critical pressure can probe
weaknesses, point to areas for further development, and suggest alternative or
complementary concepts and methods. For now, our own task is merely to see just
how far human life history can take us in delineating the celestial dynamics at
work in this particular system.
Getting Motivated
For the purposes of
this study, we reduced human life history to a set of 12 basic motives—that is,
goal-oriented behaviors regulated by the reproductive cycle. For survival, we
included two motives—survival itself (fending off immediate threats to life),
and performing routine work to earn a living. We also asked about the
importance of acquiring wealth, power, and prestige, and about the importance
of acquiring a mate in both the short term and the long term. In the context of
these novels, short-term mate selection would mean flirtation or illicit sexual
activity; long-term mate selection would mean seeking a marital partner. Taking
account of “reproduction” in its wider significance of replicating genes one
shares with kin (“inclusive fitness”), we asked about the importance of helping
offspring and other kin. For motives oriented to positive social relations
beyond one’s own kin, we included a question on “acquiring friends and making
alliances” and another on “helping non-kin.” And finally, to capture the
uniquely human dispositions for acquiring complex forms of culture, we included
“seeking education or culture” and “building, creating, or discovering
something.”
When we submitted
scores on the twelve separate motives to factor analysis, five main factors
emerged:
Social Dominance, Constructive Effort, Romance, Subsistence, and Nurture. Seeking
wealth, power, and prestige all have strong positive loadings on Social
Dominance, and helping non-kin has a moderate negative loading. (That is,
helping non-kin correlates negatively with seeking wealth, power, and
prestige.) Constructive Effort loads most strongly on the two cultural motives,
seeking education or culture, and creating, discovering, or building something,
and it also has substantial loadings on two pro-social or affiliative motives:
making friends and alliances and helping non-kin. Romance is a mating motive,
chiefly loading on short-term and long-term mating. Subsistence combines two
motives: survival, and performing routine tasks to gain a livelihood. Nurture
loads most heavily on nurturing/fostering offspring or other kin, and that
motive correlates negatively with short-term mating. Helping non-kin also loads
moderately on this factor, bringing affiliative kin-related behavior into
association with generally affiliative social behavior.
Male and female
antagonists both display a pronounced and exclusive emphasis on Social
Dominance (figure 1). Male protagonists score higher than any other character
set on Constructive Effort and on Subsistence. Female protagonists score higher
than any other character set on Romance, but their positive motives are fairly
evenly balanced among Constructive Effort, Romance, and Nurture. In these
novels, female protagonists are largely restricted to the nubile age range.
That restriction corresponds with a pronounced emphasis on Romance as a motive.
The opposition between
dominance and affiliation in the novels can clearly be linked to a robust and
often replicated finding in psychological studies of motives and personality. Summarizing research into basic motives, David
Buss observes that in cross-cultural studies the two most important dimensions
of interpersonal behavior are “power and love” (“Evolutionary Psychology” 21).
Surveying the same field and citing still other antecedents, Paulhus and John
observe that in debates about “the number of important human values,” there are
two, above all, that are “never overlooked” (1039). They designate these values
“agency and communion” and associate them with contrasting needs: the need for
“power and status” on one side and for “approval” on the other (1045).[1]
Paulhus and John link the
contrasting needs for power and approval with contrasting forms of bias in
self-perception. “Egoistic” bias attributes exaggerated “prominence and status” to one’s self, and “moralistic” bias
gives an exaggerated picture of one’s self as a “nice person” and “a
good citizen” (1045, 1046). Adopting these terms, we can say decisively that
the novels in this study, taken collectively, have a moralistic bias. In
protagonists, striving for personal predominance is strongly subordinated to
communitarian values. Protagonists and their
friends typically form communities of affiliative and cooperative behavior, and
antagonists are typically envisioned as a force of social domination that threatens the very principle of community.
If agonistic structure virtually personifies the two
most important dimensions of interpersonal behavior, it clearly runs very deep
in human nature. It is not peripheral to the worlds imagined in these novels.
It is the figure in the carpet, the central design that radiates out to the
whole image, giving order and meaning to the details.
A disposition for thinking agonistically is evidently
an important feature of human nature. We take it as a basic working hypothesis
that any important feature of human nature must enter into fictional worlds in
some fashion. How it enters into fictional worlds is a question for
empirical research. The findings from Paulhus and John on egoistic and
moralistic bias offer a highly suggestive clue about how agonistic structure
enters into the depiction of characters in fiction. Needs for power and
approval are clearly linked with dominance and affiliation. Bias in
self-perception is a form of imagination. It is an imaginative distortion of
reality prompted by basic, powerful needs. Could it be that this kind of
distortion also forms a central organizing principle in imaginative fictions?
Could some characters be exaggerated images of “good” people, fulfilling
moralistic fantasies? And could other characters be exaggerated images of powerful
people, fulfilling egoistic fantasies? We think this is in fact the case. In
the next chapter, the dichotomy between moralistic and egoistic bias provides a
chief conceptual framework for our analysis of individual characters.
Choosing a Partner
Most of the novels in this
study are “love stories.” That is, the plots usually include making choice of a
marital partner. Evolutionary psychologists have identified mating preferences
that males and females share and also preferences in which they differ. Males
and females both value kindness, intelligence, and reliability in mates. Males
preferentially value physical attractiveness, and females preferentially value
wealth, prestige, and power. These sex-specific preferences are rooted in the
logic of reproduction. Physical attractiveness in females correlates with youth
and health in a woman—hence with reproductive potential. Wealth, power, and
prestige enable a male to provide for a mate and her offspring. [2] We
anticipated that scores for mate selection would correspond to the differences
between males and females found in studies of mate selection in the real world.
Since protagonists typically evoke admiration and liking in readers, we
anticipated that protagonists would give stronger preference than antagonists to
intelligence, kindness, and reliability. We reasoned that a preference for
admirable qualities in a mate would evoke admiration in readers.
We asked questions about
selecting mates in both the short term and the long term. In the results of the
factor analyses for mate selection, the loadings for short-term and long-term
mating are almost identical and divide with the sharpest possible clarity into
three distinct factors: Extrinsic Attributes (a desire for wealth, power, and
prestige in a mate), Intrinsic Qualities (a desire for kindness, reliability,
and intelligence in a mate), and Physical Attractiveness (that one criterion by
itself).
We anticipated differences in mate preferences
in the short and long term, but our respondents evidently read the question on
short-term mating to mean something different from what we had in mind. We had
in mind illicit sexual activity. But respondents gave scores on short-term
mating to many characters who do not engage in illicit sex. In many cases, the
respondents evidently interpreted short-term mating to mean any romantic
excitement in its early phases, even for relations that eventually culminate in
marriage. The scores on selecting mates in the short and long term are
essentially equivalent. We give the results here only the long term, but the profiles
of individual characters in the next chapter include instances of the short
term.
Female protagonists and
antagonists both give a stronger preference to Extrinsic Attributes—wealth,
power, and prestige—than male protagonists or antagonists, but female
antagonists exaggerate the female tendency toward preferring Extrinsic
Attributes (figure 3). The emphasis female antagonists give to Extrinsic
Attributes parallels their single-minded pursuit of Social Dominance. Female
protagonists give a more marked preference than male protagonists to Intrinsic
Qualities—intelligence, kindness, and reliability.
We did not anticipate that
male protagonists would be so strongly preoccupied with Physical Attractiveness
relative to other qualities, nor did we anticipate that male antagonists would
be so relatively indifferent to Physical Attractiveness. The inference we draw
from these findings is that the male desire for physical beauty in mates is
part of the ethos the novels. It is part of the charm and romance of the
novels, part of the glamor. Male antagonists’ relative indifference to Physical
Attractiveness seems part of their general indifference to affiliative
relationships.
If one were to look only
at the motive factors, one might speculate that male antagonists correspond
more closely to their gender norms than female antagonists do. Male antagonists
could be conceived as personified reductions to male dominance striving. The
relative indifference male antagonists feel toward any
differentiating features in mates might then look like an exaggeration of the
male tendency toward interpersonal insensitivity. Conceived in this way, male
antagonists would appear to be ultra-male, and female antagonists, in contrast,
would seem to cross a gender divide. Their reduction to dominance striving
would be symptomatic of a certain masculinization of motive and temperament.
They would be, in an important sense, de-sexed. Plausible as this line of
interpretation might seem, it will not bear up under the weight of the evidence
about male antagonists' relative indifference to Physical
Attractiveness in a mate. Like female antagonistic dominance striving, that
also is a form of de-sexing. Dominance striving devoid of all affiliative disposition
constitutes a reduction to sex-neutral egoism. The essential character of male
and female antagonists is thus not a sex or gender-specific tendency toward
masculinization; it is a tendency toward sexual neutralization in the isolation
of an ego disconnected from all social bonds.
Agonistic structure
evidently works powerfully enough so that it turns one set of major characters,
antagonists, into personifications of Social Dominance, the basic motivational
component in agonistic structure. Hardly any feature of human nature, one might
think, would run deeper than sex. Along with parenting, sex is the one feature
of human nature that most closely links “proximal” motives, active desires,
with the “ultimate” regulative principle of inclusive fitness. Hence the
sometimes obsessive, single-minded preoccupation that evolutionary
psychologists have displayed in the scientific study of human mating. Any
imaginative construct powerful enough to neutralize sex in a whole class of
major characters surely constitutes a central organizing principle in the
novels.
Developing a Personality
Repudiating a conventional
distinction between “the novel of character” and “the novel of incident,” Henry
James poses a set of rhetorical questions. “What is character but the
determination of incident? What is incident but the illustration of character?
What is either a picture or a novel that is not of character? What else
do we seek in it and find in it?” (55). In The Mayor of Casterbridge,
more succinctly but in an equally emphatic way, Hardy (following Novalis)
declares, “Character is Fate” (ch. 17). In what then does character consist? Current
research into personality commonly distinguishes five broad factors. Extraversion
signals assertive, exuberant activity in the social world, versus a tendency to
be quiet, withdrawn and disengaged. Agreeableness signals a pleasant,
friendly disposition and a tendency to cooperate and compromise, versus a
tendency to be self-centered and inconsiderate. Conscientiousness refers
to an inclination toward purposeful planning, organization, persistence, and
reliability, versus impulsivity, aimlessness, laziness, and undependability. Emotional
Stability reflects a temperament that is calm and relatively free from
negative feelings, versus a temperament marked by extreme emotional reactivity
and persistent anxiety, anger, or depression. Openness to Experience
describes a dimension of personality that distinguishes open
(imaginative, intellectual, creative, complex) people from closed
(down-to-earth, uncouth, conventional, simple) people.[3]
Basic motives are the
building blocks of human life history—the desire for physical satisfactions and
the avoidance of pain and danger; the desire for sex and companionship, the
love of children, the need for social bonding, the hunger for status, the
impulse to acquire information and to engage in imaginative play. Personality
traits are dispositions for acting on such motives. They vary considerably
between individuals and remain relatively stable within individuals. They are
thus chief distinguishing features of individual identity. When we speak of
“human nature,” we focus first of all on “human universals,” on cognitive and
behavioural features that everyone shares. We typically use personality, in
contrast, to distinguish one person from another—for example, a friendly,
careless extravert in contrast to a cold, conscientious introvert. In reality,
personality factors are themselves human universals, integral parts of our
common human nature. Each of the five factors has a common substratum. Individuals
differ only in degree on each factor (Nettle, Personality). There is no
individual, for instance, who does not display some degree of “Extraversion”—no
human black hole, so psychologically dense that no matter can escape from it.
There is, it is true, a severe genetic abnormality, a medical condition, that
renders some very small number of individuals incapable of sensing pain. Such
individuals do not live to maturity—they get too physically damaged. For the rest
of us, some degree of “Neuroticism,” the inverse of “emotional stability,” is
an unfortunate necessity, part of the cost of living in this our vale of tears.
The underlying commonality
in Extraversion/Introversion is the necessity to engage in some way with an
external environment. We can measure the degree to which organisms move outward
to engage with the world or retreat into the inner self. Some authors give us
much of the full particularity of the world, and others give us generously of
their own inner visions. Hawthorne greatly admired Trollope for the vivid
verisimilitude in his depictions of ordinary life—“just as real as if some giant
had hewn a great lump out of the earth and put it under a glass case, with all
its inhabitants going about their daily business, and not suspecting that they
were made a show of " (cited in Trollope 122-23).
Agreeableness is a measure
of affiliative sociality, and since humans are social animals, most humans have
some measure of affiliative sociality. Extreme defects in Agreeableness are measures
of sociopathic personality disorders. (Chekhov’s story “The Darling” offers an
affectionate and humorously condescending portrait of an opposite extreme, a
character constituted exclusively by empathic sensibility.) Novelists depict
human communities, and they also tacitly fashion communities with their
readers. In negotiating our place in any community, we must balance off our
dispositions for egoistic striving and dominance on the one side with our needs
for affiliative sociality on the other. Novelists depict these negotiations and
in the very act of depicting them enter into parallel negotiations with their
readers.
Conscientiousness is a
measure of any given person’s disposition for organizing, planning, and
carrying through on the tasks of life. Locating present action within a
temporal continuum containing past and future is part of the specifically human
cognitive apparatus (Darwin, The Descent 1: 88-89). Without some measure
of conscientiousness, a person could not function at all. As we have argued
throughout this study, motives and plots are organized around goal-directed
actions. The ability to sustain an effort toward achieving a goal is a central
feature of Conscientiousness. Character is in this respect integral with
narrative form. Carelessly spontaneous personalities can stand in as
protagonists in special cases, usually comic, but as a rule, insofar as
protagonists converge in temperament with the imaginative spirit in which a
work is written, they display powers of organization like those that make it
possible to produce any work at all. No wonder, then, that protagonists on
average rate high in Conscientiousness. (In this respect, David Copperfield,
Amy Dorrit, and Esther Summerson are more typically Dickensian protagonists
than Pip.)
Emotional Instability is a
measure of emotional reactivity in the range of negative affect. Emotional
reactivity varies in intensity from individual to individual, but experiencing
pain is normal and necessary. Without fear and sorrow, people would have no
means of registering dangers or feeling the sense of loss. The ability to
experience emotional pain, like the ability to experience physical pain, is an
indispensable adaptive trait. In authors like Hardy, the ability to feel pain
acutely and in a prolonged, chronic way offers imaginative access to one of the
chief dimensions of imaginative life—the dimension of tragedy.
Openness to Experience registers
curiosity and a readiness to absorb experience of an imaginative, intellectual,
and aesthetic character. That sort of readiness is often foregrounded by the
novelists in our study. From Austen to Dickens, from Eliot to Hardy, the hunger
for “education” and “culture” distinguishes the inner protagonistic circle. Our
readiness for culture—our disposition for producing emotionally charged
symbolic forms—is the single most important feature of human nature that
distinguishes us as a species from all other species (Carroll, “An Evolutionary
Paradigm,” “Rejoinder,” Literary Darwinism 197-203; Deacon 21-22; Dissanayake,
Art; Dutton; Mithen; Panksepp and Panksepp; Tomasello et al.; Tooby and
Cosmides, “Does Beauty Build?”; Wade; E. O. Wilson, Consilience). Consequently,
in making Openness to Experience a distinctively protagonistic attribute, the
authors in our study are not only reflecting a local norm of English and
European civilization in the nineteenth century. They are presenting heroic
images of human nature itself. Depicting this sort of heroism is not mere
sentimentalism or idealism. It is a form of realism that reaches deeper than
“social realism.” It reaches into our inmost character as a species.
In conclusion, then, if we
strip away the now standard triad of race, class, and sex, what is left? More
than has been taken away. Beneath ethnic and class identity, beneath even the
two basic human morphs of male and female, there are elemental features of
human nature, the bedrock of personal identity. The composition of that bedrock
can be assessed with the five factors of personality: the biologically elemental
interaction between an organism and its environment; the capacity of all higher
organisms to feel pain and react against it; the disposition of all mammals for
affiliative bonding; and the specifically human capacities for organizing
behavior over time, carrying out plans, and generating imaginative culture.
The conceptual scope of
the five personality factors, transcending as it does the standard elements of
identity politics, has an important bearing on the question we pose at the
beginning of this chapter. If personality should prove a key element in
agonistic structure—if it were to correlate strongly with other attributes of
characters, the emotional responses of readers, and agonistic role
assignment—that correlation would give strong evidence that agonistic structure
is in fact a central organizing principle in the novels, not a wandering
asteroid, but a gravitational center for the whole system of meaning in the
novels.
Assuming that protagonists
and their friends would be generally affiliative, we predicted that
protagonists and their friends would on average score higher on the personality
factor Agreeableness. Assuming that protagonists would be more closely aligned
than other characters with the perspectives of authors, we predicted that that
protagonists would score higher than antagonists and minor characters on the
personality factor Openness to Experience. We made no specific predictions on
the other factors.
Female protagonists score
higher than any other set on Agreeableness, Conscientiousness, and Openness,
and they score in the positive range on Stability (figure 6). Male protagonists
look like muted or moderated versions of the female protagonists. The
personality profiles of male and female antagonists are very similar to one
another—both somewhat extraverted, highly disagreeable, and low in Stability
and Openness. Female antagonists are somewhat more conscientious than male
antagonists.
In the value structures
implicit in the organization of characters in agonistic structure, Introversion,
Agreeableness, Conscientiousness, Stability, and Openness are all positively
valenced features. Agreeableness is the most strongly marked part of this
normative array. Being agreeable is a trait that distinguishes good characters
generally, but being conscientious and open to experience are more specifically
characteristic of protagonists. With respect to personality, female
protagonists are clearly the normative character set.
The total profile for
protagonists is that of quiet, steady people, curious and alert but not
aggressive, friendly but not particularly outgoing. The antagonists, in
contrast, are assertive, volatile, and unreliable, but also dull and
conventional. Openness would be associated with the desire for education or
culture and with the desire to build, discover, or create, and that whole
complex of cognitive features is one of the two basic elements in Constructive
Effort. As one would anticipate, then, Openness correlates with Constructive
Effort (r = .41). The main antagonistic motive factor is Social
Dominance, which correlates strongly and negatively with Agreeableness (r =
-.54). Antagonists score in the extreme range both on Agreeableness (negatively)
and on Social Dominance (positively).[4]
This suite of strong correlations,
keying into the chief motive factors distinguishing protagonists and
antagonists, emphatically demonstrates that personality is integral to
agonistic structure. If personality is also central to individual identity, and
if individual identity is a central principle in the creation of characters in
novels—as surely most readers will grant—then agonistic structure must be a
central organizing principle in the novels.
Our study is designed to
register only differences in personality among characters, but the categories
of personality are just as useful for distinguishing the personal identity of
novelists, and the personal identity of the author enters deeply into the
meaning of novels. Novelists are almost necessarily intuitive psychologists
capable of discriminating fine shades of difference in qualities of temperament.
Readers employ similar powers for identifying the qualities of personality that
help distinguish one novelist from another. Personality enters into the
concrete particularity of external reality depicted in novels (extraversion)
and the depth of inward experience evoked in characters (introversion); the
warmth and humanity of some novelists (Trollope, say), and the relative
coldness or antagonism of others (Flaubert, for instance); the degree to which
“duty” or “integrity” influences an author’s norms of judgment for his or her
characters (so prominent, for instance, in George Eliot); the capacity for
suffering (a signal quality in Hardy and in all the Brontës); the willingness
of an author to violate conventional expectations (as Jane Austen so seldom
does, and Emily Brontë so often does), and the scope of cultural interest
reflected in the author’s range of reference (as in the vast repertory of
metaphoric allusion forming the web of motifs in Middlemarch). In
discussing novels and novelists, in chapters three, eight, and nine, we draw on
such concepts for descriptive and analytic purposes.
We
have to register one important limitation in the instrument we used for
obtaining scores on personality factors for our characters. The larger the
instrument, the higher the levels of resolution. We toyed with the idea of
using a very large instrument but, with regret, rejected that idea. The larger
the instrument, the higher the level of resolution, but also, the greater the
drain on the limited patience of respondents to an online questionnaire. We
chose then an instrument that is quite short, consisting of only ten items (two
for each factor), the Ten Item Personality Inventory (TIPI, constructed by Gosling,
Rentfrow, and Swann). This instrument has proven empirical validity within its
range, but in the nature of the case, some resolution is lost, and some
adjustment has to be made. For our purposes, the most important such adjustment
concerns the personality factor Openness to Experience. For Openness, the two
items on which characters received scores are "Open to new experiences,
complex" and "Conventional, uncreative.” Someone not already familiar
with the broad construct Openness to Experience, as it is formulated, for
instance, by Costa and McCrae, might think that the term
"conventional" signifies conventional social attitudes and not
consider the term in its cognitive aspect, as lack of imagination and
intellect. The scores on Openness for a number of individual characters
strongly suggest that in interpreting the term “conventional,” our respondents
did in fact tend to focus more on social attitudes than on cognitive interests.
For instance, from among Jane Austen’s protagonistic characters, the following characters
all receive relatively low scores—average or below average—on Openness to
Experience: Elinor Dashwood from Sense and Sensibility, Fanny Price from
Mansfield Park,
Fitzwilliam Darcy from Pride and Prejudice, and George Knightley from Emma.
Even Anne Elliott from Persuasion scores only in the moderate range
(.36). All of these characters are intelligent and cultivated but also have
relatively conservative views toward social behavior. Darcy scores below
average (-.15) on Openness, but he is described as “clever,” and he devotes
great care and attention to enhancing his library. The relatively heavy
weighting of social attitudes in the scores on Openness can be illustrated by
comparing Anne Elliott’s score with that of Heathcliff from Wuthering Heights. Heathcliff’s
score on Openness (.37) is almost identical to Anne’s, but Heathcliff has only
a rudimentary education and displays no general cultural curiosity. He is
merely indifferent to conventional standards of conduct. Anne is quiet and
relatively conventional in social attitudes, but she is also highly educated,
and she engages in serious, interesting conversations on literary subjects.
In
compensation for having less resolution in the measurement of cultural
curiosity, we gain more resolution in the measurement of cultural conservatism.
The analytic utility of that feature of this particular instrument makes itself
felt with particular force in our commentary of the relations between
personality and ideology in the novels of Jane Austen.
Getting Emotional
Human behavior is
organized through motives—goal-directed action prompted by needs rooted in the
adaptive history of the species. Sex is a motive, and we seek mates. Social
affiliation is a motive, and we seek friends and seek to make alliances.
Nurturing offspring is a motive, and we seek to provide food, shelter, and
education for our children. The most immediate, proximal mechanism for
activating motives are emotions—feeling states that are caused and accompanied
by physiological and neurochemical changes manifesting themselves, on the
phenomenal level, as qualities of sensation. Emotions prompt to action, and
they also form the basis for our evaluative responses in social relations.
Those evaluative responses guide our own social behavior and help us to
organize our goal-directed actions. (For various definitions of emotion see
Plutchik, Emotions 18-19.)
One of the chief working
hypotheses in this study is that when readers respond to characters in novels,
they respond in much the same way, emotionally, as they respond to people in
everyday life. They like or dislike them, admire them or despise them, fear
them, feel sorry for them, or are amused by them. In writing fabricated
accounts of human behavior, novelists select and organize their material for
the purpose of generating such responses, and readers willingly cooperate with
this purpose. They participate vicariously in the experiences depicted and form
personal opinions about the qualities of the characters. Authors and readers
thus collaborate in producing a simulated experience of emotionally responsive
evaluative judgment.[5]
We sought to identify
emotions that are universal and that are thus likely to be grounded in
universal, evolved features of human psychology. Emotions at that conceptual
level would be on the same level as the basic motives extrapolated from human
life history. Over the past several decades, adaptationist psychologists have
made substantial progress in identifying basic emotions. Much of this work was
pioneered by Paul Ekman. By isolating emotions that can be universally or
almost universally recognized from facial expressions, Ekman and other
researchers ultimately produced a core set of seven “basic” emotions: anger,
fear, disgust, contempt, joy, sadness, and surprise. Different researchers
sometimes use slightly different terms, register different degrees of intensity
in emotions (for instance, anxiety, fear, terror, panic), organize the emotions
in various patterns and combinations, or link them with self-awareness or
social awareness to produce terms like embarrassment, shame, guilt, and envy (A.
Buss; Lewis; Panksepp 143). This core group of seven emotions nonetheless has
widespread support as a usable taxonomy of basic emotions (Lewis and
Haviland-Jones; Plutchik, Emotions).
Because emotions activate
goal-directed action, they can be characterized through their relations to
desired objects or states of being (Ekman, “Basic Emotions” 46, Emotions;
Nettle, Personality 85-90, 107-13). Fear, anger, disgust, and contempt are
all aversive emotions; they activate movements of negation or escape against
threats or obstacles to desired objects or states. Fear signals danger and
activates responses such as flight, hiding, or submission. Anger is a response
to inhibiting force and activates hostile corrective action. Disgust manifests
repugnance at putrid and toxic substances. Contempt is closely related to
disgust but combines it with a sense of moral and/or social disdain. Contempt
displays social disaffection joined with a feeling of superiority to any
potentially threatening force in the object of dislike. Sadness and enjoyment,
though differently valenced, both have a direct relation to a desired object or
state. Sadness activates psychological mechanisms for coping with loss, for
instance, the loss of a loved one, of social status, or of a cherished
possession. Enjoyment registers pleasure and satisfaction that draws us toward
objects or keeps us focused on them. As Ekman puts it, the enjoyable emotions
“motivate our lives” (Emotions 199). They direct us toward things that
we most value and most seek—food and physical comforts, mates, our children and
loved ones, our friends, wealth and property, social status, knowledge, the
arts, and amusements—all the activities that engage our powers, advance our
interests, and fulfill our needs. Surprise, unlike the other basic emotions,
has no intrinsic valence. It is a response to the unexpected and brings
attentive curiosity into play. The response to surprise is to pause and
re-orient one’s self. Surprise is usually brief and is preliminary to the
valenced responses of aversion and seeking in the other emotions (Frijda 63;
Plutchik, Emotions 109.)[6]
The questionnaire
contained a list of ten emotional responses. To produce this list, we started
with the core of seven terms from Ekman and adapted them for registering graded
responses specifically to persons or characters. We used four of the seven
terms unaltered: anger, disgust, contempt, and sadness. We also retained fear
but divided it into two distinct items: fear of a character, and fear for
a character. Ekman observes that the positive emotions have been less carefully
observed and differentiated than the emotions that reflect emotional upset. The
simple terms “joy” or “enjoyment” cover a wide spectrum of possible pleasurable
or positive emotions, ranging from “amusement” to “schadenfreude” to “bliss” (Emotions
191). In adapting the term “joy” or “enjoyment,” we sought to register some
qualitative differences and also devise terms appropriate to responses to a
person. We chose three terms: liking, admiration, and amusement. “Liking” is an
emotionally positive response to a person, but it does not contain a specific
element of approval or disapproval. “Admiration” combines positive emotionality
with a measure of approval or respect (Darwin, The Expression 289;
Dutton 190-92; Plutchik, “The Nature” 349). By itself, “surprise,” like “joy,”
seems more appropriate as a descriptor for a response to a situation than to a
person. Consequently, we did not use the word “surprise” by itself. Instead,
along with “admiration,” we used “amusement,” which combines the idea of
surprise with an idea of positively valenced emotionality. “Amusement” extends
emotional response to take in responses to comedy. Sadness and fear take in
responses to tragedy; and anger and contempt, mingled with amusement, take in
responses to satire.
The
questionnaire includes one further term: indifference. A number of researchers
have included a term such as “interest” to indicate general attentiveness, the
otherwise undifferentiated sense that something matters, that it is important
and worthy of attention. “Indifference” can be regarded as the inverse of
“interest.” “Indifference” provides a qualitatively neutral measure of intensity
in emotional reaction to a character.
We predicted (1) that
protagonists would receive high scores on the positive emotional responses
“liking” and “admiration”; (2) that antagonists would receive high scores on
the negative emotions “anger,” “disgust,” “contempt,” and “fear of” the
character; (3) that protagonists would score higher on “sadness” and “fear-for”
the character than antagonists; and (4) that major characters (protagonists and
antagonists) would score lower on “indifference” than minor characters.
Factor analysis produced
three clearly defined emotional response factors: (1) Dislike, which includes
anger, disgust, contempt, and fear of the character, and which also
includes negative correlations with admiration and liking; (2) Sorrow, which
includes sadness and fear for the character and a negative correlation
with amusement; and (3) Interest, which consists chiefly in a negative
correlation with indifference.
Male and female
protagonists both score relatively low on Dislike and relatively high on Sorrow
(figure 1). Male and female antagonists score very high on Dislike—higher than
any other set—low on Sorrow, and somewhat above average on Interest. Female
protagonists score high on Interest, but male protagonists, contrary to our
expectations, score below average on Interest. They score lower even than good
minor males, though not lower than the other minor characters.
Once one has isolated the
components of agonistic structure and deployed a model of reading that includes
basic emotions as a register of evaluatively polarized response, most of the
scores on emotional response factors are predictable. There is, however, one
surprising and seemingly anomalous finding that emerges from the scores on
emotional responses—the relatively low score received by male protagonists on
Interest. This finding ran contrary to our expectation that protagonists, both
male and female, would score lower on indifference than any other character
set. We think this finding can be explained by the way agonistic polarization
feeds into the psychology of cooperation. Male protagonists in our data set are
relatively moderate, mild characters. They are introverted and agreeable, and
they do not seek to dominate others socially. They are pleasant and
conscientious, and they are also curious and alert. They are attractive
characters, but they are not very assertive or aggressive characters. They
excite very little Dislike at least in part because they do not excite much
competitive antagonism. They are not intent on acquiring wealth and power, and
they are thoroughly domesticated within the forms of conventional propriety.
They serve admirably to exemplify normative values of cooperative behavior, but
in serving this function they seem to be diminished in some vital component of
fascination, some element of charisma. They lack power, and in lacking power,
they seem also to lack some quality that excites intensity of interest in
emotional response.
In
these novels, the aggressive pursuit of Social Dominance—wealth, prestige, and
power—is morally demonized. The desire for Social Dominance is overwhelmingly
the single most distinctive motivational trait of both male and female
antagonists. That motivational trait correlates with low scores on the
affiliative personality factor Agreeableness and high scores on the emotional
response factor Dislike. Despite this strongly valenced cluster of
correlations, male and female antagonists score higher on Interest (lower on
indifference) than male protagonists. Readers dislike antagonists, but it is
sometimes the case that antagonists are more exciting than protagonists,
especially male protagonists. We did not anticipate this finding, but it led us
a good ways toward our ultimate conclusions about the socially adaptive
functions of agonistic structure. The evolved and adapted dispositions for
suppressing dominance is evidently so powerful that it partially suppresses the
natural tendency readers have to be particularly interested in protagonists,
male as well as female. The kind of fascination exercised by egoists indulging
their drive for power runs counter to the needs of the egalitarian ethos, and
in the novels in this study, taken as a whole, the egalitarian ethos evidently
gets the final say. It is the egalitarian ethos, and not the fascination with
power, that regulates the motivational profile of male protagonists. That
profile drives the emotional responses of readers, but the needs and
expectations of readers also exercise a strong influence on how authors depict
characters. This reciprocal causation is the topic of the next section.
A Circulatory System for a Social Ethos
Because the features attributed to characters correlate
with role assignments and with the emotional responses of readers, we can infer
relations between the attributes of characters and the outlook of authors and
readers. On the average, the features that distinguish “good” and “bad”
characters, and especially the features that distinguish protagonists and
antagonists, reflect the positive and negative values authors have invested in
their characters and anticipate their readers will share. In general, good
characters, and especially protagonists, reflect the normative values of the
novels—values shared by authors and their readers. Bad characters reflect the
inverse of normative values. Normative values provide a common frame of
reference for authors and their readers. 
The causal sequence in the diagram in figure 1 forms a
feedback loop. The designs or intentions of the author, in the top left-hand
corner of the diagram, serve as a starting point in the causal sequence, and
the end-point in the sequence, the ethos of a given culture, feeds back into
the designs or intentions of the author.
Authors and readers can
and often do operate in tension with the normative values of their culture, but
if authors are to communicate at all, their own idiosyncratic dispositions must
define themselves in relation to the shared framework. By comparing the scores of individual novels with those of the larger
data set, we can identify the features that distinguish one novel or novelist
from another. In chapter eight, for instance, we examine the way scores for
Austen’s male protagonists deviate from the average for the whole set of
novels. We also analyze the way scores on emotional response differ between her
novels and the other novels. By reflecting on the comparisons, we draw
inferences about basic thematic and tonal features of her imagined world.
Novelists
designate characters as persons intent on achieving goals. The success or
failure of the character in achieving goals is the main action in the
story—broadly, the “plot.” Goals are the end-objects of motives—for instance,
the desire to survive, to get married, to make friends, to obtain education, or
to assist one’s friends. Readers recognize characters as agents with goals and
have emotional responses to the characters. In an obvious sense, an author is
the first causal force in this sequence. The author creates characters and
designates their features and fortunes. For a main character, the
author fabricates a situation, identifies the hopes and fears of the
character, invents a sequence of actions organized around those hopes and
fears, and determines the outcome for that sequence of actions. In all of this, outside of recognizing what
the author has stipulated, the reader has no part. The reader must take it as
the author gives it. But in giving it, the author does not neglect to consult
the reader, at least prospectively. The author anticipates the effects that his
or her designs will have on the minds and emotions of readers.[7]
Despite
the power exercised by authors, the causal force between an author and his or
her readers does not move in only one direction, from author to readers. In anticipating
the effects that their designations will have on readers, authors are
themselves the cunning servants of their readers. They are themselves
constrained in constructing meaning by their own sense of what readers expect
and demand. Dickens’ revision of the end of Great Expectations offers a
case in point. Having done a little judicious pre-publication market testing by
consulting a savvy friend, Dickens decided that the original, unhappy
conclusion he had written for his novel would not sell nearly so well as a
hopefully upbeat ending, and he changed the ending accordingly. The author’s
ability to manipulate the responses of readers depends on keeping his or her
depictions within the range of the readers’ expectations or desires. Authors
rule, but only because they provide their subjects with what the subjects want.
Authors dominate the feelings and thoughts of the audience, but only because
they allow the feelings and thoughts of the audience at least partially to
determine the parameters within which the authors work.
Authors
of a canonical standing on a level with that of the authors sampled in this
study do not just passively reflect the established and conventional values and
beliefs of their culture. That conception of the inert passivity of the
authorial mind is, in our view, an important limitation in Foucauldian cultural
theory and the New Historical literary criticism that flows from it. Great
writers tap into the deepest levels of the human psyche, connect their
contemporary cultural forms with basic human passions, and give their own
idiosyncratic and distinctive stamp to the world they envision. Despite his
willingness to play to his audience, Dickens is still “the inimitable Dickens.”
Great and original authors create new possibilities of understanding, but no
matter how original and independent they might be, all authors feed off of the
meanings that are available within their culture: the literary forms and
traditions with which they work, and the forms of cultural imagination—ideological,
religious, and philosophical—in which they participate. Authors, readers, and
the larger culture are all locked into a reciprocal and interdependent
relationship. The arrows in the diagram leading back from a cultural ethos to
the designs of an author, and thus closing the feedback loop, are intended to
reflect that reciprocal relationship.
Conclusion: Mirror, Mirror, on the Wall
Agonistic structure in
these novels clearly serves as a central organizing principle. It is a center
of gravity for the whole solar system constituted by motives, personality
traits, mate selection, and emotional responses. The characters display an
integrated array of agonistically polarized attributes, and readers respond to
those attributes in emotionally polarized ways. The antagonists display a
single-minded preoccupation with wealth, prestige, and power—egoistic striving
wholly segregated from social affiliations. That motive profile extends itself
into their criteria for choosing mates. Male antagonists have no particular
preferences in mates, and female antagonists seek only to marry for wealth and
status. The sociopathic dispositions revealed in motives and mating correspond
to low scores on the personality factor Agreeableness. Antagonists are both
emotionally isolated and also incurious. They are interested in nothing except
enhancing their power and prestige. The protagonists, in contrast, care about
friends and family, respond to romantic attractions, and become readily
absorbed in cultural pursuits. They are affectionate, reliable, and open to
experience. They are also on average younger and more physically attractive
than antagonists. Agonistic structure thus presents a sharply etched
picture—youth, beauty, and positive emotional energy meeting resistance and opposition
from malevolent forces seeking only personal domination for its own sake. The polarized
emotional responses of readers correlate strongly with this integrated array of
attributes. Readers respond with aversion and disapproval to antagonists and
with admiration and sympathy to protagonists.
Humans share with amoebas a fundamentally
dichotomized orientation to the world. “Approach” and “avoidance” are the two
mechanisms that govern an amoeba’s activity. Chemical signals direct it to
approach nutrients and to retreat from toxins. People do the same thing. They
approach those things—food, sex, warmth, friends, status—that make them feel
good, and they turn away with aversion from those things that make them feel
bad. Egoistic displays of dominance evidently have a toxic impact on the nervous
system of our respondents.[8]
Amoebas react. Humans react and judge. As the scores
on emotional responses indicate, judgments can be complex, nuanced, ambivalent.
Even so, those complications are only that, complications. They work variations
on a basic theme, and that theme is polarized evaluative response. As a team of
personality psychologists led by Gerard Saucier explains, in many contexts, across
a diverse array of concerns, psychologists identify “a global evaluation factor
(good vs. bad)” (857). When personality psychologists use statistical
techniques to reduce multiple personality attributions to superordinate
factors, they can choose the number of factors to extract. If only one factor
is extracted, that factor constitutes a “contrast between desirable and
undesirable qualities” (Saucier and Goldberg 854). Scores on motives, mating,
and personality reveal in detail what counts as desirable and undesirable
qualities in characters. Scores on emotional responses lock down these
evaluative judgments by placing them in the court of first and last appeal—the
court of actual feeling.
Most of the novels
included in this study are “classics. Classics gain access to the deepest
levels of human nature. They evoke universal passions and fulfill deep
psychological needs, but they do not always produce mimetically accurate
representations of human nature. They hold a mirror up to nature, but this
mirror, unlike that in Snow White, is under no obligation to tell the
simple, unvarnished truth. The images produced are filtered through an
imaginative lens that adds its own twist to the images it reflects. In the
novels in this study, agonistic structure creates a virtual imaginative world
designed to give concentrated emotional force to the clash between dominance
and affiliation. That imaginative virtual world provides a medium in which
readers participate in a shared social ethos. The social ethos shapes agonistic
structure, and agonistic structure in turn feeds back into the social ethos,
affirming it, reinforcing it, integrating it with the changing circumstances of
material and social life, and illuminating it with the aesthetic, intellectual,
and moral powers of individual artists.