Biased Use of Cross-Cultural and Historical Perspectives on Male Homosexuality in Human Sexuality Textbooks.

A quarter century ago, distinctions between man-man sex and man-boy sex were not as sharp as they are today; both were generally regarded as criminal and pathological (e.g., Gebhard, Gagnon, Pomeroy, & Christenson, 1965). In the courts, man-man sex was characterized with terms such as "loathsome and disgusting" (Knutson, 1979/80); in psychiatry, terms such as "dread dysfunction" were used (e.g., Socarides, 1970). An important reason that man-boy sex was seen as a crime and a pathology was the belief that it turned boys into homosexuals (e.g., Masters, 1962). In the 1970s, in the wake of gay liberation and the women's movement, distinctions between the two sharpened. On the one hand, man-man sex became more tolerated in liberal circles, was depathologized by the American Psychiatric Association, and became increasingly decriminalized. On the other hand, man-boy sex became even more anathematized, pathologized, and criminalized (cf. Finkelhor, 1984; Okami, 1990). Based on the rape and incest models advanced by the women's movement, man-boy sex was now seen as pathological because it was viewed as a form of power abuse, producing intense psychological disturbance (Okami, 1990).

Consistent with this new perspective, Masters, Johnson, and Kolodny (1985) drew sharp moral distinctions between man-man sex and man-boy sex in an early edition of their textbook Human Sexuality. On the one hand, they presented man-man sex (i.e., homosexuality) as normal and healthy. On the other, they presented man-boy sex (i.e., pedophilia) as pathological and harmful. In a lengthy discussion entitled "Is There a Positive Side to Pedophilia?" Masters et al. critiqued an interview study conducted by Sandfort (1983) on a sample of 25 Dutch boys aged 10 to 16 involved in ongoing sexual relationships with men. Sandfort reported that the boys experienced their relationships, including the sexual aspects, predominantly in positive terms, that evidence of exploitation or misuse was absent, and that the boys tended to see the pedophile as a teacher, as someone they could talk to easily and with whom they could discuss their problems. Against Sandfort's findings, Masters et al. argued that the study was methodologically flawed and speculated that possibly "the boys were so intimidated by their pedophile that they were afraid to say anything against him" (p. 451). They discounted Sandfort's conclusion that the relationships were positive, arguing that man-boy relationships are "inherently abusive and exploitive" and are always negative. They asserted that they were opposed to these relationships no matter how beneficial either party claimed them to be.

Masters et al.'s (1985) critique of Sandfort's (1983) study was flawed by a series of errors, as Bauserman (1990) documented. Prominent among these was their misrepresentation of Sandfort's methodology, most likely due to their failure to consult more detailed presentations of his research (e.g., Sandfort, 1984). Aside from the errors discussed by Bauserman, an even more basic one was present in their textbook, which forms the focus of the current investigation.

MISUSE OF HISTORICAL AND CROSS-CULTURAL PERSPECTIVES ON HOMOSEXUALITY

Errors of Commission

In keeping with the view that cross-cultural and historical perspectives are needed to achieve a scientifically valid understanding of human sexuality (cf. Bullough, 1976; Ford & Beach, 1951), Masters et al. (1985) provided perspective on homosexuality by discussing a series of cultures in other times and places that approved of certain types of male-male sex. This use of perspective was problematic, however, because they used predominately pedophilic (i.e., man-boy) examples. Given that they drew sharp moral distinctions between man-boy and man-man sex in our society, condemning the former while endorsing the latter, it was highly inconsistent for them to have then used examples of the former to illustrate the potential normalcy and social acceptability of the latter.

In their chapter on homosexuality, Masters et al. (1985) stated that to "provide appropriate background for a contemporary look at homosexuality and bisexuality, we will first look hack in history" (p. 409). They then discussed three cultures that had approved of homosexual relations--all three of which were of the man-boy type rather than the man-man type. They noted that homosexuality was widely accepted as natural in ancient Greece, mostly occurring "between grown men and young adolescent boys" (p. 409, italics added)--they had acknowledged earlier that sexual relations between two adult men were "frowned upon" (p.10). They wrote further that homosexual behavior was common in the early days of the Roman Empire, including marriages between men. Their implication of widespread, socially approved man-man sex is inaccurate, however. Cantarella (1992), in a scholarly review of bisexuality in the ancient world, noted that, at the end of the Republic and during the early days of the Empire, sexual relations between men and boys "had now become in practice an absolutely normal relationship, socially accepted, engaged in with total freedom, and celebrated by the poets" (p. 141). Sexual relations between two adult males, on the other hand, were more problematic because the "passive adult male was considered effeminate and was sneered at and held in low esteem" (p. 155). As a third historical example, Masters et al. stated that Boswell (1980) found that "for many centuries Catholic Europe showed no hostility to homosexuality" (p. 409). However, as the poems Boswell presented from the Early and High Middle Ages indicate, the predominant type of homosexuality practiced and tolerated was pederasty--that is, sexual relations between men and young adolescent boys.

In a later chapter, guest author Paul Gebhard (Masters et al., 1985, pp. 620-637) provided additional perspective by reviewing male homosexuality in a variety of approving societies. He discussed the widely occurring ancient Greek form, which involves a sexual relationship between an older male and a younger adolescent male in which the older male also acts as a teacher and guardian. Gebhard noted that this type of male-male sexuality is found around the Mediterranean; in many societies in the Middle East, Africa, and Asia; and in some Melanesian islands. In addition to this, he discussed three other examples of the man-boy type (i.e., Sambia, East Bay, and Java). He also discussed two examples of the man-man type, the berdaches and xaniths, who were effeminate males playing the passive role in sexual relations with masculine males in Native American societies and in Oman, respectively. Primary sources reveal that these latter two examples also illustrate the man-boy type, because berdaches and xaniths usually began their sexual careers with adult men when they were young adolescents, if not younger (Wikan, 1982; Williams, 1992).

In short, Masters et al. (1985) included in their textbook nine historical and cross-cultural examples of societies approving of male-male sex to provide perspective on homosexuality. All nine were relevant to the man-boy type but only two were relevant to the man-man type. Given their unqualified condemnation of man-boy sex in our society, it was inconsistent to use predominately man-boy examples from other times and places to inform the issue of man-man sex in our society. This bias represents an error of commission, using examples to inform issues with which they are not relevant according to the authors' own moral position.

Errors of Omission

Despite including nine historical and cross-cultural examples that were all relevant to man-boy sex, Masters et al. (1985) did not use any of them to provide perspective when discussing man-boy sex in our culture. This omission is problematic because these examples could have informed their critique of Sandfort's (1983) study. Masters et al. speculated that the boys in this study were too intimidated to say anything against their pedophile and that their experiences were all negative. Primary source material from the cross-cultural examples appearing elsewhere in their textbook, however, questions such speculations. Davenport (1965) reported that East Bay Islander boys aged 7 to 11 who had sexual relations with married men would then "discuss their homosexual activities freely and without shame in the presence of their parents and friends" (p. 200). Omani boys aged 12 or 13, after discovering their homosexual interests, actively sought sex with men as prostitutes (i.e., xaniths); this pursuit reflected sexual interest rather than economic need (Wikan, 1982). Williams (1996) reported that the Native American men he interviewed who had boyhood sexual relations with significantly older males typically reflected on these experiences with fond memories, regardless of their current sexual orientation. Such findings, had they been discussed in relation to Sandfort's study, could have substantially informed any speculations offered. Additionally, the ancient Greek form of man-boy sex discussed by Masters et al. and Gebhard (1985), with its functions of pedagogy and guidance, could have informed Sandfort's report that the boys saw their pedophile as a teacher and as one with whom they could discuss their problems.

These omissions were errors because their use could have made Sandfort's (1983) report, so much at odds with current Western beliefs, more intelligible. Despite the suitability of the historical and cross-cultural data for Sandfort's results, it is important to consider limitations on their use for providing perspective on contemporary Western man-boy sex. Sandfort examined an unrepresentative sample of man-boy relations--those that were predominantly positive. Clinical research has presented a much different picture characterized by a wide variety of negative reactions in response to these relations, such as feelings of guilt or shame, symptoms of depression or anxiety, and sexual identity concerns (Mendel, 1995; Urquiza & Capra, 1990; Watkins & Bentovim, 1992). Clinical samples are also unrepresentative, however, as three recent reviews of the nonclinical literature have shown (Bauserman & Rind, 1997; Rind & Tromovitch, 1997; Rind, Tromovitch, & Bauserman, 1998). In a review of 59 studies based on college samples, Rind et al. found that boys' sexual experiences with significantly older individuals, many of whom were males, were 37% positive, 29% neutral, and 33% negative. Girls' sexual experiences with older persons, on the other hand, were predominately negative (72%), and only occasionally positive (11%) or neutral (18%). Moreover, male students, who as boys were willing partners in these relations, were psychologically as well adjusted as male students who had no such experiences. On the other hand, male students who were unwilling in these relations or female students with these experiences regardless of level of willingness were somewhat less well adjusted than controls. The other two reviews, one of which focused on nationally representative samples (Rind & Tromovitch, 1997), reported comparable reactions, psychological correlates, and sex differences.

The findings from these recent reviews imply that willing man-boy sex accompanied by positive reactions may be better informed by the ancient Greek model than by models based on the female experience (e.g., rape and incest models). For unwilling man-boy sex with negative reactions, on the other hand, the latter models may be more appropriate, in which case the cross-cultural and historical perspectives seem much less informative. In short, the data on psychological correlates of and reactions to man-boy sex suggest that the cross-cultural and historical data have value for providing informative perspective on a nontrivial proportion of such cases.

A BRIEF REVIEW OF CROSS-CULTURAL AND HISTORICAL DATA ON MALE HOMOSEXUALITY

The claims of errors of commission and omission made previously are further supported by a brief review of male-male sex across time and place. Scholarly reviews of cross-cultural male homosexuality have shown that it has generally appeared in three main forms: transgenerational, transgenderal, and egalitarian (Adam, 1985; Carrier, 1980; Ford & Beach, 1951; Greenberg, 1988; Gregersen, 1983; Herdt, 1991a, 1991b; Murray, 1992; Trumbach, 1977, 1989; Williams, 1992). Transgenerational homosexuality involves sexual relations between boys or youths and significantly older males (i.e., older adolescents or adults). Transgenderal homosexuality refers to sexual relations between a masculine male and a cross-gendered male who takes on a nonmasculine gender role and acts as the passive partner. Egalitarian homosexuality consists of sexual relations between males who do not change gender roles and who are similar in age and social status. These relations typically occur between two boys or two youths, usually do not exclude heterosexual relations or marriage, and have no implication for gender identity or social identity based on sexual orientation. It is noteworthy that these scholarly reviews draw no moral distinctions between man-boy and man-man sex. Man-boy sex is presented as a form of homosexuality that often serves useful social functions rather than being classified as a form of pedophilia, socially and individually destructive in nature and hence intolerable.

These reviews have concluded that the "gay" pattern in the modern West, which involves egalitarian relations between men whose sexual interests and behaviors are generally restricted to other adult men, has rarely appeared as a pattern in other times and places. Gregersen (1983) concluded that the Western homosexual pattern "is quite exceptional from a cross-cultural point of view and seems to be a fairly recent phenomenon" (p. 297). Herdt (1987) noted that the "Gay Movement is a totally new social phenomenon" (p. 204). Adam (1985) concluded that the current structure of the gay world shows a radical break with cross-cultural forms and merits its own analysis. The reviews further indicate that, outside Western culture, boyhood sexual involvements with other males, both peers and elders, are well within the range of normal experience. Adam concluded from the cross-cultural data that there is "a special propensity for homosexual relations among unmarried male youths" (p. 20). Finally, the data indicate that transgenerational homosexuality (i.e., man-boy sex) has been the most common form of same-sex relationship in which adult males cross-culturally and historically have been involved (Herdt, 1995).

These conclusions imply that the unqualified use of historical and cross-cultural data on male homosexuality to provide perspective on Western man-man sex, as Masters et al. (1985) did, is inappropriate because the gay pattern is different. Further implied is that use of these data for providing perspective on Western man-boy sex, which Masters et al. did not do, can be informative because of the commonness and normalcy of boy-boy and man-boy sex in other cultures. Given these findings, Sandfort's (1983) conclusions become less surprising. It is important to add that the cross-cultural and historical data provide little, if any, data that would be relevant to age-differentiated female homosexuality.

Transgenerational Homosexuality

Numerous societies, past and present, have accepted, even institutionalized, sexual relations between boys and older males. Adam (1985) described two major categories of transgenerational homosexuality: the ancient model and the Melanesian model. In the former, often referred to as pederasty, the relationship generally occurs between an adult male and an adolescent boy; the relationship often, but not always, takes on the social function of an "erotic apprenticeship," in which the older male mentors the boy and helps him develop into an adult. The relationship ends when the boy reaches maturity. For example, in ancient Greece pederastic relations lasted from the time the boy was 12 until he was about 17 or 18 (Cantarella, 1992; Percy, 1996), and in early modern Japan these relations went from the time the boy was 11 to about 19 (Saikaku, 1990; Schalow, 1989). In Europe, in addition to ancient Greece, socially sanctioned pederastic relations occurred among certain ancient Germanic and Celtic tribes, and appear to have been common among the early Indo-Europeans (Bremmer, 1980; Greenberg, 1988). Highly romanticized pederastic relations were deeply rooted in Albania under Ottoman rule, and continued to the beginning of this century (Bremmer, 1980; Greenberg, 1988; Trumbach, 1977). Evidence suggests that pederasty was prevalent among the pre-Roman Etruscans (Greenberg, 1988), and it was accepted as normal and widely practiced later in ancient Rome (Cantarella, 1992). Boswell (1980) provided evidence that pederastic relations were tolerated in Europe until the 13th century. Although repressed after that time, pederastic relations remained the dominant form of homosexuality in Europe until the industrial revolution (Greenberg, 1988; Saslow, 1989; Trumbach, 1989).

Outside Europe, pederastic relations were common and accepted in certain societies in Africa, including the Mamlukes of medieval Egypt; Islamic societies across North Africa (e.g., Morocco, Siwa) and the Azande, Mongo, Mossi, Thonga, and Zulus (Greenberg, 1988; Gregersen, 1983). In the Near East and central Asia, pederasty was also common and accepted in Islamic societies (e.g., Afghanistan, Moslem India, Pakistan, Persia, Turkey, and the Usbek Khans) (Baldauf, 1988; Greenberg, 1988). In the Far East, pederasty was common and widely practiced, not only in early modern Japan, but also in imperial China, 17th century Siam, ancient Korea, and parts of Indo-China (Adam, 1985; Greenberg, 1988; Gregersen, 1983; Hinsch, 1990). Pederasty was socially sanctioned and practiced in certain areas of Java (Gebhard, 1985), Sumatra (Money & Ehrhardt, 1972), Melanesia (Adam, 1985), Polynesia (Greenberg, 1988), and in various Native American societies in the New World (Greenberg, 1988; Herdt, 1987; Williams, 1992).

The Melanesian model occurred in various societies throughout Melanesia, Australia, and Amazonia (Adam, 1985; Herdt, 1987). The essential features of the Melanesian form are that sexual relations occur between prepubescent boys from about 7 to 13 and adolescent boys or young men, and these relations are socially sanctioned generally because they are considered essential to "growing" the boys, providing them with the semen they lack (Herdt, 1987). Semen is transmitted orally, anally, or through body smearing, depending on the society, and is believed to mature the boys physically and is often believed to transmit hunting or warrior prowess to them as well (Herdt, 1987, 1991b). Keesing (1982) and Herdt (1987) noted that practices remarkably similar to those observed in Melanesia occurred among certain Native American tribes in Amazonia. Williams (1992) recounted similar age-structured relations among the ancient Mayans, in which institutionalized marriages occurred between prepubescent boys and adolescent males.

Transgenderal Homosexuality

Numerous societies have also sanctioned and institutionalized sexual relations between masculine and cross-gendered males (Greenberg, 1988; Herdt, 1991a; Williams, 1992). Many Native American societies had berdaches (i.e., cross-gendered males) and held them in high esteem for their spiritual and artistic contributions, for mediating between men and women, and for being a valuable resource for helping and educating older children and adolescents (Herdt, 1991a; Williams, 1992). Berdache tendencies were considered to emerge very early in a boy's life, and a boy became a berdache typically at or before puberty. Berdaches were typically, but not always, involved in passive sex with men with a masculine gender role; berdaches did not, however, have sex with other berdaches. They often began their homosexual relations with older boys and men before puberty (Williams, 1996). Outside the New World, transgenderal homosexuality has been documented among the reindeer-herding peoples of Siberia, related peoples of the Bering Sea area of Alaska (e.g., Aleuts, Kodiak Islanders), various Polynesian societies (e.g., old Hawaii, Marquesas Islands, Samoa, Tahiti), Hindu India, various societies in Africa (e.g., Lango, Zanzibar, Zulus), and the Omanis in the Middle East (Carrier, 1980; Williams, 1992). In these societies, males also tended to begin their cross-gendered status and same-sex behavior at a young age: between 10 and 15 years old among the Kodiak Islanders, at 12 or 13 years old among the Omanis, and before puberty among the Polynesians (Wikan, 1982; Williams, 1992, 1996).

THE CURRENT STUDY

The errors of commission and omission made by Masters et al. (1985) may be most readily explained as biases stemming from their differing moral positions on Western man-man sex and man-boy sex, respectively. As previously discussed, these differing positions emerged in our society in the 1970s and have become predominant since then. The question arises of how common these biases are in more recent human sexuality textbooks. Rind (1995) demonstrated wide-spread bias among human sexuality textbooks in their coverage of psychological correlates of nonadult-adult sex, in which they typically exaggerated negative correlates and inappropriately generalized from the experiences of girls to those of boys. Based on these results, as well as the strong moral distinctions made between man-man and man-boy sex in the post-1970s era, it was expected that other authors of human sexuality textbooks would also tend to commit errors of commission and omission in their use of historical and cross-cultural perspectives on male homosexuality. It was the purpose of the current study to examine this possibility.

METHOD

Sample of Textbooks Included

Eighteen human sexuality textbooks published in the U.S. were obtained for analysis (see Table 1), including the 14 used in the Rind (1995) study plus 4 new ones (Carroll & Wolpe, 1996; Francoeur, 1991; Greenberg, Breuss, & Mullen, 1993; Zgourides, 1996). Newer editions for half of the original textbooks, as well as the 4 new ones, were obtained from publishers at the 1996 and 1997 Eastern Psychological Association conferences. Dates of publication for the 18 textbooks ranged from 1987 to 1997, with a mean of 1993.5 (SD = 2.79). The distribution of dates was negatively skewed, with most textbooks (72.22%) being published between 1993 and 1997. Although more than 18 human sexuality textbooks were on the market during the period from which the current sample was drawn, the obtained sample contained the most widely used textbooks in colleges and universities, indicating its practical importance.(1)

Table 1. Sample of Human Sexuality Textbooks Analyzed for Use of
Cross-Cultural and Historical Perspective on Homosexuality

Textbook                       Year of       Pages with Examples(a)
                             Publication

Allgeier & Allgeier            1995         14, 21-22, 488-89
  (4th ed.)
Byer & Shainberg               1994         36, 374
  (4th ed.)
Carroll & Wolpe                1996         8-13, 17, 20, 22,
  (1st ed.)                                 247-48, 250-52, 552
Crooks & Baur (5th ed.)        1993         275
Francoeur (2nd ed.)            1991         34, 38-39, 441-44
Greenberg, Breuss,             1993         387
 & Mullen (1st ed.)
Haas & Haas (3rd ed.)          1993         10, 403, 425
Hyde & DeLamater               1997         3, 12, 19, 34,
  (6th ed.)                                 396-97, 526-27
Katchadourian (5th ed.)        1989         368, 573-74, 576
Kelley & Byrne (1st ed.)       1992         12, 199, 320, 325, 331
Kelly (4th ed.)                1994         26, 313-14, 319
Luria, Friedman,               1987         9, 24-26, 436-38
  & Rose (1st ed.)
Masters, Johnson, &            1995         9-10, 373-74, 376-79
  Kolodny (5th ed.)
Rathus, Nevid, &               1997         3, 12-14, 370-71
  Fichner-Rathus (3rd ed.)
Strong & DeVault               1997         25-27
  (2nd ed.)
Turner & Rubinson              1993         12-13, 254
  (1st ed.)
Wade & Cirese (2nd ed.)        1991         16, 552
Zgourides (1st ed.)            1996         122, 131

(a) Listed pages contain mentions or discussions of cross-cultural or historical examples of societies with sanctioned forms of homosexuality, or of reviews of these societies.

Stimulus Material

Analysis of the use of perspective in the textbooks was based on finding all included historical or cross-cultural examples of specific societies or groups of related societies that approved of and practiced transgenerational, transgenderal, or egalitarian homosexuality, and then evaluating the appropriateness of their use. For each textbook, all pages containing these examples were located by (a) searching the index for the key words historical perspective, cross-cultural perspective, homosexuality, and pederasty,; (b) searching the index for well known examples (e.g., ancient Greece, New Guinea); (c) examining the chapters in which these pages were found for further examples; and (d) examining the introductory chapter, the chapter on homosexuality, and the chapter that included coverage of pedophilia. All relevant pages were photo-copied for later evaluation by judges. The textbooks themselves were also used in a separate analysis to establish whether a given textbook presented man-man sex and man-boy sex in our society as morally and conceptually distinct, as Masters et al. (1985) did. If so, then its use of historical and cross-cultural examples of the latter to provide perspective only on the former could then be logically and validly viewed as bias. The chapters on homosexuality and pedophilia were examined to assess whether such distinctions were drawn.

Coding Instrument

A coding instrument was developed to evaluate the use of perspective in the textbooks. A separate standard sheet, consisting of rows and columns, was used for each textbook. On each row, judges recorded information about a mention of a specific society or a group of related societies. Separate mentions of the same example were coded in separate rows. In the first few columns, the names of the examples to be coded along with their location (i.e., chapter, page number) were already provided for the judges. Each remaining column was headed by instructions informing judges how to code a given example. Judges coded the typology of the example (transgenerational, transgenderal, egalitarian, or some combination of these), how it was described (e.g., man-man sex, man-boy sex, not described), how it was labeled (homosexual, pederastic, pedophilic, not labeled), and why it was used (e.g., to provide perspective on gay sexuality, pedophilia, or children's sexuality, or to discuss origins of homosexuality). In the last column, judges answered whether its use was appropriate (yes, no, or mixed). On a separate sheet of paper "appropriate" was defined. If a transgenerational example was used to provide perspective on Western homosexuality, its use was inappropriate. If this example was used to address specific issues deemed to be relevant to Western homosexuality, such as the origins of homosexual orientation, then its use was appropriate (e.g., the Sambians, with their continual same-sex activities from age 7 to their early 20s, are relevant to theories of origin such as learning theory or script theory). If this type of example was used to provide perspective on Western man-boy sex, it was also appropriate. A transgenderal example was recorded as appropriate if used for perspective on Western homosexuality. A mixed rating could be used if the example was perceived by a judge to have both appropriate and inappropriate aspects.

At the bottom of each sheet judges indicated whether they thought the textbook drew moral distinctions between man-man sex and man-boy sex in our society. Evidence that it did, specified by instructions on a separate sheet, included (a) labeling man-man sex homosexuality and presenting it as normal, but labeling man-boy sex pedophilia and presenting it as pathological; (b) explicitly distancing the former from the latter by claiming, for example, that homosexual men are unlikely to have sex with boys and disdain men who do; and (c) grouping man-boy sex morally and conceptually with rape, incest, and/or man-girl sex.

Coding Procedure

Four judges familiar with the previously presented cross-cultural and historical data on homosexuality coded the textbooks. Another professor who is also a sex researcher, a graduate student in human sexuality, a lawyer with an advanced degree in history and knowledgeable about sex in history, and the author acted as judges. Each judge independently read the material from the 18 textbooks and coded it. Except for the author, the judges were masked with respect to the names of the textbook authors when coding the photocopied material. Judges worked on the photocopied material first and then evaluated the actual textbooks for evidence of drawing moral distinctions between man-man and man-boy sex in our society. After coding all the material, the judges used instructions from a separate sheet to assign values to the entries in 3 of the columns. For codings on how the example was described, instructions were that transgenerational examples were to be assigned a 0 (i.e., biased) if they were described as relations between men, a 2 (i.e., unbiased) if they were described as relations between boys and older males, and a 1 (i.e., mixed) if they were not described. Transgenderal examples were assigned a 2 regardless of actual description. For codings on how the example was labeled, based on the assumption that the textbook referred to man-man sex as homosexuality and man-boy sex as pedophilia in our society, transgenerational examples were to be rated as 0 (i.e., biased) if they were labeled homosexual, 1 (i.e., mixed) if they were not labeled, and 2 (i.e., unbiased) if labeled pederastic or pedophilic. For codings on whether the example was appropriate, codings judged inappropriate were assigned 0, mixed were assigned 1, and appropriate were assigned 2. In a second step, the judges shared their results with one another and attempted to resolve discrepancies through discussion, which involved going hack to the scholarly literature reviews on cross-cultural and historical homosexuality, if necessary, to argue for a particular coding.

In addition to coding examples, as discussed above, the author recorded the number of times scholarly reviews of cross-cultural or historical homosexuality (e.g., Carrier, 1980; Ford & Beach, 1951) were cited in the chapter on homosexuality and in the chapter discussing pedophilia. The rationale was that these reviews are at least as much about man-boy (transgenerational) sex as man-man (transgenderal) sex. Hence, they provide legitimate perspective for both types of sex in our culture. To use them for man-man sex, but not man-boy sex, would then constitute bias.

Interjudge Reliability

Percent agreement was computed for 5 of the codings: typology, description, labeling, appropriateness, and drawing distinctions. Across the 18 textbooks, judges assigned values to 119 examples of specific or related societies. For these examples, mean interjudge percent agreements were 87% or higher for typology, description, and labeling, while interjudge reliability alphas were .95 or higher. Disagreements for these codings were resolved by discussion. For appropriateness, mean interjudge percent agreement was 85%, while interjudge reliability alpha was .97. For this coding, differences were retained: The appropriateness value of an example became the average value of the four codings. Agreement was 100% for drawing distinctions.

RESULTS

Drawing Distinctions

All 18 textbooks drew moral and conceptual distinctions between man-man sex and man-boy sex in our society. The former, labeled homosexuality, was presented as normal and acceptable. The latter, labeled pedophilia, was presented as pathological and harmful and was discussed along with other topics such as rape, incest, and man-girl sex. These pervasive distinctions indicate that using man-boy examples from other times and places to provide perspective on man-man sex in our society constitutes bias.

Typology

Societies were classified according to their predominant type of homosexuality. Table 2 lists the 10 most frequently cited societies or groups of related societies that sanctioned male homosexuality along with their typology. Ancient Greece and Sambia were the most often used, occurring in 94.4% and 66.7% of the textbooks respectively. Nine of these 10 societies are most noted for their sanctioned transgenerational homosexuality, whereas only one is most noted for its transgenderal homosexuality--none is most noted for egalitarian homosexuality between adults. As is shown in the table, transgenerational examples were used almost exclusively in chapters discussing homosexuality or other nonpedophilic topics, despite the fact that the sexual behavior practiced in these societies would be defined as pedophilia if it occurred in our society. Altogether, chapters on homosexuality included 21 separate societies, of which 81% were transgenerational and 19% were transgenderal. Other chapters, such as the introductory chapter, contained 19 separate societies, of which 68% were transgenerational, 16% were transgenderal, and 16% were egalitarian. The egalitarian examples were of the boy-boy type. In one textbook (Carroll & Wolpe, 1996) all 9 societies mentioned in its chapter on pedophilia were transgenerational.

Table 2. Frequency of Use of Cross-Cultural and Historical Examples of Societies with Sanctioned Male Homosexuality in a Sample of 18 Human Sexuality Textbooks

                                   Frequency of Mention By:

                                  Textbook      Chapter-Type(b)

Society                Type(a)    N      %      Homo   Other   Pedo

Ancient Greece           Mb      17     94.4      9      12      1
Sambia                   Mb      12     66.7      9       5      1
Ancient Rome             Mb       8     44.4      4       5      0
Melanesia/New Guinea     Mb       8     44.4      7       1      0
Berdaches                MM       7     38.9      5       2      0
East Bay                 Mb       7     38.9      3       4      0
Siwans                   Mb       7     38.9      5       2      0
High Middle Ages         Mb       5     27.8      4       2      0
Keraki                   Mb       3     16.7      2       1      0
Imperial China           Mb       2     11.1      1       2      1

Note. Listed examples are mentioned in 2 or more textbooks. Examples mentioned in only 1 textbook are as follows. (1) Homosexuality chapter: Arabs, Azande, Kimam, Sumatra, Thailand (transgenerational); Mohave, Polynesia, xaniths (transgenderal). (2) Pedophilia chapter: Africa, Arabs, Egypt, Muslim India, Japan, Turkey (transgenerational). (3) Other chapters: Arabs, Aranda, Etoro, Islamic societies (transgenerational); Hindu India, Mayans (transgenderal).

(a) Mb = man-boy (transgenerational); MM = man-man (transgenderal). Societies were classified by predominant form.

(b) Homo = homosexuality chapter; Pedo = pedophilia chapter; Other = nonhomosexuality and nonpedophilia chapters.

Biased Use of Transgenerational Homosexuality

Given that all the textbooks drew sharp distinctions between man-boy and man-man sex in our society, it follows that they should not use cross-cultural or historical man-boy examples to provide perspective on man-man sex in our society. These examples instead should be confined to providing perspective on man-boy sex in our society, to which they are most relevant.

Test of proportions. A conservative test of bias was that the proportion of textbooks using transgenerational examples in their chapter on homosexuality would exceed the proportion of textbooks using these examples in their chapter discussing pedophilia. All of the 17 textbooks (100%) that had a chapter on homosexuality included transgenerational examples; only one (5.88%) of these textbooks (Carroll & Wolpe, 1996) used these examples in their discussion of pedophilia. McNemar's test of correlated proportions, used because paired observations from each textbook were being analyzed (cf. Glass & Hopkins, 1996), revealed a significant difference, [chi square] (1, N = 17) = 16, p [is less than] .001, with a large effect size, r = .97, indicating strong bias--rs of .10, .30, and .50, respectively, indicate small, medium, and strong effects (Cohen, 1988).

Additionally, bias would be indicated by using transgenerational examples more frequently in chapters other than those on homosexuality (e.g., introductory chapters) than in the chapters discussing pedophilia. Fourteen of the 18 textbooks (77.78%) used these examples in other chapters compared to only one usage in the 18 chapters discussing pedophilia (5.56%). This difference in proportions was statistically significant, [chi square] (1, N = 18) = 13, p [is less than] .001, with a large effect size, r = .85.

Verification of inappropriate use. Using transgenerational examples in nonpedophilia chapters does not necessarily indicate bias. For example, Hyde and DeLamater (1997) used the Sambian example to discuss learning theory and homosexual orientation; this example is significant for evaluating learning theory because boys, after spending 10 or more years engaged exclusively in homosexual behavior, almost all become exclusively heterosexual. Similarly, later in their book Hyde and DeLamater discussed ritualized man-boy sex in Melanesia to evaluate further the origins of homosexual orientation, with a focus on script theory. These instances represented appropriate use of nonwestern man-boy sex for discussing Western man-man sex. On the other hand, Hyde and DeLamater used the pederastic examples of the Siwans of Africa and the ancient Greeks to provide perspective on attitudes toward homosexuality, when these examples are relevant to attitudes about pedophilia, using the distinctions all textbook authors have drawn. This use was therefore inappropriate. To verify that the overuse of transgenerational examples in nonpedophilia chapters compared to the pedophilia chapters constituted bias, it was necessary to analyze the codings for appropriateness described previously, which could range from 0 (bias) to 2 (no bias). The mean appropriateness rating was compared to a value of 1; if it fell significantly below 1, then inappropriateness would be verified.

Two tests were conducted to examine inappropriateness. Because they were conceptUally correlated, in that the way a textbook presented examples in the chapter on homosexuality was likely to be related to the way it presented examples in chapters not on homosexuality or pedophilia, the level of significance for each was set at p = .05/2 = .025, according to a Bonferroni correction. In the 17 chapters on homosexuality, the mean value of the 52 uses of transgenerational examples was .28 (SD = .60), which fell significantly below the value of 1, t(51) = -8.66, p [is less than] .001, one-tailed, effect size r = .77 (positive effect sizes indicate bias). In other chapters not on homosexuality or pedophilia in the 18 textbooks, the mean value of 42 uses was .59 (SD = .73), which also fell significantly below 1, t(41) = -3.64, p [is less than] .001, one-tailed, effect size r = .49. These results confirm that the transgenerational examples were generally not being used appropriately in nonpedophilia chapters.

Examination of other biases. Aside from inappropriate use, other possible biases were inappropriate descriptions of transgenerational examples, as well as inappropriate labeling. For example, Byer and Shainberg (1994) claimed that homosexuality between men was accepted in ancient Greece and Rome: This description was inappropriate, given that these cultures accepted sexual relations between men and boys while scorning the passive adult (Cantarella, 1992; Percy, 1996). Moreover, labeling the relations in Greece and Rome as homosexuality was inappropriate, given that these authors discussed the same relations in their chapter on child sexual abuse as pedophilia. As in the analyses in the previous section, bias was indicated in the current analyses by means that fell significantly below 1 (0 indicated bias; 2 indicated no bias). Further, as in the previous section, the significance level was set at p = .025 for the tests in each pair in this section.

Across the 17 textbooks with a chapter on homosexuality, no bias was shown in how these relations were described, as indicated by a mean rating that was not less than 1 (M = 1.40, SD = .63), t(51) = 4.59, p [is greater than] .99, one-tailed; similarly, across the 18 textbooks in other chapters not concerned with homosexuality or pedophilia, no bias was found (M = 1.43, SD = .74), t(41) = 3.77, p [is greater than] .99, one-tailed. On the other hand, bias was indicated for labeling in the case of the chapters on homosexuality (M = .12, SD = .32), t(51) = -19.77, p [is less than] .001 one-tailed, effect size r = .94, as well as other chapters (M = .36, SD = .53), t(41) = -7.82, p [is less than] .001, one-tailed, effect size r = .77.

An overall index of biased use of transgenerational examples was computed by summing the codings assigned to description, labeling, and appropriateness. This index could range from 0 (biased) to 6 (unbiased). Overall bias was examined by comparing the mean value of this index with 3, the midpoint of the scale; a mean significantly less than 3 indicated bias. For the 17 chapters on homosexuality, the mean index of 1.80 (SD = 1.01) fell well below 3, t(53) = -8.53, p [is less than] .001, one-tailed, effect size r = .77. In the 18 other chapters, the mean index also fell significantly below 3 (M= 2.38, SD = 1.59), t(41) = -2.54, p [is less than] .001, one-tailed, effect size r = .37. These results confirm a strong tendency towards biased use of transgenerational examples.

Biased Use of Scholarly Reviews of Homosexuality

As discussed previously, scholarly reviews of homosexuality are not reviews of man-man sex. They are reviews of both man-boy and man-man sex across culture and time, with the former type weighing heavily. Hence, using Ford and Beach (1951), Carrier (1980), Gregersen (1983), or others to make the point that homosexuality has been accepted and practiced by many societies is equally appropriate for providing perspective on Western man-boy and man-man sex. Therefore, bias is indicated by using these examples for one but not the other. In the 17 chapters on homosexuality, these reviews were used in 10 cases (58.82%), compared to no uses (0%) in chapters dealing with pedophilia, [chi square] (1, N = 17) = 10, p [is less than] .001, with a large effect size, r = .77, indicating a strong bias.

Biased Neglect of Transgenderal Homosexuality

In the chapters on homosexuality, transgenderal rather than transgenerational examples should predominate, because the former type is closer to Western gay sexuality because it often involves sexual relations between two adult males (Greenberg, 1988; Williams, 1992). Bias is therefore indicated by a focus on the latter type of examples. All 17 chapters (100%) contained transgenerational examples, but only 7 of these (41.18%) included transgenderal examples. This difference in proportions was statistically significant, [chi square] N = 17) = 10, p [is less than] .001, with a large effect size, r= .77, indicating a strong bias. This underuse of transgenderal examples relative to transgenerational ones occurred despite the availability of numerous examples of the former (cf. Ford & Beach, 1951; Greenberg, 1988).

DISCUSSION

Sharp moral distinctions emerged between man-man sex and man-boy sex in our society in the 1970s. Prior to this time, both forms were reviled and viewed as pathological and criminal. Following gay liberation and the women's movement, however, man-man sex became more tolerated while the urgency to condemn and suppress man-boy sex increased. Consistent with this new view, Masters et al. (1985), in an early edition of their human sexuality textbook, endorsed man-man sex (i.e., homosexuality) but reproved man-boy sex (i.e., pedophilia). In this sense, their writings simply reflected the liberal academic thinking of the day. What was problematic in their treatment, however, was that they concurrently used historical and cross-cultural examples of predominately man-boy sex to provide perspective on Western man-man sex but not on Western man-boy sex. It was the purpose of the current study to investigate the extent of these errors of commission and omission in more recent human sexuality textbooks. Analyses of 18 human sexuality textbooks revealed pervasive bias. As in the Masters et al. textbook, all textbooks in the current sample approved of man-man sex and reproved man-boy sex. Despite this, they all used cross-cultural and/or historical examples of the latter to provide perspective on the former. Only one of the 18 textbooks used these man-boy examples to provide perspective on Western man-boy sex. These biases obfuscate the nature of male-male sex in an ethnocentric fashion.

Among the 10 most commonly used societies or groups of related societies, nine were of the man-boy type while only one was of the man-man type. Sometimes the use of man-boy sex in other societies was appropriate, as in examining the origins of a homosexual orientation. Most uses, however, were inappropriate, given the moral distinctions that authors drew between man-man sex and man-boy sex in our society. Most uses of man-boy sex in other places and times had as their purpose to show that homosexuality, as opposed to pedophilia, can be and often has been accepted as normal, thus challenging deeply-ingrained Western views of man-man sex. The extent of bias, aside from its pervasiveness, is indicated by the effect sizes in the statistical analyses. Averaging across the effect sizes associated with the biased use of transgenerational examples, the overall index of appropriateness of use, the biased use of scholarly review citations, and the biased use of transgenerational rather than transgenderal examples, the mean effect size based on Fisher z transformations (Rosenthal, 1984) was huge, r = .80.

Illustrations of Biased Use of Perspective

Although the statistical analyses demonstrate pervasive and strong bias, it is instructive to review several specific examples to illustrate the internal inconsistencies to be found in the current sample of textbooks. In their chapter on homosexuality, Hyde and DeLamater (1997) discussed ritualized homosexual behavior in Melanesia. They quoted an anthropologist who observed:

   When a boy is eleven or twelve years old, he is engaged for several months
   in homosexual intercourse with a healthy older man chosen by his father....
   Men point to the rapid growth of adolescent youths, the appearance of peach
   fuzz beards, and so on, as the favorable results of this child-rearing
   practice (p. 397).

They commented that we would surely term this behavior "homosexual," and added that it "is fortunate that anthropologists were able to make their observations over the last several decades to document these interesting and meaningful practices before they disappear" (p. 397). In a later chapter, when discussing such relations in our society, their language switched to terms such as "child sexual abuse," "victims," and "perpetrators." They no longer characterized these practices as interesting and meaningful, but as psychologically damaging in most cases, leading to extreme reactions such as posttraumatic stress disorder. In their chapter on homosexuality, in order to illustrate that "society's response to homosexual behavior and to the rights of homosexuals has had a long history of alterations between tolerance and persecution" (p. 325), Kelley and Byrne (1992) listed two tolerating examples from history, both of which were of the man-boy type: ancient Greece and Western Europe until the 13th century. In their chapter covering pedophilia, these historical examples were not used to provide perspective. Instead, sensationalistic examples of sex murderers were used, such as John Wayne Gacy, prominently shown in a photo wearing a clown's outfit, who murdered 32 young males (mostly older teenagers and young men in their 20s), and a man in China who murdered a 2nd grade boy by piercing his eardrums and veins with needles. In his chapter on homosexuality, Zgourides (1996) discussed pederasty in ancient Greece at length to provide perspective on homosexuality. He later speculated that many instances of homophobia in our society stem from our condemnation of homosexual acts, particularly anal intercourse. To argue that Westerners' disgust at anal intercourse is a cultural trait, he pointed out that "[b]oys and men of the Siwan of Africa practice it, as do the boys and men of the Kiraki [sic] of New Guinea" (p. 131). When discussing pederasty in our society, he did not mention Greek pederasty or cite the practices of the Siwans or Keraki; instead, his discussion was now on homosexual male pedophiles who "victimize" boys usually through anal intercourse, wherein the "harmful effects ... on the victim are many" (p. 346).

Concluding Remarks

An important function of scientific presentations is to advance an objective understanding of phenomena of interest. Using cross-cultural and historical perspectives is recognized in sexology as a valuable means by which to advance such understanding of sexological phenomena (Bullough, 1976; Ford & Beach, 1951; Gebhard, 1985), particularly those of a taboo nature such as homosexuality (Boswell, 1980; Greenberg, 1988). The human sexuality textbooks in the current sample extensively used cross-cultural and historical data to provide perspective on male homosexuality; their use, however, generally failed to advance an objective understanding of this phenomenon. These data were cited selectively and inappropriately. Their use seemed to be guided more by the textbook authors' own moral positions than by a striving for scientific objectivity. The fallout from this bias is that students, who rely on these textbooks as authoritative sources of scientific information, are not being challenged to understand the various forms of male-male sex in a scientifically objective fashion.

(1) According to David Lee (personal communication), who is sponsoring editor for psychology at Houghton Mifflin and handles Allgeier and Allgeier's textbook, the major market leaders among current human sexuality textbooks are, in alphabetical order: Allgeier and Allgeier (1995), Crooks and Baur (1993), Hyde and DeLamater (1997), Masters, Johnson, and Kolodny (1995), Rathus, Nevid, and Fichner-Rathus (1997), and Strong and DeVault (1997). All these textbooks were included in the current sample. At the 1996 and 1997 Eastern Psychological Association conferences, the author asked publishers' representatives what the market leaders are. Responses were in agreement with the above list.

REFERENCES

Adam, B. D. (1985). Age, structure, and sexuality: Reflections on the anthropological evidence on homosexual relations. Journal of Homosexuality, 11, 19-33.

Allgeier, E. R., & Allgeier, A. R. (1995). Sexual interactions (4th ed.). Lexington, MA: D. C. Heath.

Baldauf, I. (1988). Die Knabenliebe in Mittelasien: Bacabozlik [Bacabozlik: Boylove in Central Asia]. Berlin, Germany: Verlag das Arabische Buch.

Bauserman, R. (1990). Objectivity and ideology: Criticism of Theo Sandfort's research on man-boy sexual relations. Journal of Homosexuality, 20, 297-312.

Bauserman, R., & Rind, B. (1997). Psychological correlates of male child and adolescent sexual experiences with adults: A review of the nonclinical literature. Archives of Sexual Behavior, 26, 105-142.

Boswell, J. (1980). Christianity, social tolerance, and homosexuality: Gay people in Western Europe from the beginning of the Christian era to the fourteenth century. Chicago: University of Chicago Press.

Bremmer, J. (1980). An enigmatic Indo-European rite: Paederasty. Arethusa, 13, 279-298.

Bullough, V. L. (1976). Sexual variance in society and history. New York: Wiley.

Byer, C. O., & Shainberg, L. W. (1994). Dimensions of human sexuality (4th ed.). Madison, WI: Brown and Benchmark.

Cantarella, E. (1992). Bisexuality in the ancient world. New Haven, CT: Yale University Press.

Carrier, J. M. (1980). Homosexual behavior in cross-cultural perspective. In J. Marmor (Ed.), Homosexual behavior: A modern reappraisal (pp. 100-122). New York: Basic Books.

Carroll, J. L., & Wolpe, P. R. (1996). Sexuality and gender in society. New York: Harper Collins.

Cohen, J. (1988). Statistical power analyses for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Crooks, R., & Baur, K. (1993). Our sexuality (5th ed.). Redwood City, CA: Benjamin/Cummings.

Davenport, W. H. (1965). Sexual patterns and their regulation in a society of the Southwest Pacific. In F. A. Beach (Ed.), Sex and behavior (pp. 164-207). New York: Wiley.

Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press.

Ford, C., & Beach, F. (1951). Patterns of sexual behavior. New York: Harper & Row.

Francoeur, R. T. (1991). Becoming a sexual person (2nd ed.). New York: Macmillan.

Gebhard, P. H. (1985). Sexuality in cross-cultural perspective. In W. Masters, V. Johnson, R. Kolodny (Eds.), Human sexuality (2nd ed.) (pp. 620-637). Boston: Little, Brown and Company.

Gebhard, P. H., Gagnon, J. H., Pomeroy, W. B., & Christenson, C.V. (1965). Sex offenders: An analysis of types. New York: Harper & Row.

Glass, G. V., & Hopkins, K. D. (1996). Statistical methods in education and psychology. Needham Heights, MA: Allyn & Bacon.

Greenberg, D. (1988). The construction of homosexuality. Chicago: University of Chicago Press.

Greenberg, J. S., Breuss, C. E., & Mullen, K. D. (1993). Sexuality: Insights and issues (3rd ed.). Madison, WI: Brown & Benchmark.

Gregersen, E. (1983). Sexual practices: The story of human sexuality. New York: Franklin Watts.

Haas, K., & Haas, A. (1993). Understanding sexuality (3rd. ed.). St. Louis, MO: Mosby.

Herdt, G. (1987). The Sambia: Ritual and gender in New Guinea. New York: Harcourt Brace Jovanovich.

Herdt, G. (1991a). Representations of homosexuality: An essay on cultural ontology and historical comparison (Part I). Journal of the History of Sexuality, 1, 481-504.

Herdt, G. (1991b). Representations of homosexuality: An essay on cultural ontology and historical comparison (Part II). Journal of the History of Sexuality, 1, 603-632.

Herdt, G. (1995). The development of desires in Sambian (New Guinea) boy inseminating rituals. Paper presented at the Eastern Region Annual Conference of the Society for the Scientific Study of Sexuality (SSSS) in Atlantic City, April 8, 1995.

Hinsch, B. (1990). Passions of the cut sleeve: The male homosexual tradition in China. Berkeley, CA: University of California Press.

Hyde, J. S., & DeLamater, J. (1997). Understanding human sexuality (6th ed.). New York: McGraw-Hill.

Katchadourian, H. A. (1989). Fundamentals of human sexuality (5th ed.). New York: Harcourt Brace Jovanovich.

Keesing, R. M. (1982). Prologue: Toward a multidimensional understanding of male initiation. In G. Herdt (Ed.), Rituals of manhood: Male initiation in Papua New Guinea (pp. 1-39). Berkeley, CA: University of California Press.

Kelley, K., & Byrne, D. (1992). Exploring human sexuality. Englewood Cliffs, NJ: Prentice Hall.

Kelly, G. F. (1994). Sexuality today: The human perspective (4th ed.). Guilford, CT: The Dushkin Publishing Group.

Knutson, D. C. (1979/80). Introduction. Journal of Homosexuality, 5, 5-23.

Luria, Z., Friedman, S., & Rose, M. D. (1987). Human sexuality. New York: Wiley.

Masters, R. E. L. (1962). Forbidden sexual behavior and morality: An objective re-examination of perverse sex practices in different cultures. New York: Julian Press.

Masters, W., Johnson, V., & Kolodny, R. (1985). Human sexuality (2nd ed.). Boston: Little, Brown and Company.

Masters, W., Johnson, V., & Kolodny, R. (1995). Human sexuality (5th ed.). New York: Harper Collins.

Mendel, M. P. (1995). The male survivor. Thousand Oaks, CA: Sage Publications.

Money, J., & Ehrhardt, A. (1972). Man and woman, boy and girl: The differentiation and dimorphism of gender identity from conception to maturity. Baltimore: Johns Hopkins University Press.

Murray, S. O. (1992). Age-stratified homosexuality: Introduction. In S. O. Murray (Ed.), Oceanic homosexualities (pp. 3-23). New York: Garland.

Okami, P. (1990). Sociopolitical biases in the contemporary scientific literature on adult human sexual behavior with children and adolescents. In J. R. Feierman (Ed.), Pedophilia: Biosocial dimensions, pp. 91-121. New York: Springer-Verlag.

Percy, W. A. (1996). Pederasty and pedagogy in archaic Greece. Chicago: University of Illinois Press.

Rathus, S. A., Nevid, J. S., & Fichner-Rathus, L. (1997). Human sexuality in a world of diversity (3rd ed.). Boston: Allyn and Bacon.

Rind, B. (1995). An analysis of human sexuality textbook coverage of the psychological correlates of adult-nonadult sex. The Journal of Sex Research, 32, 219-233.

Rind, B., & Tromovitch, P. (1997). A meta-analytic review of findings from national samples on psychological correlates of child sexual abuse. The Journal of Sex Research, 34, 237-255.

Rind, B., Tromovitch, P., & Bauserman, R. (1998). A meta-analytic examination of assumed properties of child sexual abuse using college samples. Psychological Bulletin, 124, 22-53.

Rosenthal, R. (1984). Meta-analytic procedures for social research. Beverly Hills, CA: Sage Publications.

Saikaku, I. (1990). The great mirror of male love. (P. G. Schalow, Trans.). Stanford, CA: Stanford University Press. (Original work published in 1687).

Sandfort, T. (1983). Pedophile relationships in the Netherlands: Alternative lifestyle for children? Alternative Lifestyles, 5, 164-183.

Sandfort, T. (1984). Sex in pedophiliac relationships: An empirical investigation among a nonrepresentative group of boys. The Journal of Sex Research, 20, 123-142.

Saslow, J. M. (1989). Homosexuality in the renaissance: Behavior, identity, and artistic expression. In M. B. Duberman, M. Vicinus, & G. Chauncey, Jr. (Eds.), Hidden from history (pp. 90-105). New York: New American Press.

Schalow, P. G. (1989). Male love in early modern Japan: A literary depiction of the "youth." In M. B. Duberman, M. Vicinus, & G. Chauncey, Jr. (Eds.), Hidden from history (pp. 118-128). New York: New American Press.

Socarides, C. W. (1970). Homosexuality and medicine. Journal of the American Medical Association, 212, 1199-1202.

Strong, B., & DeVault, C. (1997). Human sexuality (2nd ed.). Mountain View, CA: Mayfield.

Trumbach, R. (1977). London's sodomites: Homosexual behavior and Western culture in the 18th century. Journal of Social History, 11, 1-33.

Trumbach, R. (1989). The birth of the queen: Sodomy and the emergence of gender equality in modern culture, 1660-1750. In M. B. Duberman, M. Vicinus, & G. Chauncey, Jr. (Eds.), Hidden from history (pp. 129-140). New York: New American Press.

Turner, J. S., & Rubinson, L. (1993). Contemporary human sexuality. Englewood Cliffs, NJ: Prentice Hall.

Urquiza, A. J., & Capra, M. (1990). The impact of sexual abuse: Initial and long-term effects. In M. Hunter (Ed.), The sexually abused male: Prevalence, impact, and treatment (Vol. 1) (pp. 105-135). Lexington, MA: Lexington Books.

Wade, C., & Cirese, S. (1991). Human sexuality (2nd ed.). New York: Harcourt Brace Jovanovich.

Watkins, B., & Bentovim, A. (1992). Male children and adolescents as victims: A review of the current knowledge. In G. C. Mezey & M. B. King (Eds.), Male victims of sexual assault (pp. 27-66). New York: Oxford University Press.

Wikan, U. (1982). Behind the veil in Arabia: Women in Oman. Baltimore: Johns Hopkins University Press.

Williams, W. L. (1992). The spirit and the flesh: Sexual diversity in American Indian culture. Boston: Beacon Press.

Williams, W. L. (1996). Two-spirit persons: Gender nonconformity among native American and native Hawaiian youths. In R. C. Savin-Williams & K. M. Cohen (Eds.), The lives of lesbians, gays, and bisexuals (pp. 416-435). New York: Harcourt Brace.

Zgourides, G. (1996). Human sexuality: Contemporary perspectives. New York: Harper Collins.

Manuscript accepted July, 2 1998

Bruce Rind

Temple University

Correspondence should be sent to Bruce Rind, Ph.D., Department of Psychology, Temple University, Philadelphia, PA 19122; e-mail: [e-mail]