Legal, Social, and Biological Definitions of Pedophilia

Archives of sexual behavior

Ames, Ashley M., & Houston David A.
Type of WorkArticle
Publication LanguageEnglish

The paper

  1. briefly reviews some of the historical attitudes toward the sexual use of children,
  2. examines on an atomistic level what is known about those    individuals who act on their arousal to children, and
  3. examines how societal values have influenced the concept of     adult-child sex and how this influence has impacted on the quest for    scientific knowledge.

Behavior that constitutes the sexual use of children in a socio-legal sense is not identical [334] to what constitutes the sexual use of children in a biological sense. Accordingly, “socio-legal pedophilia” has different dimensions than “sociological pedophilia.” The confusion surrounding the differences between these two conceptualizations has hindered research into pedophilia, hampering our efforts to understand this problem. We must distinguish between violations of socio-legal norms and the more biologically dysfunctional problem of true pedophilia (i.e., sexually attracted by biologically prepubescent children).

Historical overview

The historical record reveals no universal consensus on the appropriateness of adult-child sex outside the nuclear family. Nonetheless, while little recorded history exists pertaining to children, the evidence that does exist indicates that the sexual use of children has a long history. The sexual use of pubescent boys was practiced in Ancient Greece where young males, barely past puberty and as yet “unbearded,” were solicited by warriors as protégés and lovers (Vanggaard, 1972). These boys always took the passive role, since men who allowed themselves to be sodomized were held in the lowest contempt (Ungaretti, 1978).

The growth of Christianity led to the condemnation of these practices, though apparently not to protect the young males as much as to prevent onanism (Goodrich, 1976). These condemnations served to protect male children somewhat, but no such protection was afforded female children. Medieval Canon law officially forbade the marriage of girls less than 12 years old, but it was not uncommon to find girls of 10 married to very old men (Nobile, 1976). The streets of 14th century Florence were alive with children of both sexes acting as prostitutes (Goodrich, 1976).

Historical accounts from the 18th century indicate that adult-child sex (particularly same-sex pairings) were an accepted practice in China, Japan, Africa, Turkey, Arabia, Egypt, and the Islamic areas of India (Trumbach,1977). In 19th century London the going price for a virginal, 12-year-old girl, of good background, was reported to be 400 pounds (Bullough, 1964).

Two recent “travel guides for gentlemen” show that the desire for sexual interactions with children is still apparent in the Western world. The Discreet Gentlemen’s Guide to the Pleasures of Europe (1975) reported on where one can find “Lolita-eyed nymphettes who make pocket money with every orifice but the natural one.” Along the same lines, Mankoff’s Lusty Europe (1973) revealed where one can find prostitutes as young as 10 years old. The full scope of such sexual practices is indicated in a report on child pornography given to the Congress of the United States. This report stated that at least 250 publications exist in the U.S. devoted to the graphic depiction of erotica [335] using children from 3 to 5 years of age. The U. S. Humane Society estimated that 1.2 million children annually are involved in illegal commercial enterprises revolving around sexual exploitation. Another 100,000 children are estimated to be sexually abused annually by individuals (Kemp, 1980).

Adult sexual interest in children is not confined to the Western world. Ritualized homosexuality involving young boys is practiced in 10 to 20% of Melanesian societies (Herdt, 1984). Variations of the manner in which sex is allowed are found from one tribe to another with some using fellatio, others anal intercourse, and still others masturbation. In these societies, girls as young as 8 years old are also used for special ceremonies that involve making medicine from seminal fluids gathered from the child after intercourse. Young brides of the Kimam Papuan tribe are “tested” by many men prior to the wedding ceremony to insure their fitness for marriage (Serpenti, 1984).

The historical and cross-cultural record clearly reflects the long-standing existence of the use of children in sexual practices. The accounts given thus far have all centered on extra-familial liaisons. Evidence indicates that intra-familial adult-child sex has an equally pervasive history.       

Historical and cross-cultural perspectives on incest

The factor of relatedness has resulted in a different conceptualization of adult/child sex when the act occurs between a child and a relative. The ubiquity of the incest taboo has been debated by sociologists and anthropologists. Sociologists have generally contended that the prohibition against incest is societally determined and is not universal (Solomon, 1978), while anthropologists have generally contended that the incest taboo is one of the most universal of all rules (Fox, 1972). Whether universal or just extremely widespread, the prevalence of the incest taboo indicates a common cross-cultural process that has led to its adoption in varying degrees in divergent cultures.

In spite of the debate over how and why the incest taboo came about, empirical evidence is convincing that intra-family mating lead to a selective disadvantage (Lindzey, 1967). In Middle Eastern societies, where relatives frequently married, increased rates of genetic disease have been found (Feldman, 1980). Also in Japan, where as many as 5% of all marriages were between first cousins, Schull and Neel (1965) found that inbred offspring were not as healthy as outbred offspring. Evidence from the United States is similar; children of incestuous matings display inordinately high rates of mental retardation and physical anomalies (Adams and Neel, 1967).

Some form of the incest taboo appears to be in place not only cross-culturally but also across species. Among primates, where paternity is often in [336] question, either the males or the females of the species leave the home troop at sexual maturity, thus rendering incestuous matings unlikely. Female chimpanzees in estrus have been observed to display violent behavior to ward off the sexual advances of a brother (Goodall, 1971).

Due to the widespread cross-species and cross-cultural nature of the incest taboo, socio-biologists contend that avoidance of incest has evolved as an adaptive pattern. Bischof (1975) contends that natural selection seems to favor behavior patterns that render incest very improbable. Breuer (1984) stated that “in the light of today’s ethological knowledge it must be assumed that the notion of incest was not ‘invented’ at all, but rather that men inherited from their ancestors unconscious behavior patterns of inborn incest avoidance that later were woven into their culture.”

Whether the avoidance of incest is a biological or a sociological imperative, intra-familial adult/child sex involves a second dimension of deviance over and above the societal prohibitions against the sexual use of children. If we assume that the practice of incest involves the breaking of two norms, we can expect to find lower rates of incest than rates of extra-familial adult-child sex within the population. Evidence on the prevalence of sexual abuse reviewed below generally supports this statement.

This hypothesis of double deviance serves to confound research into the etiology of incest. Natural questions that have arisen but are as yet unanswered concern the specificity of the incestuous father’s arousal. Certainly, some pedophiles marry and have children (Abel et al., 1987). Instances of inappropriate sexual age preferences are found less frequently among incestuous fathers than among non-incestuous child molesters (Quinsey et al., 1979). If this finding remains constant under replication, somewhat different processes are involved in the etiology of incest and in the etiology of pedophilia.    

Prevalence of adult/child sex

Ascertaining a definitive estimate of the prevalence of adult/child sex is understandably difficult. Estimates derived from crime statistics involve only reported cases and are likely to be an underrepresentation. Retrospective reports carry their own inherent problems but are perhaps our best source of knowledge. Finkelhor (1979) questioned 530 female and 266 male college students about their sexual experiences in childhood. Results revealed that  

  • 11% of the females and  
  • 4% of the males reported having had an unwanted sexual experience with an adult before the age of 12. 

Females reported that 

  • 46% of such encounters involved a relative, 
  • 33% an acquaintance, and 
  • 24% a stranger.

Males were  less often abused by a relative (17%) and  * more often by acquaintances (53%) and  * strangers (30%). [337]

Apparently, a large percentage of the population has experienced some form of unwanted sexual abuse. Data on the differential incidence of offending across perpetrators indicates that a relatively small number of men are responsible for a large number of offenses. Self-report data gathered from non-incarcerated pedophiles indicate that some pedophiles are responsible for a very large number of victims. Reports obtained from 232 child molesters,who were guaranteed confidentiality, revealed that on average they had victimized 76 children (Abel et al., 1987). Incarcerated pedophiles also report a high incidence of encounters and, on average, report having 11 more victims than those for which they were prosecuted (Groth et al., 1982). An accurate estimate of the number of pedophiles is probably impossible to obtain with the current data. What can be said is that pedophiles commit many offenses across a lifetime of offending and are never prosecuted for many of the offenses that they commit.        

Nature of the offender

A great deal of the existing literature on pedophiles is of an a-theoretical nature. Many of the studies that have attempted to find causal variables in pedophiles have either not used control groups or have attempted such controls based solely on the type of offense (e.g., Kopp, 1962). Furthermore, the existing data are almost exclusively obtained from incarcerated pedophiles, thus limiting the generalizability of findings. Nonetheless, some valuable information has been found and should help to illuminate the nature of the men who seek out children as sexual partners. In a sample of  34 child molesters incarcerated in Canada, only 26% had committed their crimes under the influence of drugs or alcohol. Consistent with current beliefs about the intergenerational effect of abuse, 53% of this sample claimed to have been victims of sexual abuse themselves (Earls et al.,1984). These findings have been replicated in a U.S. sample of 68 pedophiles, incarcerated in Massachusetts for sexual abuse of children. Of these men, 57% reported having been victims of sexual abuse in childhood. This finding compared to 23% of 107 rapists in the same institution. Significantly more child molesters than rapists had siblings with psychiatric histories. Pedophiles also had more chronic medical problems and lower IQs than rapists (Bardet al., 1987).

Attempts to make comparisons of the psychological functioning of sexual offenders based on the age of the victim have revealed some interesting findings, although the utility of using control groups based upon such comparisons has been questioned (Quinsey, 1986). Men whose offenses were committed against children have higher frequencies of avoidant and dependent [338] personalities and lower frequencies of antisocial and borderline personalities than men who agress against adult women (Ames et al., 1987).This finding supports the stereotype of the weak, passive, socially isolated, and inept man who turns to children for sexual fulfillment. This stereotype is further supported by the findings of  Wilson and Cox (1983) < > who interviewed 77 members of a self-help club for pedophiles in London. Using the Eysenck Personality Questionnaire these men presented themselves as introverted, shy, sensitive, lonely, depressed, and humorless.

Men who are involved in incestuous relationships do not present the same picture of pathology as do men who are involved with non-related children. In contrast to the findings cited above, incest very frequently involves alcohol abuse (Maisch, 1972; Virkkunen, 1974). Gebhard et al. (1965) reported that incestuous fathers seldom display diagnosable mental problems but usually fall at the ends of the spectrum of dominance; some incestuous fathers are very dominant, others, very passive, with few at intermediate levels. Even though little agreement exists in the literature about the nature of the men who practice incest, agreement is found about the nature of the family where incest is found. De Francis (1969) reported that the greater the family disorganization, the higher the proportion of relatedness in cases of sexual abuse. Along these same lines, Avery-Clark et al. (1981) and Patterson (1982) have reported that extreme chaos and frequent crisis situations are often found in incestuous families. They also reported that incestuous families frequently have high rates of general disorganization and disturbed marital relationships.

What then can be said about the nature of pedophiles? Given the existing research, about half of incarcerated pedophiles have themselves experienced sexual abuse as children. As a group, they are less prone than other offenders to abuse drugs or alcohol. They often report being sickly as children and having chronic illnesses. They describe themselves in terms of being shy and introverted, and they are often given a DSM-III (American Psychiatric Association, 1980) diagnosis that supports this view they hold of themselves.

The nature of the society

Societies have dealt with adult-child sex with a great deal of variance. Even the presumed universal taboo of father-daughter incest has been inconsistently enforced throughout European and American history (Maisch, 1972). This legal ambivalence is perhaps a reflection of a social ambivalence about the appropriateness of adult/child sex.

Societies that practice a form of ritualized homosexuality between men and boys are usually societies in which women have very low status (Allen, 1984). In addition to holding low status, women in these societies are often looked upon as a toxic force to be avoided except for procreative purposes [339] (Herdt, 1984). These beliefs leave men little choice but to turn to younger males for recreational sexual liaisons. These societies also tend to be warrior societies that highly value virility. Age-appropriate homosexual acts would not be tolerated since one of the partners (whoever was the recipient of the other's penis) would be viewed as submissive and thus unworthy of his place with other men. After observing several of these cultures as they exist today, Herdt (1984) has said that the belief that the semen given to the child during these practices will facilitate his growth is merely a rationalization for homoerotic play.

Children of high social status frequently escaped being subjected to these practices. This observation is evidenced by the practice in Ancient Rome of having young boys of high status wear a gold ball around their necks so that any potential abusers would know that they were not to be used (De Mause, 1974, p. 44). The use of female children also seems to have been primarily restricted to girls of low social status. When outrage was expressed about an adult-child liaison, the liaison frequently involved a man of low status and a child of high status. When cases involved parties from the same social status, concern was more for the father’s loss of a commodity (his daughter's virginity) than for any loss incurred by the child (Rush, 1980).

During the 17th, 18th, and 19th centuries, childhood was much briefer, and children worked in factories or were apprenticed at very young ages. Children no less than women were the property of the father (Radbill, 1980). The line between childhood and adulthood was not clear, and biological rather than chronological dictates were used to determine the end of childhood. This situation was true not only in our own pre-industrial society but in all primitive cultures yet studied. The end of childhood for the female was marked by the commencement of menstruation and for the male by his ability to physically assume a man’s role (De Mause, 1974).

The trend in Western society has been to arbitrarily set dates as milestones along the way to adulthood, instead of observing the more natural biological markers. This prolongation of childhood has led to a limbo period wherein young people are physiologically, but not sociologically, prepared for reproduction. At this point in our history, a very real conundrum exists for the researchers of adult-child sex. This problem is reflected in the question of what truly marks the point beyond which sexual interaction with a child is pathological and not just criminal. The remainder of this paper is devoted to exploring that question.

Where does pedophilia begin?

We turn to the discipline that rests at the intersection of sociology and biology, socio-biology. A basic tenet of sociobiology is that traits producing [340] behaviors resulting in a higher rate of reproductive success become more highly represented in the population over time. The selection of fecund sexual partners can be viewed as such a behavior.

Men who consistently prefer prepubescent children as sexual partners are practicing a very unsound reproductive strategy. In contrast, men who prefer post-pubescent girls are practicing a sound reproductive strategy. Socio-biological research has consistently found a distinct preference among males of all primates for young females (Mackey, 1980; Paterson and Pettijohn, 1982). In an interesting study with males between age 7 and adulthood, Cross and Cross (1971) found that all males, regardless of age, rated the faces of 17-year-olds as more appealing than the faces of any other age group.These findings are suggestive of a pervasive male preference for youthful females. From the standpoint of sound reproductive strategy and increased reproductive success, this preference makes intuitive sense.

The biological line of demarcation between childhood and adulthood seems to be a more natural discriminator in efforts to classify men as pedophiles or non-pedophiles. As in all classification systems, those individuals who fall at or near the intersection are hardest to classify. If, however, one is using the secondary sex characteristics of the victim as the crucial discriminator, whether the man is aroused by children or by post-pubescent girls should be fairly apparent. If the offender’s victims are consistently without secondary sexual characteristics, the man is a pedophile.

The importance of body-shape in the construct of pedophilia is demonstrated in research by Freund (1965). Using a group of homosexually oriented and heterosexually oriented pedophiles, Freund found that erectile responses were more dependent on their explicit preference for an immature or mature body-type than upon their preference as to gender of sexual partner. In a series of similar experiments involving the elicitation of erectile responses via visual stimuli, Quinsey et al. (1979) found that incest offenders were not as aroused by models of an inappropriate age as were pedophiles. This finding supports the hypothesis that incest follows a different etiological course than does pedophilia. In a third set of experiments, Howell (1979) found that pedophiles demonstrated a marked preference for slimness and a small body build when compared to other offenders. All these findings suggest that for the pedophile the lack of secondary sex characteristics is the arousing criterion, whereas for non-pedophiles, the presence of secondary sex characteristics is the arousing criterion. If this line of thinking proves to be correct, age is an irrelevant factor in the diagnosis of pedophilia. Instead of age, body type should become the crucial discriminant.

The desire among men to have sexual interactions with children is not uncommon. We suggest that a distinction should be made between biological children and socio-legal children. The end of biological childhood is clearly marked in females by the onset of menarche. While the age of the onset of [341] menarche can vary from one culture to another, it is marked by discernible signs. The redefinition of body-type from flat and slim to mature is a universal determinant of the end of biological childhood. The end of socio-legal childhood is not so apparent and is extremely variable across cultures and times. Laws governing child molestation reflect this socio-legal childhood, regardless of its discrepancy with biological childhood. This discrepancy has served to cloud what should be a natural distinction between offender types, between child molestation and rape. By making this distinction, important differences between these populations will perhaps be found and the etiology of pedophilia will become more apparent.


Abel, G., Mittelman, M., and Becker, J. (1987). Sexual offenders: Results of assessment and recommendations for treatment. In Ben-Aron, M. H., Hucker, S. J., and Webster, C. D.(eds.), Clinical Criminology: The Assessment and Treatment of Criminal Behavior, Butterworth,Toronto, Canada.

Adams, M., and Neel, J. (1967). Children of incest. Pediatrics 40: 55.

Allen, M. (1984). Homosexuality, male power, and political organization in North Vanuatu: A comparative analysis. In Herdt, G. H. (ed.), Ritualized Homosexuality in Melanesia, University of California Press, Berkeley, pp. 83-126.

American Psychiatric Association. (1980). Diagnostic and Statistical Manual of Mental Disorders,3rd ed., APA, Washington, DC.

Ames, A., Fleming, D., Knight, R., and Straus, H. (1987). The incidence of antisocial personality disorder among rapists and child molesters. Unpublished manuscript.

Avery-Clark, C., O'Neil, J., and Laws, D. (1981). A comparison of intra-familial sexual and physical child abuse. In Cook, M., and Howells, K. (eds.), Adult Sexual lnterest in Children,Academic Press, Toronto, Canada, pp. 3-39.

Bard, L., Caster, D., Cerce, D., Knight, R., Rosenberg, R., and Schneider, B. (1987). A descriptive study of rapist and child molesters: Developmental, clinical and criminal characteristics. Behav. Sci. Law 5: 203-220.

Bischof, N. (1975). Comparative ethology in incest avoidance. In Fox, R. (ed.), Biosocial Anthropology, Malaby, London, England.

Bruer, G. (1984). Sociobiology and the Human Dimension, Cambridge University Press, Cambridge,England.

Bullough, V. (1964). The History of Prostitution, University Books, New Hyde Park, NY.

Cross, J., and Cross, J. (1971). Age, sex, race, and the perception of facial beauty. Dev. Psychol. L5: 433-439.

De Francis, V. (1969). Protecting the Child Victim of Sex Crimes Committed by Adults, American Humane Society, Children's Divisions, Denver, CO.

De Mause, L. (1974). The History of Childhood, Psycho-history Press, New York.

The Discreet Gentlemen's Guide to the Pleasures of Europe. (1975). Bantam Books, New York.

Earls, C., Bouchard, L., and Laberge, J. (1984). Étude descriptive des délinquents sexuels incarcérés dans les pénitenciers Quebéquois, (Cahier de Recherce No. 7, Institute Philippe Pinel de Montreal), Montreal, Canada.

Feldman, M. (1980). Genetics and social behavior. In Markl, H. (ed.), Evolution of Social Behavior: Hypothesis and Empirical Tests, (Dahlem Konferenzen), Berlin, Germany.

Finkelhor, D. (1979). Psychological, cultural and family factors in incest and family sexual abuse. J. Marr. Fam. Counsel. 4: 41-49.

Fox, R. (1972). Alliance and constraint: Sexual selection and the evolution of human kinship systems. In Campbell, B. (ed.), Sexual Selection and the Descent of Man 1871-1971, Heineman, London.

Freund, K. (1965). Erotic preference in pedophilia. Behav. Res. Ther. 3: 229-234.

Gebhard, P., Gagnan, J., and Pomeroy, W. (1965). Sex Offenders: An Analysis of Type, Harper & Row, New York.

Goodall, J. (1971). The behavior of free-living chimpanzees in the Gombe stream area. Anita. Behav. Monogr. 1: 161-311.

Goodrich, M. (1976). Sodomy in medieval secular law. J. Homosex. 1: 295-302.

Groth, A., Longo, R., and McFadin, J. (1982). Undetected recidivism among rapists and child molesters. Crime Delinquency 28: 450-458.

Herdt, G. (1984). Ritualized Homosexuality in Melanesia, University of California Press, Berkeley,pp. 1-81.

Howell, K. (1979). Some meanings of children for pedophiles. In Cook, M., and Wilson, G.(eds.), Love and Attraction, Pergamon, New York, pp. 519-526.

Kemp, C. (1980). The Battered Child, University of Chicago Press, Chicago.

Kopp, S. (1962). The character structure of sex offenders. Am. J. Psychother. 16: 64-70.

Lindzey, G. (1967). Some remarks concerning incest, the ancient taboo and psychoanalytic theory. Am. Psychologist 22: 1051.

Mackey, W. (1980). A socio-biological perspective on divorce patterns of men in the United States. J. Anthropol. Res. 36: 419-430.

Maisch, H. (1972). Incest, Stein & Day, New York.

Mankoff, A. (1973). Mankoff’s Lusty Europe, Viking Press, New York.

Nobile, P. (1976). Introduction. In Kramer, W. (ed.), The Normal and Abnormal Love of Children, Sheed, Andrews & McMeel, Kansas City, MO.

Paterson, C., and Pettijohn, T. (1982). Age and human mate selection. Psychol. Rep. 51: 70.

Patterson, G. (1982). Coercive Family Process, Castalia, Eugene, OR.

Quinsey, V. (1986). Men who have sex with children. In Weisstub, O. N. (ed.), Law and Mental Health: International Perspectives, Vol. 2, Pergamon, New York.

Quinsey, V., Chaplin, T., and Carrigan, W. (1979). Sexual preferences among incestuous and non-incestuous child molesters. Behav. Ther. 10: 562-565.

Radbill, S. (1980). Children in a world of violence: A history of child abuse. In Kempe, C. H., and Heifer, R. E. (eds.), The Battered Child, University of Chicago Press, Chicago.

Rush, F. (1980). The Best Kept Secret, Prentice-Hall, Englewood Cliffs.

Schull, W., and Neel, J. (1965). The Effects of Inbreeding in Japanese Children, Harper & Row,New York.

Serpenti, L. (1984). The ritual meaning of homosexuality and pedophilia among the Kimam-Papuans of South Irian Jaya. In Herdt, G. H. (ed.), Ritualized Homosexuality in Melanesia,University of California Press, Berkeley, pp. 318-335.

Solomon, R. (1978). Socio-biology, morality, and culture. In Stem, G. S. (ed.), Morality as a Biological Phenomenon, (Dahlem Konferenzen) Berlin, Germany, pp. 283-387.

Trumbach, R. (1977). London's sodomites: Homosexual behavior and Western culture in the eighteenth century. J. Soc. Hist. 11: 1-33.

Ungaretti, J. (1978). Pederasty, heroism and the family in classical Greece. J. Homosex. 3:291-300.

Vanggaard, T. (1972). Phallos: A Symbol and Its History in the Male World, Cape, London.

Virkkunen, M. (1974). Incest offenses and alcoholism. Med. Sci. Law 124-128.

Wilson, G., and Cox, D. (1983). Personality of pedophile club members. Pers. Indiv. Diff.4: 323-329.