Motivation and Emotion, Volume 1, Number 3, 1977

On Falling in Love in Conformance with the

Romantic Ideal'

James R. Averill 2 and Phyllis Boothroyd

University o f Massachusetts, Amherst

Eighty-five subjects (single, married, and divorced) completed an anony-

mous questionnaire describing their attitudes toward, and experiences with,
romantic love. Included in the questionnaire was a description of the
romantic "'ideal, "" and subjects were asked to rate how closely their most
and least intense experiences o f love conformed to this paradigm case. Ap-
proximately equal numbers o f subjects indicated that their most intense ex-
perience either did (N = 34) or did not (N = 33) conform closely to the
romantic ideal. The conformity did not vary as a function o f age, sex, or
marital status. There was a small but significant (.31) correlation between
favorable attitudes toward the romantic ideal and having an experience that
conformed to that ideal. The circumstances most frequently mentioned as
conducive to falling romantically in love were the appearance of the loved
one and a chance meeting. The results are discussed in terms of the sociiTl
role o f romantic love.

Love is a passion: One does not enter into love quietly and with delibera-
tion; rather, one is " g r i p p e d , " "seized," and " o v e r c o m e " by love. That, at
least, is the romantic stereotype or ideal. 3 But how many people actually
conform to this ideal? And what are the circumstances conducive to falling
in love? For example, is love primarily a matter for youth, a kind o f adoles-
cent infatuation, or do older persons also fall in love? H o w does a person's

~This research was supported, in part, by a grant (MH 22299) from the National Institute of
Mental Health.
2Address all correspondence to Dr. James R. AveriU, Department of Psychology, University of
Massachusetts, Amherst, Massachusetts 01002.
3In speaking of the romantic "ideal," we are not making a value judgment; rather, we are
simply referring to the form of love frequently depicted in song and literature, and which will
be described more fully below.
235
This journal is copyrighted by Plenum. Each article is available f o r $7.50 f r o m Plenum Pub-
lishing Corporation, 2 2 7 West 17th Street, New Y o r k , N.Y. 10011.
236 Averill and Boothroyd

attitudes toward the romantic ideal influence the chances of his or her
falling in love? And what effect does a love experience gone bad, as indi-
cated by divorce, have on subsequent tendencies to fall in love?
The literature on romantic love has burgeoned in recent years (cf.
Huston, 1974; Murstein, 1974; Otto, 1972; Rubin, 1973), but there is sur-
prisingly little information on questions such as the above. Yet such ques-
tions are important, not just for the study of love, but also for the under-
standing of emotion in general. Since the time of Aristotle, and even before,
the experience of passivity (being overcome) has been considered one of the
hallmarks--even a defining characteristic--of emotion (Peters, 1962). Most
traditional theories of emotion have attempted to explain the experience of
passivity in biological terms, i.e., emotions are instinctive, " g u t " reactions,
independent of "higher" mental processes. But the view that emotions are
biologically primitive reactions is based more on prejudice than on fact (for
an historical review of this issue, see Averill, 1974). In contradistinction to
the traditional view, an argument can be made that the experience of passiv-
ity during emotion is basically an interpretation of one's own behavior
(Averill, 1976; Schafer, 1976). That is, responses such as anger, fear, and
romantic love do not just happen to an individual; rather, they are actions
that a person performs, but for which he does not--or cannot--accept
responsibility. The interpretation of a response as emotional may be made
simultaneously with, or even in anticipation of, the performance of the act,
in which case the individual may actually experience being overcome.
Often, however, the interpretation may be retrospective, i.e., the act is re-
appraised after the fact as conforming to appropriate Cultural norms. 4
These norms are seldom explicitly stated--there are no manuals on the
proper way to fall in love, say, as opposed to making love. Emotional
norms are conveyed primarily by means of paradigm cases, of which the ro-
mantic ideal of falling in love is one example.
It is not being suggested that love is always experienced passively, nor
that all persons endorse the romantic ideal. As just described, "falling" in
love, as depicted in song and literature, is simply a paradigm case that helps
give more customary expressions of love their meaning and significance.
The purpose of the present study was to determine the extent to which per-
sons interpret their own experiences of love as conforming to this paradigm,
and to explore in a preliminary fashion the circumstances conducive to
falling romantically in love.

'Emotional interpretations that are retrospective on one occasion may become prospective on
subsequent occasions. This is an important link in the chain o f emotional development. The
process can be illustrated in miniature by a commonly used technique o f hypnotic induction.
The subject is first led to interpret a naturally occurring response, such as swaying when the
eyes are closed, as being due to the suggestion of the hypnotist. The subject is then more likely
to dissociate subsequent responses from the self and attribute them instead to external causes.
On Falling in Love 237

METHOD

Subjects

In order to obtain a sample that was diverse in terms of both age and
prior experience, subjects were recruited from two sources: (a) upper level
psychology courses at the University of Massachusetts (62 subjects) and (b)
a club for divorced and other nonmarried persons in the Amherst area (25
subjects). The club members were mostly professional people, with some
students and semiprofessionals.
When being recruited, subjects were told that the study concerned the
nature of, and attitudes toward, romantic love. It was emphasized that no
embarrassing questions would be asked. Subjects enrolled in psychology
courses were given class credit for participating, while those from the club
volunteered out of interest and as a favor to the second author, who also
was a club member.
Two of the student volunteers indicated that they had never been in
love, and they were eliminated from the sample. This left a total of 85 sub-
jests, about equally divided between males (41) and females (44). Ages
ranged from 18 to 54, with a mean of 26.3. Of the subjects, 60 were single, 7
were married, and 18 were divorced.

Procedure

Subjects met in small groups, from 4 to 15 persons at a time. They

first completed the "Attitudes toward Love" scale developed by Knox and
Sporakowski (1968). This scale consists of such items as: "When you are in
love, you are usually in a daze." "Love at first sight is often the deepest and
most enduring type of love." "Regardless of other factors, if you truly love
another person, that is enough to marry that person." Each item is rated on
a 5-point scale, from "strongly agree" (1) to "strongly disagree" (5). Thus,
a low score on the Knox scale represents an endorsement of the cultural
stereotype of romantic love.
After completing the Knox scale, subjects indicated how many times
they had been in love, and answered some additional questions regarding
their attitudes toward love. They were then given the following newspaper
account of a couple who fell in love:

On Monday, Cpl. Floyd Johnson, 23, and the then Ellen Skinner, 19, total stran-
gers, boarded a train at San Francisco and sat down across the aisle from each other.
Johnson didn't cross the aisle until Wednesday, but his bride said, " I ' d already
made up my mind to say yes if he asked me to marry him." "We did most of the
talking with our eyes," Johnson explained. Thursday the couple got off the train in
238 Averili and Boothroyd

Omaha with plans to be married. Because they would need to have the consent of the
bride's parents if they were married in Nebraska, they crossed the river to Council
Bluffs, Iowa, where they were married Friday.

This account has been cited by Burgess and Wallin (1953, p. 151) and
Rubin (1973, p. 192-193) to illustrate the romantic ideal.
Subjects were reminded that instances of the above type also can be
found in literature (e.g., Romeo and Juliet), and they were asked to indicate
on a 10-point scale how closely their most and least intense experiences of
love conformed to the romantic ideal? Subjects were then asked to describe
the circumstances surrounding each of these experiences.

RESULTS

The degree to which the most intense experiences of love were inter-
preted as conforming to the romantic ideal is depicted in Figure 1. The dis-
tribution of subjects appears to form a U-shaped curve. Specifically, 34
subjects (40%) indicated that their experiences conformed closely to the
romantic ideal (ratings of 8, 9, or 10), while nearly the same number of sub-
jects (33) rated their experiences as quite dissimilar to the romantic ideal
(ratings of 1, 2, or 3). Relatively few subjects' ratings were in between.
A stepwise regression analysis (cf. Draper & Smith, 1966) was applied
to the data in Figure 1, in order to test for linear, quadratic, and higher-
order trends. Only the quadratic component of the regression equation was
statistically significant [F (1,8) = 13.29, p < .01], accounting for 62% of the
variance. These statistics confirm the curvilinear or U-shaped nature of the
curve in Figure 1.
The above results refer to the subjects' m o s t intense experiences of
love. However, for any given subject the most intense experience might
have been quite mild on an absolute scale. Is it possible that the distribution
of conformity ratings depicted in Figure 1 simply reflects differences in
intensity? To answer this question, a measure of intensity that is indepen-
dent of the conformity ratings is required. The duration of the experience
provides such a measure. (That duration is a rough measure of intensity is
indicated by the fact that 55 subjects said their most intense experiences
lasted 1 year or longer, while only 22 subjects indicated that their least
intense experiences lasted that long.)

5The examples presented to subjects do not characterize all aspects of the romantic ideal. How-
ever, the account of Corporal Johnson and Ellen Skinner, together with the reference to such
literary figures as Romeo and Juliet, highlights three of the main features of romantic love,
namely, suddenness of onset (love at first sight), a commitment to future involvement (such
as marriage), and turbulent emotional reactions (Romeo and Juliet).
On Falling in Love 239

co 15
I-
L)
1.1J
rn

Li_
o
n,-
LLI
rn

Z
~0

5
LL/ v
I I I I I I I I I ....... ! ........

I 2 3 4 5 6 7 8 9 I0
NO CLOSE
RESEMBLANCE RESEMBLANCE

Fig. 1. The number of subjects who indicated that their most intense ex-
periences of love either bore no resemblance, or conformed closely, to the
romantic ideal.

Table I. Duration of the Most Intense Experience of Love as a Ftmc-

tion o f Whether that Experience Did or Did Not C o n f o r m to the
Romantic Ideal

Duration of experience
1 week 1 month 6 months 1 year Longer
Conformed
to ideal 1 2 7 6 18
(N = 34)
Did n o t
conform t 2 7 1 20
(N = 33)

Table I presents the duration of the most intense experiences of those

subjects whose love did or did not conform to the romantic ideal (con-
formity ratings of 8, 9, or 10 vs. 1, 2, or 3). It is obvious that there were no
differences in intensity, as measured by duration, between these two types
of experiences, although not all of the experiences of either type were par-
ticularly enduring.
The above results do not mean that intensity had no effect on con-
formity ratings. This is indicated by the fact that only 14 subjects gave their
least intense experiences of love high conformity ratings (8, 9, or 10), while
44 subjects rated such experiences as dissimilar to the romantic ideal (1, 2,
or 3). Twenty-three subjects fell in the midrange¢
6These figures add u p to a total o f 81 subjects. T h e 4 other subjects indicated that they had
been in love only once, and their data are included among the most intense experiences.
240 Ave~iland Boothroyd

In summary, as the imensity of love increased from least to most,

there was a greater tendency to rate an experience as conforming to the
romantic ideal. However, among the most intense experiences reported by
subjects, there were no differences in intensity between those that did and
those that did not conform to the romantic ideal.
Let us turn now to some of the personal and situational factors that
helped determine how subjects interpreted their experiences.

Attitudes Toward Love

There was a small but statistically reliable correlation between scores

on the Knox scale and conformity ratings (r = -.31, p < .01). Since a low
score on the Knox scale is indicative of a romantic attitude, there was a ten-
dency for people who agreed with the romantic ideal to interpret their expe-
riences as conforming to that ideal. Of course, no causal relationship can be
inferred from this correlation: People might have agreed with the ideal be-
cause they had experienced romantic love; they might have experienced
romantic love because of prior favorable attitudes toward the ideal; or,
most likely, both a favorable attitude toward, and falling in love in accord-
ance with, the romantic ideal may reflect the influence of other personal
and contextual variables.

Sex and Marital Status

Subjects were divided according to sex and marital status, and a re-
gression analysis (Finn, 1974) was performed on how closely their most in-
tense experiences conformed to the romantic ideal. Neither of the main
effects of sex and marital status, nor their interaction, approached statis-
tical significance. Males were as likely to describe their experience as con-
forming to the ideal as were females, and the same was true for single,
married, and divorced subjects.
The responses of divorced subjects were of special interest, since it
might be assumed that they would be most impervious to the romantic
ideal. Of the 18 divorced subjects, 14 reported having had their most intense
experiences after the dissolution of their marriage. The experiences of these
14 subjects were distributed as follows: Three conformed closely to the
romantic ideal (rated 8, 9, or 10), six were dissimilar to the ideal (rated 1, 2,
or 3), and five were rated in the middle (4, 5, 6, or 7). The sample is small,
but it does indicate that divorce provides no sure "immunity" against fall-
ing romantically in love on a subsequent occasion.
On Falling in Love 241

Age

As might be expected f r o m the experiences o f the divorced subjects,

there was no significant relationship between age and whether or not an ex-
perience o f love was interpreted as conforming to the romantic ideal. This
was true with regard to both the subjects' current ages (r = .09) and their
ages at the time of the experience (r = - . 11). It thus appears t h a t falling in
love in conformance with the romantic ideal is not a p h e n o m e n o n confined
to the p r e v i o u s - - o r present--generation; nor is it a matter o f youth, regard-
less o f generation. Rather, one m a y fall in love at any age, provided the cir-
cumstances are right.

Situational Variables

A variety of possible circumstances that might influence falling in love

were listed in the questionnaire, along with an " o t h e r " category. The
number o f subjects who endorsed each circumstance is presented in Table
II. (Only those subjects whose experiences either did or did not c o n f o r m
closely to the romantic ideal are tabulated.) The most frequently endorsed
circumstances had to do with c o m m o n interests, the appearance o f the other
person, and meeting by chance. C o m m o n interests were i m p o r t a n t regard-
less of how closely the experience conformed to the romantic ideal. H o w -
ever, the appearance of the other person was checked significantly more
often when the experience was o f the romantic variety (chi square = 11.19,

Table II. Number of Subjects Who Indicated that a Particular Circumstance

Influenced Their Most Intense Experience of Love
Experience did
not conform to Experience conformed
romantic ideal to romantic ideal
Circumstance (N = 33) (N = 34)
Lonely 8 13
Moved to a new place 3 7
Broke up with another 3 4
Difficulties with parents 1 2
Chance meeting 11 20
Appearance of other person 7 22
Common interests 15 18
Long-term friendship 7 7
An expression of love by the other
person 6 5
Other 8 13
242 Averili and Boothroyd

d f = 1, p < .01). The same was true with regard to meeting by chance. In
the latter instance the difference between types of experience did not reach
traditional levels of statistical significance (chi square = 2.55, d f = 1, p =
•11); but, as will be explained below, additional analyses support a relation-
ship between a chance meeting and falling in love in conformance with the
romantic ideal.
Within-subject analyses (sign tests) were performed on the data of
subjects who rated their most intense experiences as conforming to the
romantic ideal (8, 9, or 10) and their least intense experiences as being dis-
similar (1, 2, or 3). There were 15 such subjects. Two variables, the appear-
ance of the loved one and a chance meeting, were involved significantly
more often in the experiences that conformed to the romantic ideal
(p < .05). In spite of the fact that these analyses confound intensity with
type of experience, they provide support for the results of the between-sub-
jects analyses described above•
In short, of the factors listed in Table II, the appearance of the loved
one and a chance meeting seem to be the most important in determining
whether or not an experience of love is interpreted as conforming to the
romantic ideal. This is not surprising, since the romantic ideal emphasizes a
sudden falling in love. Such suddenness is unlikely to happen among friends
or acquaintances (although romantic love is not precluded by prior friend-
ship, as indicated by the responses of seven subjects). Also, when a meeting
is by chance, there is little else to go on initially except appearance.' Once
established, common interests may play a major role in determining
whether or not the relationship will become intense and endure, but that is
true whether or not the experience conforms closely to the romantic ideal.
In addition to the factors listed explicitly in Table II, subjects were
provided with an " o t h e r " category, in which they could state any additional
circumstances surrounding their experience. This category elicited a wide
variety of idiosyncratic responses, with some form of sexual frustration or
"opening up" being the only factor mentioned more than once (by two
males whose experience conformed to the ideal and one female whose ex-
perience did not). This infrequent mention of sex is worthy of comment in
view of the fact that romantic love has often been considered to be the pro-
duct of frustrated sexual desires (cf. Freud's famous definition of love as
"aim-inhibited sex"). Some of the subjects in the present study may have

7,,Appearance,, is a broad concept, which can encompass many aspects of a person, e.g., phys-
ical attractiveness, mannerisms, dress, and other accoutrements of social class, as well as the
more intangible aspects of personality, such as attitudes, beliefs, and aspirations. Of these,
physical attractiveness is perhaps the most important during the initial stages of romantic in-
volvement. For an excellent review of this poorly understood--and little investigated--area of
psychology, see Berscheid and Walster (1974).
On Falling in Love 243

been reluctant to mention sex as an explicit factor in their love relationship,

or they may not have recognized its influence. However, an examination of
the open-ended descriptions that subjects gave of their experiences, as well
as data from prior empirical research (cf. Burgess & Wallin, 1953, Ch. 7),
lends little support to the view that sexual frustration per se is a primary de-
terminant of romantic love.

DISCUSSION

To summarize briefly, 40% of the subjects indicated that their most

intense experiences of love were similar to the romantic ideal, while a like
percentage indicated that their experiences were dissimilar. There was little
middle ground. It is unlikely that these results are idiosyncratic to the
present subjects, for the sample was quite heterogenous on a number of
variables (e.g., age, sex, and marital status), none of which had a significant
influence on the conformity ratings. Of those personal and situational vari-
ables which did influence conformity ratings, a favorable attitude toward
the romantic ideal, the appearance of the loved one, and a chance meeting
were the most important? This list is, of course, incomplete. The present
study was not designed to investigate in any depth the conditions under
which persons fall in love. Therefore, the remainder of this discussion will
be devoted to an analysis of the distribution of conformity ratings depicted
in Figure 1.
Let us begin with the fact that 40°7o of the subjects indicated a high de-
gree of conformity with the romantic ideal (ratings of 8, 9, or 10). This
would seem to belie the contention by some recent commentators (e.g., Kit-
patrick, 1974) that romantic love is dead and in need of an obituary. Of
course, it might be objected that 40% is an overestimate of the number of
people whose experiences of love actually conform to the romantic ideal.
For one thing, most of the experiences of romantic love reported by the
younger subjects already had ended (or probably will end) in separation.
And with additional experience, what now seems like the epitome of love
may later be reappraised as infatuation, or even silliness. But this argument
can be turned around. As time goes on, the likelihood increases that a
person will have an intense love experience; and, as we have seen, the more

*With regard to the nearly 400/0 who indicated that their most intense experiences did n o t con-
form to the romantic ideal, these subjects were generally quite vague in describing the circum-
stances surrounding their relationships, perhaps because their experiences were not particu-
larly "dramatic." The themes most frequently emphasized in the open-ended accounts were a
gradual buildup of feelings, common interests and goals, cbmpatibility, understanding, and
comfort in the presence of the other person.
244 Averill and Boothroyd

intense an experience is, the more likely it is to be interpreted as conform-

i n g - - o r not c o n f o r m i n g - - t o the romantic ideal. Stated differently, an in-
crease in intensity causes a polarization in the interpretation of love, b o t h
toward and away f r o m the romantic ideal.
There is another sense, however, in which 40% m a y b e - - i n fact, un-
doubtedly i s - - a n overestimate o f the number o f persons whose experiences
c o n f o r m to the romantic ideal. The romantic ideal is multidimensional, and
subjects in the present study apparently used a variety o f criteria for de-
termining how similar their experiences were to the paradigm cases pre-
sented. In a number of instances, the conformity was quite evident, as is
illustrated by the following verbatim description:

The most intense [experience] left me completely enraptured with the person.
Nothing else seemedto matter. She was everything I had ever hoped and looked for
in another person, It was very beautiful, mystical, encompassing most of my
thoughts, feelings and actions. I was extremely happy, after being lonely and for-
lorn, but I hadn't been looking for anyone else. I had wanted to be alone for a while,
but just couldn't help myself. I fell in love with her after talking to her once.
(Conformity rating of 9)

On the other hand, subjects sometimes rated their m o s t intense experi-

ences as similar to the romantic ideal, even though such c o n f o r m i t y is not
readily apparent f r o m their open-ended descriptions. The following is an
example:

My present, most intense romance is thriving just fine. Now I have my feet on the
ground but I love him and he loves me very much. We've been together for 6 months
and I really feel this is a much more mature, and somewhat practical relationship.
We act crazy but we do have parallels in opinions and future dreams which give us
more substance upon which to build a relationship. (Conformity rating of 10)

Unfortunately, the descriptions most subjects gave o f their experi-

ences were not sufficiently detailed to allow an independent assessment
o f conformity to the romantic ideal. A very rough approximation would
be that between 5 % and 10% o f all the subjects had experiences that
conformed closely to the romantic ideal in a literal sense. But regardless o f
how m a n y subjects "literally" fell in love in accordance with the ideal, it is
still worthy o f note that 40% o f them interpreted their most intense expe-
riences as so conforming.
Let us turn now to the bimodal nature of the conformity ratings, i.e.,
the fact that subjects interpreted their experiences as either conforming or
not conforming to the romantic ideal. This fact might be taken as support
for a biological interpretation of romantic love, such as that by Money
and Ehrhardt (1972). These authors analyze romantic love as an imprinting
p h e n o m e n o n - - e i t h e r it happens or it doesn't. The critical period for the
imprint, they speculate, is initially in adolescence, although there appear to
On Falling in Love 245

be no upper age limits. The primary releaser-stimuli for falling in love are
the physical characteristics of the loved one (a speculation that also receives
support from the present data). With regard to the adaptive significance of
the process, Money and Ehrhardt suggest that "falling in love is, perhaps,
not only Nature's way of guaranteeing the mating of the species, but also of
keeping the couple together long enough for sexual affection to be joined by
parental affection, and thus of ensuring the proper care and protection of
the young" (p. 192).
There are, however, aspects of the present data that do not accord
well with an analysis of romantic love in biological terms. Specifically, we
have just seen that many of t h e subjects who gave high conformity ratings
were evidently reinterpreting their experiences as conforming to the
romantic ideal. This fact suggests a sociological explanation for the bimodal
distribution of conformity ratings. But before getting to that, it is necessary
to make a brief digression.
The bimodal distribution of conformity ratings suggests that the
romantic ideal and its opposite (or set of opposites) forms a typology rather
than a continuum. It might be helpful, therefore, to compare the naive
(everyday) psychology of romantic love with the more formal typologies of
personality theory. A typology, such as the introversion-extraversion
dichotomy of Jung (1933), assumes that people can be categorized into dis-
crete types. 9 The ideal of a type represents a complex blend of character-
istics to which few people actually conform. Indeed, the ideal type is not so
much a description of actual people as it is an abstraction or tool to aid in
the interpretation of behavior. Similarly, the romantic ideal is not so much
a description of actual behavior as it is a standard by which people interpret
their experiences of love. But an ideal type may do more than simply aid in
the interpretation of behavior: it can also help shape or determine the
behavior it purportedly explains. This is especially true of popular (as
opposed to theoretical) ideals. Thus, peopl e may try to emulate the ideal
and, even when they fail, may still interpret their behavior as being more in
conformance than it actually is.

~Typologies in this sense can be compared with trait theories. In Eysenck's (1952) scheme, for
example, introversion and extroversion represent the extremes of a continuum, with persons
distributed normally in between. It is perfectly conceivable that the conformity ratings in the
present study would also form a normal distribution, though they didn't. Of course, the kind
of distribution one obtains is a function, in part, of the method o f measurement. If different
aspects of the romantic ideal (e.g., suddenness of onset, total involvement, emotional turmoil)
were tested separately, the responses along each dimension might be normally distributed. But
the intent of the present study was not to dissect the romantic ideal into its components;
rather, it was to see how persons respond to the ideal as a paradigm case, with its attendant
vagueness. Evidently, a subject whose love conformed closely to the ideal on one of its dimen-
sions tended to give major weight to that dimension in determining the relevance of the ideal
as a whole to his or her experience.
246 Avefilland Boothroyd

The above line of reasoning does not explain why the romantic ideal
was established in the first place, what function it serves within the social
system, or why some people interpret their experiences as conforming to it
while others react against it. Although the present data do not address these
issues directly, the authors are in sympathy with the type of argument pre-
sented by Greenfield (1965). Greenfield has suggested that the romantic
ideal is one way of encouraging couples to marry, even though marriage
may be counter to their immediate self-interests. In our culture, for exam-
ple, the individual is encouraged to be self-reliant, independent, and--in a
sense--economically selfish. This is the "rational" way to behave. On
another level, however, society also encourages the individual to support a
wife and children (or, in the case of the wife, unselfishly to give up a career
in the marketplace for a less prestigious domestic role). How can
these conflicting demands be resolved? The answer, according to Green-
field, is found in the romantic ideal. One marries because one has fallen in
love--an event that is beyond self-control.
It is not necessary to agree with the details of Greenfield's analysis in
order to recognize the principles involved. Stated most generally, standard
emotional reactions--of which romantic love is but one example--are ways
of coping with conflicting societal demands or expectations (cf. Averill,
1976). In this respect they may be compared with hysterical reactions, which
are ways of coping with intrapsychic conflicts. In both instances, the indi-
vidual is encouraged to engage in behavior that is discouraged on another
level. The conflict is resolved by interpreting the act as occurring indepen-
dently of the self.
A corollary to the above analysis is that persons who have not inter-
nalized the relevant, though potentially conflicting, demands of society also
will not interpret their experiences of love as conforming to the romantic
ideal. Indeed, they may reject that ideal as unnecessary, at best, and harm-
ful, at worst. In this respect, it is relevant to note that there appears to be a
strong "counterculture" bias among many recent commentators who reject
the romantic ideal (see especially the collection of essays by Otto, 1972).
Unfortunately, we do not know whether such a bias also was characteristic
of those subjects in the present sample who interpreted their experiences as
dissimilar to the romantic ideal.
The above observations are admittedly speculative, and other explan-
ations could be offered for the bimodal distribution of conformity ratings.
At the present time, however, the explanations offered are perhaps less im-
portant than the issues raised. If nothing else, the present results illustrate
how little we know, even in a descriptive sense, about how people interpret
their own emotional experiences, especially when it comes to being
"gripped" or "overcome" by emotion. As noted at the beginning of this
On Falling in Love 247

article, the experience of passivity (being overcome) has long been con-
sidered one of the hallmarks of emotion. Yet we still cannot say with any
precision exactly what constitutes this experience, how frequently it occurs,
whether it is an integral part of an emotional reaction or whether it is pri-
marily a retrospective interpretation of behavior, whether it is more char-
acteristic of some emotions than others, and of what variables it might be a
function. Such information is essential to any general theory of emotion.

REFERENCES

Averill, J. R. An analysis of psychophysiological symbolism and its influence on theories of

emotion. Journal for the Theory of Social Behavior, 1974, 4, 147-190.
Averill, J. R. Emotion and anxiety: Sociocultural, biological, and psychological determinants.
In M. Zuckerman & C. D. Spielberger (Eds.), Emotions and anxiety: New concepts,
methods, and applications. New York: LEA-John Wiley, 1976.
Berscheid, E., & Walster, E. Physical attractiveness. In L. Berkowitz (Ed.), Advances in exper-
imental social psychology (Vol. 7). New York: Academic Press, 1974.
Burgess, E. W., & Wallin, P. Engagement and marriage. Philadelphia: Lippincott, 1953.
Draper, N. R., & Smith, H. Applied regression analysis. New York: Wiley, 1966.
Eysenck, H. J. The scientific study o f personality. London: Routledge & Kegan Paul, 1952.
Finn, J. D. General model for multivariate analysis. New York: Holt, Rinehart & Winston,
1974.
Greenfield, S. M. Love and marriage in modem America: A functional analysis. Sociological
Quarterly, 1965, 6, 361-377.
Huston, T. L. (Ed.), Foundations of interpersonal attraction. New York: Academic Press,
1974.
Jung, C. G. Psychological types. New York: Harcourt, Brace & World, 1933.
Kilpatrick, W. The demythologizing of love. Adolescence, 1974, 9, 25-30.
Knox, D. H., Jr., & Sporakowski, M. J. Attitudes of college students toward love. Journal of
Marriage and the Family, 1968, 30, 638-642.
Money, J., & Ehrhardt, A. A. Man & woman, boy & girl. Baltimore: Johns Hopkins Univer-
sity Press, 1972.
Murstein, B. I. Love, sex, and marriage through the ages. New York: Springer, 1974.
Otto, H. A. (Ed.), Love today. New York: Association Press, 1972.
Peters, R. S. Emotions and the category of passivity. Aristotelian Society Proceedings, 1962,
62, 117-134.
Rubin, Z. Liking and loving. New York: Holt, Rinehart & Winston, 1973.
Schafer, R. A new language for psychoanalysis. New Haven: Yale University Press, 1976.