We thematically content analyzed 100 romantic fiction novels, recently published by Harlequin Books, with a family therapeutic approach providing theoretical background. The novels fall into 3 major motif types: Cinderella, Beauty and the Beast, and Taming of the Shrew. As hypothesized, we found a blatantly patriarchal value system characteristic of the novels: traditional gender roles, marked inequality in spousal structure, lack of congruent communication patterns, consistent naturalizing of aggressive spousal behaviors. The discussion draws on feminist and family therapy literature, and deals with the destructive messages the romantic fiction genre delivers to millions of readers regarding gender roles and marriage.
The target article below was today published in PSYCOLOQUY, a refereed journal of Open Peer Commentary sponsored by the American Psychological Association. Qualified professional biobehavioral, neural or cognitive scientists are hereby invited to submit Open Peer Commentary on it. Please email or consult the websites below for Instructions if you are not familiar with format or acceptance criteria for PSYCOLOQUY commentaries (all submissions are refereed).
To submit articles and commentaries or to seek information:
EMAIL: email@example.com URL: http://www.princeton.edu/~harnad/psyc.html http://www.cogsci.soton.ac.uk/psyc
1. Feudalism, described by Weber (1922/1957, pp. 375-376) as a type of patriarchal authority, prevailed in Europe roughly from the 8th to the 14th centuries. One of its major characteristics was a remarkably rigid, seigniorial system of social stratification, where lords exercised a wide variety of rights over their vassals, who owed them specific services. Nominally, contemporary Western societies are guided by principles antithetical to feudalism. True, for several democracies it has taken some 200 years to legally adopt the slogan of the French Revolution: egalite, liberte et fraternite (consider such examples as German race laws under the Third Reich of the 1930's; racial segregation in the U. S. A. in the 1950's, no voting rights for women in Switzerland until the 1970's). But the era of feudal values is over now, is it not? Not if one belongs to the 50 million readers who annually purchase 175 million copies issued by only one of several romantic fiction publishers (Harlequin, 1999). "Feudal notions" wrote Modleski (1982, p. 50), "... are still the staple of Harlequin Romances..."
2. Throughout the years fascinating issues have arisen relating to various aspects of romantic fiction; these have been recently addressed by such empirical researchers as Cherland & Edelsky, (1993, on pre-adolescents in Canada), Christian-Smith (1990, 1993, on US teenagers, including a content analysis of 34 adolescent romance novels), Gilbert (1993, on Australian teenagers), Radway (1983; 1991, on US housewives), Whissell (1996, a content analysis of 25 popular Women's novels), and Willinsky & Hunniford (1993, on Canadian teenagers). The topic has justifiably attracted feminist inquiry, as well, such as Bridgwood (1986), Douglas (1980), Hallam & Marshment (1995), Jones (1986), Light (1984), Treacher (1988), and Walkerdine (1990), with advocates lining up on both sides of the barricades.
3. Our current effort is aimed at a previously untouched facet of this complex phenomenon: We are interested in the messages delivered by the romantic fiction genre regarding gender roles and marriage. The methodology we employ (content analysis) is not suitable for studying "the ultimate effects the fantasy resolution has on the women who seek it out again and again" (referred to by Radway, 1991, p. 15, as "the crucial question"). Yet we assume that the norms and values contained in these books leave a mark on the 50 million women (250 million in Coward, 1984, p. 190; over fifty percent of all women, in Light, 1984) who annually read them. Based both on the above cited sources and our brief perusal of some romantic novels, we have some hypotheses about the nature of this mark: We expect to find in a sample of romantic novels published in the last decade traditional gender roles, inequality in spousal structure, a lack of congruent communication patterns, and a consistent naturalizing of aggressive spousal behaviors. Our assumption about the effects of reading on readers is neither far-fetched nor outlandish: From Plato (Republic, Book III, 391) to contemporary censors and book-burners it has been axiomatic that we are subject to the influence of the material we hear and read. The same assumption constitutes the basis of bibliotherapy (cf. Reed, 1993, pp. 404-405 for a comparison of reading with therapy).
4. Of approximately 5000 novels published between 1990 and 1998 by Harlequin Books (Mills & Boon, as well as Silhouette), readily available in bookstores and libraries, 100 were randomly sampled and submitted to thematic content analysis. Most of the variables selected for analysis derive from the theory of family therapy. Some others were empirically established during our reading of the novels. The variables extracted were as follows.
4. i. For the novel: general theme, the occurrence of a reunion following a previously severed relationship, the promise of a child at the denouement, the occurrence of lies or secrets.
4. ii. For each of the two main protagonists: marital status, relative power, predominant communication pattern (see # 23, below), prior sexual experience, use of physical and verbal aggression, being a pursuer or a distancer (see # 16, below), sacrificing for the sake of spousehood a career or a dwelling place at the denouement, lack of living parents or siblings.
4. iii. For the couple: type of spousal structure (complementary, symmetrical or reciprocal; see # 18, below), the presence and the direction of a difference in social status, wealth, education, age, and being distressed.
5. 20 % of the sampled novels was independently read and coded by both authors. The inter-coder reliability, aggregated over all the variables, was 89%.
6. The quantified results of the content analysis have been cast in three tables. Tables I and II contain findings relevant to the novel as a whole, and to the couple appearing in it.
Table I. Descriptions of the novel and of the couple by theme type (in percents).
______________________________________________________________ Variable Theme Type ______________________________________________________________ Cinderella Beauty and Taming of Total the Beast The Shrew n= 42 n=36 n=19 n= 97 ______________________________________________________________ Status M>F* 88 58 68 73 F>M 2 22 11 11 F=M 10 19 21 15 ______________________________________________________________ Property M>F 88 67 74 77 F>M 5 19 11 11 F=M 7 14 16 11 ______________________________________________________________ Education M>F 81 42 68 64 F>M 2 25 11 12 F=M 17 33 21 24 ______________________________________________________________ Age M>F 100 100 100 100 F>M 0 0 0 0 F=M 0 0 0 0 ______________________________________________________________ Distress M>F 2 8 5 5 F>M 88 58 53 70 F=M 10 33 42 25 ______________________________________________________________ Spousal Structure Complem'tary 93 42 0 56 Symmetrical 5 53 100 41 Reciprocal 2 6 0 3 ______________________________________________________________ Reunion 31 28 21 28 Child/Pregnancy 43 42 21 38 Lies, secrets 62 47 68 58 ______________________________________________________________ * Male higher than Female
7. As shown in Table I, Harlequin novels have a small number of easily distinguishable themes or motifs, with 4 types covering 97 percent of the novels in our sample. The remaining three novels, not using any of these motifs, were dropped from all subsequent analysis. For the sake of identification, we shall refer to each type by the name of a well-known tale: Cinderella (35%), The Beauty and the Beast (37%), and The Taming of the Shrew (20%). An additional 8% of the sample made use of a combination of two themes: Cinderella and the Beast. (Because of their relative infrequency in the present sample, and owing to the strong similarity to the Cinderella theme, we have combined these with novels belonging to the Cinderella category in all subsequent analyses.)
8. Table II shows that while about two thirds of the novels feature previously unmarried protagonists, a considerable proportion of the heroes (28%) and of the heroines (15%) had been divorced.
Table II. Marital status of the protagonists by theme type (in percents).
_______________________________________________________________ Marital Theme Type Status _______________________________________________________________ Cinderella Beauty and Taming of Total the Beast The Shrew M F M F M F M F
n= 42 n= 36 n= 19 n= 97 _______________________________________________________________ Never Married 57 60 58 78 74 63 61 67 Never Married+ 2 0 3 3 5 5 3 2 Divorced* 21 12 19 6 11 11 19 9 Divorced *+ 12 10 8 6 5 0 9 6 Widowed 2 7 0 3 0 11 1 6 Widowed + 5 12 11 6 5 11 7 9 ________________________________________________________________
* Including 5 couples (of whom 3 have children), reunited after a previous relationship which had ended in either divorce or separation. + Plus child/ren.
9. Table III presents information relating to the male and female protagonists (whom we may more properly call the antagonist and the protagonist, respectively). In practically all of the variables there are strong differences both between men and women and within the three theme types.
Table III. Description of the main protagonists by theme type and gender (in percents).
Variable Theme Type _________________________________________________________________ Cinderella Beauty and Taming of Total the Beast The Shrew M F M F M F M F n= 42 n= 36 n= 19 n= 97 _________________________________________________________________ Power Dominant 86 5 83 42 95 95 87 36 Submissive 0 81 0 42 5 5 1 52 Equal 14 14 17 17 0 0 12 12 _________________________________________________________________ Sexual Experience None 0 55 0 47 0 53 0 52 Low 5 45 6 47 11 37 6 44 Moderate 29 0 31 3 11 0 26 1 High 67 0 64 3 79 11 68 3 _________________________________________________________________ Pursuer 81 14 58 28 84 0 73 16 _________________________________________________________________ Communication Blamer 42 38 58 36 63 79 53 45 Placator 2 14 6 25 11 0 5 15 Super-R'tnal 2 0 3 0 0 0 3 0 Congruent 24 31 6 17 11 5 14 21 None 26 17 28 22 16 16 25 19 _________________________________________________________________ Aggression Phys. & Ver. 19 7 14 6 58 37 25 12 Physical only 5 7 0 11 0 11 2 9 Verbal only 19 14 31 17 11 21 22 16 Neither 57 71 56 67 32 32 52 62 _________________________________________________________________ Gives up job or 12 50 19 47 11 58 15 51 residence _________________________________________________________________ No parents 67 69 67 67 58 63 65 67 _________________________________________________________________ No siblings 62 79 72 81 63 84 66 80 _________________________________________________________________
10. In addition to presenting separate distributions of the prevalent communication patterns for men and women (based on Virginia Satir's 1967 classification; see Table III), we also show their cross-tabulation: Table IV contains the characteristic communication patterns used by each couple in our sample. The code "None" appears in those cases where one of the protagonists does not engage in any psychologically meaningful conversation with the other. Note that though we found several examples of irrelevant communication, none of the heroes adopted this incongruent pattern as their predominant style.
Table IV. Cross-tabulated dyadic communication patterns in 97 novels (in percents).
Female Protagonist ________________________________________________________________ Male Blamer Placator Super- Congruent None Total Protagonist Rational ________________________________________________________________ Blamer 36 11 0 3 2 52 Placator 2 1 0 0 2 5 Super-R'tnal 1 0 0 2 0 3 Congruent 2 1 0 9 2 14 None 4 2 0 6 12 25 ________________________________________________________________ Total 45 15 0 21 19 100
Each of the three types identified (see Tables I and II) is based on a non-egalitarian, blatantly stratified relationship between a female and a male protagonist.
11. i. Cinderella type. In this classical fairy-tale (from our point of view analogous in its basic structure to Sleeping Beauty, Snow-white and many other tales), the Damsel in Distress is saved by Prince Charming, who promises her a married life of status, wealth and happiness. (Cf. Walkerdine, 1984). The heroine is passive: Through no fault of hers, she has been hurt by circumstances; helpless and powerless, her redemption is similarly achieved by an external agent. In the hierarchical system necessary for this type of story, the woman is effectively told 'soit belle et tais tois', while the active, strong, brave, and rich hero earns the right to win a decorative wife. In this complementary system (see # 19, below) there is no sharing, team-work, or functional communication. In our sample, 35 percent of the novels fall into this category. With its encouragement of helplessness, an external locus of control, and a passive waiting for a miracle to occur, the Cinderella-type romantic novel gives no chance to its reader to recognize and acknowledge her own power. What can be the role expectations of women (from themselves and from their spouses) who have grown up on these fairy tales, and continue to be fed the same myths in their adolescence and in their adulthood? As the results show, this type of novel contains the most blatant gender stereotypes, for its heroine has the weakest starting position, is most dependent, and, therefore, is most unlike the typical hero.
11. ii. The Beauty and the Beast type. This motif consists of a violent, insulting, sometimes physically aggressive, punishing male who is finally tamed by his beautiful, self-sacrificing victim. (This is the only theme recognized in the genre by Coward, 1984, p. 189). At the denouement the heroine discovers that all his aggression had only masked his sensitivity and love for her; in fact, his cruelty and violence were the proof of his love. Intimacy is not built gradually, there is no rehabilitation process: At the last moment the hero suddenly, magically changes from beast to beau (in a fashion analogous to the Frog-Prince's metamorphosis in the Grimm fairy tale). Thirty seven percent of the novels read belong to this category. The beast may have a temporary physical disability, but more often he is emotionally handicapped. He is cynical, distrustful of women in general, exceptionally so towards the heroine. She undergoes a series of emotional and/or physical humiliations (often humbly), before realizing that he has loved her all along. Having passed all the tests, she wins a husband, who is, again, rich, successful etc. Their relationship is not merely stratified, it is also wrought with verbal, emotional and physical abuse. Communication is degrading, it serves to illuminate the protagonists' power differential, the gap in their status, their roles as controller and controlled. The destructiveness of this theme lies in that it whitewashes and reinforces abusive relationships. (See Kurz, 1993; Lansbury, 1982; Stordeur & Stille, 1989, p. 27). The abusive partner claims that he loves his victim, while the abused one finds in his behavior proofs of love and falls in love with him. Consequently, she cannot leave the situation and waits for the metamorphosis, for turning over a new leaf. The depiction of women's willingness to take aggression as a sign of love unfortunately, yet necessarily, limits the alternative courses of action available to those reader s who find themselves in this nexus.
11. iii. Cinderella and the Beast type. This is a combination of two motifs: Inequality between the protagonists is even more heavily emphasized than in either of the two basic motifs by contrasting the penniless and weak heroine with a powerful and aggressive hero whom she "tames" and marries. Both Anderson (1974, p. 240) and Jones (1986) regarded this configuration as the typical plot of the Mills & Boon novel. Eight percent of the sample read use this alternative. (They are included in the Cinderella category).
11. iv. The Taming of the Shrew type. This motif is well known around the world. Brunwald (1966) identified it as Aarne-Thompson Type 901, and collected over 400 versions of it in 30 nations. The shrew is a woman who dares to think that she is the master of her destiny, has an independent mind, makes her own choices, will not let others control her life. This attitude equates her with males (who are expected to have all of these qualities without censure). Her lack of submissiveness is expressed through a constant struggle with her environment, her stance is bellicose, and she is depicted negative, self-centered, capricious, non-adjusting to her roles in life. The hero who falls in love with her feels challenged to change and subdue her, so as to deserve him. The Taming of the Shrew provides the motif for 20 percent of the novels read. In romantic fiction this motif begins as symmetrical in dyadic structure (see below), but becomes either complementary (the shrew is tamed in about two thirds of the cases), or the conflict remains unresolved. In the latter case the "happily-ever-after" promise contains another myth: "They'll continue to fight, but at least they won't be bored!" But whether the result is female submission, or a continuous power struggle, in none of these novels is there an alternative solution of a reciprocal structure: equality, harmony between two equally powerful individuals, team-work. It appears as if there was a contradiction between self-esteem and living in spousal harmony.
12. The highly stereotypical male and female roles found in romantic fiction have often been commented on. (See, for example, Douglas, 1980; Greer, 1971; Margolies, 1982; Modleski, 1980; Whissell, 1996). While these roles may be seen to directly follow from the motifs discussed above, it is no less appropriate to say that it is the gender roles that dictate the motifs. The rigidly traditional character of these roles is rendered even more conspicuous by their being near-empty shells: There are no personalities behind the virile males and the frequently submissive, occasionally assertive virginal heroines. (Most of the latter end up relinquishing their achievements for the sake of marriage.) When children play some part in the romance (in the now acceptable second marriages of widowed or divorced characters), they lack personality traits, too, and serve only as supernumeraries in their parents' romantic entanglements. Such absence of personality in the protagonists leads to a complete lack of insight on their part. They can be angry at each other, blame or ingratiate themselves, but they never take responsibility or ask what their share is in a conflict.
13. While the men of these novels are expected to be successful and strong, a woman's role includes marriage and serving as a womb for the next generation. In 38% of the novels in our sample the heroine affirms this expectation by either being pregnant, or at least by planning a child at the denouement. This lack of differentiation between spousehood and parenthood is familiar to family therapists; it is often women who substitute their investment in the latter for the former. By equating marriage with children, romantic fiction naturalizes for its readers the skipping of the essential stage of establishing spousehood ("coupling", in Goldenberg & Goldenberg, 1996; see also McGoldrick 1989). Yet in reality, only when this stage has been achieved, does a couple have the solid base required to meet the tasks of subsequent stages.
14. Gender roles are further punctuated by consistent and pronounced differentials in the status, wealth, education, age (see Table I), power and sexual experience (see Table III) of the protagonists. In spite of some differences among the three theme types, with the Cinderella-type stories producing the largest differentials, the basic trend is identical: In each of these variables men of romantic fiction overshadow women. Only in 12% of the novels have the male and female protagonists equal power, and in 15% -- equal social status. The novels often implicitly, sometimes explicitly suggest that women like to be dominated, like to surrender to a powerful man. The relationship between the hero and the heroine is then not one between two mature adults, but rather a father/daughter tie (the hero might be, predictably, ten to fifteen years older). The coding of distress (a strong external threat accompanied by inability to cope alone and a dependence on the other protagonist's help) results in a similar picture, with women showing more distress than men.
15. The relative sexual experience of the hero vs. the heroine deserves some comment. The novels in our sample make it clear that the more sexually experienced the hero is, the more his value increases. The opposite holds for the heroine, who is most valuable as a virgin (52% of the sample). Practically all of the non-virgins (those coded as having "Low " sexual experience) had one disappointing encounter, usually with an ex-fiance. In other words, in a classical double-standard fashion, the reader is told that men can enjoy sex without emotional involvement, while women enjoy only their single great love. This scenario constitutes a double bind for women who are expected to be sexy, to enjoy sex, but also to be monogamous, suppressing their need until the "right" moment. Sex serves as a litmus test: If she doesn't experience fireworks during her sexual encounter, t hen this is not the real thing.
16. A related issue is the sacrifice one member of the dyad makes at the romantic novel's denouement. In 51% of the cases the heroine gives up her career, plans, or dwelling place, for the sake of a marriage with the hero. (Cf. a similar finding with respect to adolescent romance novels in Christian-Smith, 1990, p. 125). In the few cases in which the hero is the one who gives up his job or makes another, comparable sacrifice, he does so in order to better himself, rather than for the sake of the spousal relationship. Though the subject of aggression is to be discussed only later, let it be mentioned here that one of the leading causes of wife battering is her desire to work outside the home (Roy, 1982b).
17. The last role-related issue to be considered is a characterization of the novels in terms of the courtship dynamics between heroine and hero. In 73% of the cases he woos, she plays hard to get. Such a pursuer-distancer relationship (see Fogarty, 1976; Guerin, Fay, Burden, & Kauto, 1987) is one of the most common presenting symptoms in couple therapy, with intimacy being attempted and constantly rejected. The more desperate the pursuer, the more crucial it becomes for the evader to hide, thus increasing the distance and the frustration for both, while preventing any possibility of intimacy. The pursuer cannot confess to misery, for fear of exposing any weakness and thus losing control. Neither can the distancer share his or her plight, the major hiding technique being the suppression of feelings. Such dyads exist in constant tension, frustration and anger, since both partners' needs are left unsatisfied. Readers of romantic fiction, however, can only conclude that to play hard to get is worthwhile, and that spousehood is built on this theme of pursuing and evading, without any other healthy role option of two equal partners.
18. To summarize this section on roles, we suggest that each and every novel in our sample virtually encapsulates Eagly's (1987) social roles theory of gender stereotypes: Starting with a division of labor, through a display of role consistent skills and behaviors, the readers arrive at highly gender-stereotyped social perceptions of the protagonists. We can only surmise, how these perceptions influence their extra-textual life...
19. Dyadic structures can be roughly described as complementary, symmetrical, or reciprocal. In complementary dyads an unwritten contract between the spouses provides for the traits of one to complement the missing traits of the other (e.g. rich and poor, strong and weak, independent and dependent ). Their inherently non-egalitarian relationship is clearly stratified and stereotyped. As we gradually distance ourselves from blatantly feudal family norms, complementary spousehoods give way to symmetrical ones. In a symmetrical dyad two partners are in a power struggle with regard to their common traits, constantly engaged in a combat to prove superiority over each other in relation to abilities and activities (e.g. who is more strong-willed, who has the last word). The dysfunctionality of such dyads is a direct result of the large amounts of energy that are wasted on rivalry, rather than on cooperation. In contrast to these, reciprocal relationships (called parallel in Lederer & Jackson, 1968, p. 161) are based on team-work between two equal partners, with each partner's contribution to, and benefits from, the partnership being based on abilities and needs, rather than on bookkeeping.
20. In our sample 56 % of the novels describe an initially complementary relationship between the protagonists; in an additional 12 % the structure changes from symmetrical to complementary throughout the novel. As with respect to other variables, as seen above, Cinderella-type novels present the most stratified, feudal, patriarchal structure: The dyadic relationships described in them are almost exclusively complementary. To anticipate our subsequent discussion of aggression, Kurz's (1993) survey showed that one of the three major causes of wife abuse was a complementary family structure, with the least amount of violence against women occurring in democratic (i.e. reciprocal) households (see also Yllo, 1994, with respect to women's status in society and levels of violence within the family).
21. In 41 percent of the novels the hero and the heroine are in a symmetrical relationship. Though this structure is most characteristic of the Taming-of-the-Shrew motif, it is not unique to it: 53% of the Beauty-and- the-Beast-type novels also have this kind of dyadic configuration. Though symmetrical relationships are not stratified, neither do they provide a model of mutuality, reciprocality, or team work of two strong individuals who have mutual respect. The protagonists are engaged in a continuous power struggle: They fight over their independence, constantly declaring how they cannot trust each other. Their struggle revolves around their anxiety of exposing their need for anybody, and most of all, for the person they will marry at the end of the novel.
22. Two additional structural issues concern the main protagonists' family of origin and prior relationship. A third issue, their marital status, will be dealt with further down.
22. i. Family of origin. Several sources commented on the orphaned status of romantic fiction's heroines (Margolies, 1982; Snitow, 1979; Treacher, 1988), a status that is amenable for placing the hero in loco parentis. While the current sample supports this finding (fully two thirds of the women have no living parents; see Table III), the same holds for the heroes. The main protagonists have no siblings, either, though in this case women outdo men (80% vs. 66%). However, this apparent near-equality in not having a family is misleading. For on the one hand, men with no family ties are depicted as independent, foot-loose, rich heirs, who do not have to share their wealth with their siblings. When men do have a family, they are well-established, prosperous, supportive of the hero. On the other hand, when women are alone, they obtain a "Little Orphan Annie" quality: lonely, fragile, in need of help. And when they do have parents or siblings, these are invalids, mental cases, in trouble with the law; rather than providing a support system for the heroine, they constitute a source of further distress.
22. ii. Reunion. More than a quarter of the novels include a reunion (see Table I): The protagonists' marriage had been bad, they separated, but now they have drastically changed and will live in marital bliss. However, such a development has little support in reality. In most of the cases years have passed since the first failure, the spouses are more likely to have become more rigid in their ways than before, and apparently neither partner has been able to establish a lasting relationship. The reunion novels raise unrealistic expectations by suggesting that such a marriage has a viable chance.
23. While the typical picture of courting, spousehood and family life found in folk tales certainly does not constitute a model worth emulating, their readers are at least spared prolonged conversations between the main protagonists. The prince and the lowly maiden may dance until midnight, but they do not provide an active model of interpersonal communication. Romantic fiction, however, is essentially a collection of dialogues, with barely changing scenery and plot. The novels in our sample can easily serve as a model of pathological spousal communication (see Moore & Kramer, 1999/2000). Dialogues in romantic fiction well represent the four incongruent patterns of communication identified by Satir (1967; see also Satir, Stachowiak, & Taschman, 1975): Blaming, placating, being irrelevant or super-reasonable. Blamers try to gain strength and self esteem through the belittling of the other. Placating involves the effacing of one's self. The irrelevant disbelieve the possibility of gaining self worth through being heard and evade any confrontation, investing all energy in side-tracking. The super-reasonable receive power from pretending to know it all, thus making their audience feel ignorant and foolish. Instead of expressing what a person experiences and requests, incongruent communication contains masked, manipulative messages, and its senders try to force the audience to comply without exposing their own vulnerability. In the fifth, congruent pattern, communicators do not manipulate, they share their thoughts and emotions about themselves without projecting them onto others, and minimize the gap between words and intentions. Congruence is crucial for intimate communication.
24. The most frequently observed pattern (see Table IV) is the blaming one for both men and women (highest in Taming-of-the-Shrew-type stories; see Table III); it is also the modal form of dyadic communication: In 36% of the novels two blamers frustrate each other. While there is a considerable number of congruent communicators (more women than men, and most of them in Cinderella-type novels), only a few of them were fortunate enough to meet: Only 9% of the novels have couples sharing congruent communication with each other.
25. There are two more noteworthy findings with respect to communication patterns. In keeping with traditional gender stereotyping, women are three times as likely to be placators as men; this pattern occurs most often in Beauty-and- the-Beast-type novels, less so among the Cinderellas, and is totally absent from Taming-of-the-Shrew-type stories. As for the absence of meaningful dialogue: For one quarter of the men, and about a fifth of the women communication never surpasses the "pass-the-salt" level. In other words, in 91% of the novels read, conversation does not serve for negotiation; it is, at least for one of the partners, either a tool for fighting, or is so technical as to be psychologically meaningless. Instead of encountering examples of each person's ability to look inside, to use self-, rather than other-oriented messages, to acknowledge their motives and feelings, and to share these feelings with others in intimate communication (see Satir, Stachowiak, & Taschman, 1975), readers receive legitimization for bickering, lying, masking, being non-authentic. These pathological norms appear not only in the spousal dyad, but extend also to parents, children, siblings, vis a vis significant topics. We must remember that this communicational wasteland is described as characteristic of each couple at the stage of building their relationship for a lifetime. We do not know what is the extent of the effect on readers; if it is strong, these books teach their public how to fail in their marriage.
26. In addition to examining patterns of communication, we have also looked at its manifest content. The major impression one receives when reading these books, is one of an ongoing battle: Encounters are dominated by misunderstandings, petty jealousies, faking indifference and hiding emotions and thoughts. Every family therapist would predict utter failure for such spousehood, in obvious contradiction to the "happily ever after" promise of the last few pages. The pathogenic nature of communication in romantic fiction is especially vivid with respect to concealment: More than one half of the novels in our sample contain a significant lie or secret between the two main protagonists; in the Taming-of-the-Shrew-type stories this phenomenon reaches 68%. The message is clear: Lies are legitimate in love. The danger inherent in this finding becomes apparent when contrasted with such sources as Bowen (1976) and Satir (1972), according to whom healthy couples tend to speak openly of every topic of concern, including painful, unpleasant or frustrating subjects. Bowen (1976) claimed that it is pre-marriage communication that is most open, gradually closing up and eating into intimacy in subsequent years. In romantic fiction even the courting stage (i.e. 99% of the book) is full of closed topics, taboos, withholding of vital information.
27. In interpersonal communication there is a continuum of expressing anger, from acting it out to being paralyzed by it. The most severe, non-sublimated form of communicating anger is through physical abuse. In our sample approximately a quarter of the protagonists (males: 27%, females: 21%) lose control over their behavior, and slap, spank, kick or shove their partners, pull their hair or throw objects at them. Male aggression often involves sexual abuse, ranging from a clearly painful, unwanted embrace or kiss, to out-and-out rape. In keeping with both their "taming" and "shrew" aspects, novels using the Taming-of-the-Shrew motif have about three times as much physical aggression (both male and female) as the rest.
28. Another outlet for the expression of anger is through verbal aggression. In the novels read this includes yelling at the other person, as well as insulting, threatening, or extorting them. Common to all of these is the aggressor's need to belittle the other person, to diminish the other's self-esteem. In romantic fiction verbal abuse is even more rampant than physical (with Taming-of-the-Shrew-type novels producing the lion's share): Nearly one half of the male heroes, and more than a quarter of the female ones, verbally aggress their partner.
29. Several disturbing themes, not unrelated to each other, arise with respect to these findings. First, romantic fiction confuses love and control. Aggression is an interpersonal control mechanism; in these novels loving a person legitimizes control over him or her. Men mistakenly think, wrote Roy (1982a), that violence earns the respect and love of women. In the genre under discussion this notion is shared by the female protagonists, as well, whether they are the source of aggression or its target.
30. Second, the widespread domestic violence present in these books naturalizes it, surreptitiously turns it, to use Norris's argument (1988, p. 13) in an analogous situation, from de facto to de jure. To be fair, it is not romantic fiction alone that is responsible for what has become normative violence in American families (see Brush, 1993); yet it certainly adds its share to the public's overall exposure to violence in the media (cf. Kurz, 1993).
31. Third, and perhaps most disturbing, is the protagonists' reaction to aggression, regardless of whether it was perpetrated on them or by them. Aggression is not pilloried in romantic fiction, leading Greer 1971, p. 183 to declare that in some of these novels "vulgarity is a strength". It is not even discussed: While on page 82 the hero may have cruelly humiliated the woman he will marry, on page 83 neither of them brings up this incident, there is no questioning, analysis or attempt at insight. This model of keeping silent about domestic violence is a familiar one: Both Daley (1981) and Kelly & Radford (1996) note that women in intimate relationships refuse to acknowledge abuse and rape, hold on to their marriage at all cost, and tend to blame themselves for any disruption of domestic tranquillity. According to Kelly & Radford (1996) women tend to minimize violence fro m men. Men control women through violence, accompanied by degrading, belittling talk. We found in our sample 15 novels that are particularly high in violence (seven Cinderella-type, two Beauty-and-the-Beast-type, and six, that is to say, one third, of The Taming-of-the-Shrew variety), and in which exactly this type of control is exercised. Similarly to the situation we have witnessed with respect to communication (see # 25, above), these couples exhibit signs of pathology even prior to marriage: He uses extortion to get her, causes utter financial and emotional dependency on him, grinds her into the ground. By the end of the story, her self esteem has become so low that she cannot but be grateful for his willingness to marry her, because her other alternative is desolation and misery.
32. Though being a far cry from presenting salutogenic family dynamics, there are in the romantic fiction of the nineties some positive, progressive themes. It is unfortunate that these sporadic occurrences are all undercut by an overwhelmingly traditional, patriarchal value system, exuding from the rest of each novel.
33. Pluralism. There is some pluralism in family models, though the mores of the 90's have made only a slight impression on romantic fiction. About two thirds of the main protagonists have never been married. Twenty eight percent of the men, fifteen percent of the women are divorced, and unwed parenthood is a rarity (see Table II). Whether these percentages are realistic or not, is beside the point (they are not). What is important to point out is that the relatively large percent of divorced protagonists at least acknowledges one aspect of current social realities. This trend barely extends to unmarried parenthood, and is totally absent with reference to two other aspects of social pluralism: ethnicity and sexual preference. The world of contemporary romantic fiction is populated with white, Anglo-Saxon (not necessarily Protestant, for religion is totally absent from this genre) heterosexuals, whose main concession to the nineties is the unhindered use of condoms. The prevalence of divorces has some further interesting aspects. On the one hand, whenever the reason for a protagonist's divorce is discussed, it is made clear that he or she were married to monsters, who were entirely and solely to be blamed for the break-up of marriage. There is no legitimization given to equally shouldering the blame, to recognizing one's shortcomings, to divorce with mutual consent. On the other hand, five percent of the novels feature women who had been abused in a former relationship and were able to break out. The implication of these few cases is that one does not have to remain in a marriage at all costs.
34. Widening horizons. Jones (1986, pp. 215-216) found that the universe represented by the romance was expanding: Compared to what had been the agenda a few decades earlier, there was more emphasis on woman's work outside home, one could find a critique of male supremacy in office and bedroom, there was less distance between male and female conceptions of desire. A "widening of horizons" was also detected by Anderson (1974, p. 248). However, Snitow's (1979) reading of the novels' message is more to the point: "Small surface concessions are made to a new female independence (several researchers, misreading I believe, claim that the new heroines are brave and more interested in jobs than families) but the novels only mention the new female feistiness to finally reassure readers that 'plus ca change, plus c'est la meme chose'". This poignant observation by Snitow may be used as a key for understanding the reference by numerous sources to the contradictory nature of the genre (e.g. Cherland & Edelsky, 1993; Christian-Smith, 1990, p. 130; Light, 1984, Modleski, 1982, p. 112; Radway, 1991, p. 211).
35. Mixed findings. As the quantified results show, while the hypothesized trends are clear, most of the findings contain some off-diagonal cases: Along with a major trend of male superiority, there is also a minor note (i.e. a small number of novels) of equality (also noted by Margolies, 1982), or of female ascendance. Yet this may well be an illusion, for equality in one area, say education, is quickly 'compensated for' by inequality in several others.
36. In their discussion of optimally functioning families, both Beavers (1976) and Skynner (1982) emphasized their democratic, egalitarian, flexible nature. As we move from optimal to adequate, and from there to dysfunctional, chaotic families, we encounter traditional values, marked dominance of one spouse and passivity of the other, a lack of respect for individuality, and an absence of clear, open, frank communication. As the above analysis of the material contained in contemporary romantic fiction has shown, its family related norms and values are thus antithetical to those commonly accepted by family therapy. The stereotypical, gender-based roles encouraged by this genre, along with its pathogenic communication model and naturalized violence are not only and by default a poor preparation for egalitarian, mature spousal relationships, they actively propagate a dysfunctional family model. We draw this conclusion while fully aware of Modleski's view, according to which "[I]t is useless to deplore the texts for their omissions, distortions, and conservative affirmations" (1982, p. 113).
37. The paradoxical support (qualified and equivocal as it may be) extended to romantic fiction by several feminist scholars (e.g. Apple, 1990; Bridgwood, 1986; Hallam & Marshment, 1995; Jones, 1986; Light, 1984; Smart 1991; Treacher 1988) deserves some clarification in itself. These sources acknowledge that the genre under discussion is patriarchal in character, and that its reading constitutes escapism, yet they feel duty bound to their constituency to defend the women readers from being always categorized as passive viewers of soap operas and readers of romance (Hallam & Marshment, 1995, p. 181), or labelled as "either masochistic or inherently stupid" (Light, 1984). Yet, in our view, no amount of re-signification, deconstruction and reconstruction, whether feminist, post-structuralist or post-modernist, can change the frightening and pathogenic implications of these novels. (For relevant aspects of the parallels between feminism and post-modernism see Assiter, 1996; Burman, 1998; Nicholson, 1995; Fraser 1995; Soper, 1994). We disagree with Anderson's lukewarm conclusion: "...if there is a readership which finds satisfaction in reading romantic fiction, while it may not give them any positive benefit, may not make them nobler or wiser, it is very unlikely to corrupt them" (1974, p. 263; a similar view was expressed by Snitow, 1979). Thus, we are more in tune with the views expressed by Douglas (1980), Dworkin (1983), Greer (1971) or Margolies (1982), who may acknowledge the need of many women to engage in "sterile self-deception" (Greer, 1971, p. 188), yet deplore both the conditions that have created this need and the results, the latter being self-defeating, harmful, disastrous. (See also Willinsky & Hunniford, 1993, about the danger present in this genre to young readers).
38. We must emphasize that the population of novels from which we have sampled (contemporary romantic fiction published by Harlequin Books) has nothing in common with the large corpus of excellent literature produced by female writers and widely available to the public. Kate Chopin, Virginia Wolf, Elizabeth Bowen, Flannery O'Connor, Katherine Anne Porter, Alice Munro, Fay Weldon and many others have made a tremendous contribution to a realistic depiction of spousehood, to the developing of the public's sensitivity to, and awareness of family processes, to a deeper understanding of the pain borne by women for endless generations.
39. We can only speculate about the reasons for the tremendous popularity of romantic fiction. Is this genre similar to fairy tales, in creating and maintaining hope in their readers (cf. Bettelheim, 1976, pp. 121-123)? Or, as Q. D. Leavis (quoted by Harding, 1967) has put it: Are these books "a refuge from actual life" or a means "to deal less inadequately with it?" It is more than likely that what Harding (1967) wrote more than 30 years ago, based on data he collected more than 60 years ago, is still valid: Some read for intrinsic pleasure, some to get away from it all, some feel they have earned the right to spend their time on romantic fiction, others are somewhat ashamed of this self-indulgence. Yet one has to keep in mind the comparison Harding drew between entertainment and drugs: Both manipulate mood without addressing the underlying problem.
40. About one hundred years ago, the essayist and writer W. D. Howells (1899/1989, pp. 269-270) attributed the popularity of "these inferior romanticists" to lack of self-knowledge on the part of the public. His view was echoed by Willinsky & Hunniford (1993): "With the romance the level of self-deception seems particularly high." We note with some sadness that the passing of a century has not changed any part of Howell's statement.
Apple, Michael W. (1990). Introduction. In Linda K. Christian-Smith, Becoming a woman through romance, pp. i-xxx. New York: Routledge.
Assiter, Alison (1996). Enlightened women -- Modernist feminism in a postmodern age. London: Routledge.
Anderson, Rachel (1974). The purple heart throbs -- the sub-literature of love. London: Hodder & Stoughton.
Beavers, W. Robert (1976). A theoretical basis for family evaluation. In Jerry M. Lewis, Robert W. Beavers, John T. Gossett, & Virginia A. Phillips (Eds.), No single thread. New York: Brunner/Mazel.
Bettelheim, Bruno (1976). The uses of enchantment. New York: Knopf.
Bowen, Murray (1976). Family reaction to death. In P. J. Guerin (Ed.) Family therapy, pp. 335-348. New York: Gardner Press.
Bridgwood, Christine (1986). Family romances: the contemporary popular family saga. In Jean Radford (Ed.), The progress of romance -- The politics of popular fiction, pp. 167-193. London: Routledge & Kegan Paul.
Brunwald, Jan H. (1966). The folktale origin of The Taming of the Shrew. Shakespeare Quarterly, 17: 345-359.
Brush, Lisa D. (1993). Violent acts and injurious outcomes in married couples: Methodical issues in the National Survey of Families and Households. In P. B. Bart & E. G. Moran (Eds.), Violence against women, pp. 240-251. Newbury Park: Sage.
Burman, Erica (1998). Deconstructing feminist psychology. In Erica Burman (Ed.), Deconstructing feminist psychology, pp. 1-29. London: Sage.
Cherland, Meredith R., & Edelsky, Carole (1993). Girls and reading: The desire for agency and the horror of helplessness in fictional encounters. In Linda K. Christian Smith (Ed.), Texts of desire: essays on fiction, femininity and schooling, pp. 28-44. London: Falmer.
Christian-Smith, Linda K. (1990). Becoming a woman through romance. New York: Routledge.
Christian-Smith, Linda K. (1993). Sweet dreams: Gender and desire in teen romance novels. In Linda K. Christian Smith (Ed.), Texts of desire: essays on fiction, femininity and schooling, pp. 45-68. London: Falmer.
Coward, Rosalind (1984). Female desire -- Women's sexuality today. London: Paladin.
Daley, Mildred (1981). Woman battering. Beverly Hills, CA: Sage.
Douglas, Ann (1980). Soft-porn culture. New Republic, August 30, 25-29.
Dworkin, Andrea (1983). Right-wing women -- The politics of domesticated females. London: The Women's Press.
Eagly, Alice (1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Erlbaum.
Fogarty, Thomas F. (1976). Marital crisis. In Philip J. Guerin (Ed.), Family therapy, pp. 325-334. New York: Gardner Press.
Fraser, Nancy (1995). Pragmatism, feminism, and the linguistic turn. In Seyla Benhabib, Judith Butler, Drucilla Cornell, & Nancy Fraser (Eds.), Feminist contentions, pp. 157-171. New York: Routledge.
Gilbert, Pam (1993). Dolly fictions: Teen romance down under. In Linda K. Christian Smith (Ed.), Texts of desire: essays on fiction, femininity and schooling, pp. 69-86. London: Falmer.
Goldenberg, Irene & Goldenberg, Herbert (1996). Family therapy - An overview, 4th ed. Pacific Grove, CA: Brooks/Cole.
Greer, Germaine (1971). The female eunuch. St. Albans: Paladin.
Guerin, Philip J., Fay, Leo F, Burden, Susan L., & Kauto, Judith G. (1987). The evaluation and treatment of marital conflict. New York: Basic Books.
Hallam, Julia, & Marshment, Margaret (1995). Questioning the 'ordinary' woman. In Beverly Skeggs (Ed.), Feminist cultural theory -- Process and production, pp. 169-189. Manchester: Manchester University Press.
Harding, D. W. (1967). The notion of "escape" in fiction and entertainment. Oxford Review, 4 (Hilary), 23-32.
Harlequin News Release ( January 1999). http://releases.twoten.press.net/r.
Harris, Thomas A. (1973). I'm OK -- You're OK. London: Pan.
Howells, W. D. (1989). Novel-writing and novel reading: An impersonal explanation. In Nina Baym & al. (Eds.), The Norton anthology of American literature, 3rd ed., vol. 2. New York: Norton. (Work originally published 1899).
Jones, Ann Rosalind (1986). Mills & Boon meets feminism. In Jean Radford (Ed.), The progress of romance -- The politics of popular fiction, pp. 195-218. London: Routledge & Kegan Paul.
Kelly, Liz & Radford, Jill (1996). "Nothing really happened": The invalidation of women's experiences of sexual violence. In M. Hester, L. Kelly, & J. Radford (Eds.), Women, violence and male power, pp. 19-33. Buckingham: Open University Press.
Kurz, Demie (1993). Social science perspectives on wife abuse: Current debates and future directions. In P. B. Bart & E. G. Moran (Eds.), Violence against women, pp. 252-269. Newbury Park: Sage.
Lansbury, Angela (1982). Foreword. In Maria Roy (Ed.), The abusive partner, pp. vii-viii. New York: Van Nostrand Reinhold.
Lederer, William. J. and Jackson Don D. (1968). The mirages of marriage. New York: Norton.
Light, Alison (1984). 'Returning to Manderley' -- Romance fiction, female sexuality and class. Feminist Review, 16: 7-25.
McGoldrick, Monica (1989) The joining of families through marriage: The new couple. In B. Carter, & M. McGoldrick, (Eds.). The changing family life cycle, pp. 93-119. Boston: Allyn & Bacon.
Margolies, David (1982). Mills & Boon -- Guilt without sex. Red Letters, No. 14: 5-13.
Modleski, Tania (1980). The disappearing act: A study of Harlequin romances. Signs: Journal of Women in Culture and Society, 5: 435-448.
Modleski, Tania (1982). Loving with a vengeance: Mass-produced fantasies for women. New York: Routledge.
Moore, M. & Kramer, D. (1999/2000). Satir for beginners: Incongruent communication patterns in romantic fiction. Et Cetera - A review of General Semantics, 56, 429-437.
Nicholson, Linda (1995). Introduction. In Seyla Benhabib, Judith Butler, Drucilla Cornell, & Nancy Fraser (Eds.), Feminist contentions, pp. 1-16. New York: Routledge.
Norris, Christopher (1988). Deconstruction, post-modernism and the visual arts. In Christopher Norris & Andrew Benjamin (Eds.), What is deconstruction? pp. 7-31. London: Academy Editions.
Radway, J. (1983). Women read the romance: The interaction of text and context. Feminist Studies, 9, no. 1.
Radway, J. (1991). Reading the romance: Women, patriarchy, and popular literature. Chapell Hill: University of North Carolina Press.
Reed, Gail S. (1993). Candide: Radical simplicity and the impact of evil. In Emanuel Berman (Ed.), Essential papers on literature and psychoanalysis, pp. 403-413. New York: NY U Press.
Roy, Maria (1982a). The nature of abusive behavior. In Maria Roy (Ed.), The abusive partner, pp. 3-16. New York: Van Nostrand Reinhold.
Roy, Maria (1982b). Four thousand partners in violence: A trend analysis. In Maria Roy (Ed.), The abusive partner, pp. 17-35. New York: Van Nostrand Reinhold.
Satir, Virginia (1967). Conjoint family therapy, revised ed. Palo Alto, CA: Science and Behavior Books.
Satir, Virginia, Stachowiak, J. & Taschman, H. A. (1975). Helping families to change. New York: Jason Aronson.
Skynner, Robin (1982). Frameworks for viewing the family as a system. In A. Bentovim, G. G. Barnes, & A. Cooklin (Eds.), Family therapy. pp. 3-36. London: Academic Press.
Smart, Carol (1991). Penetrating women's bodies: The problem of law an medical technology. In Pamela Abbott & Claire Wallace (Eds.), Gender, power and sexuality, pp. 157-175. Basingstoke: Macmillan.
Snitow, Ann Barr (1979). Mass market romance: Pornography for women is different. Radical History Review, 20: 141-161.
Soper, Kate (1994). Feminism, humanism and postmodernism. In Mary Evans (Ed.), The woman question, pp. 10-21. London: Sage.
Stordeur, Richard A. & Stille, Richard (1989). Ending men's violence against their partners. Newbury Park: Sage.
Treacher, Amal (1988). What is life without my love: Desire and romantic fiction. In Susannah Radstone (Ed.), Sweet dreams -- Sexuality gender and popular fiction, pp. 73-90. London: Lawrence & Wishart.
Walkerdine, Valerie (1984). Some day my prince will come: Young girls and the preparation for adolescent sexuality. In Angela McRobbie & Mica Nava (Eds.). Gender and generation, pp. 162-184. Basingstoke: Macmillan.
Walkerdine, Valerie (1990). Schoolgirl fictions. London: Verso.
Weber, Max (1957). The theory of social and economic organizations, edited by G. T. Parsons. New York: Oxford U Press. (Originally published 1922).
Wetherell, Margaret (1995). Romantic discourse and feminist analysis: Interrogating investment, power and desire. In Sue Wilkinson & Celia Kitzinger (Eds.), Feminism and discourse -- Psychological perspectives, pp. 128-14 4. London: Sage.
Whissell, Cynthia (1996). Mate selection in popular women's fiction. Human Nature, 7: 427-447.
Willinsky, John, & Hunniford, R. Mark (1993). Reading the romance younger: The mirrors and fears of a preparatory literature. In Linda K. Christian Smith (Ed.), Texts of desire: essays on fiction, femininity and schooling , pp. 87-105. London: Falmer.
Wilson, Elizabeth (1983). What is to be done about violence against women? Harmondsworth: Penguin.
Yllo, Kersti (1994). Reflections of a feminist family violence researcher. In Donna L. Sollie & Leigh A. Leslie (Eds.), Gender, families, and close relationships, pp. 213-236. Thousand Oaks, CA: Sage.